“Peter Sturmey and Mary McMurran have edited an excellent book on forensic case formulation for this Wiley-Blackwell series. The authors of the chapters are well-respected experts in the forensic field. Readers who work in forensic services will find the chapters on case formulation both informative theoretically and helpful clinically. Work with offenders will be enhanced by assimilating the ideas and suggestions in this book.”
Effective assessment and treatment of offenders is important for reducing the likelihood of further offending. Understanding the processes which initiate and maintain offending behavior is integral to the design and evaluation of appropriate, individually tailored interventions. Forensic Case Formulation describes the principles and application of case formulation specifically for forensic clinical practice. In this edited volume, contributors review the fundamental aspects of case formulation, including definitions, reliability, and validity. The practical applications of case formulation applied to violence, sexual offenses, personality disorder, and substance abuse are described, and illustrated with case studies. Also covered is the interrelationship between case formulation and risk assessment. Forensic Case Formulation offers rich insights into the use of case formulation with forensic clients.
Peter Sturmey is Professor of Psychology at Queens College and The Graduate Center, City University of New York. He has published extensively on developmental disabilities, applied behavior analysis, and on issues relating to staff and parent training. Mary McMurran is Professor of Personality Disorder Research in the University of Nottingham’s Institute of Mental Health. She has written extensively on personality disorders, alcohol-related aggression, and readiness to engage in therapy.
Cover image: Image © linearcurves/iStockphoto Cover design by Nicki Averill Design
Forensic Case Formulation
Kate Davidson, Professor of Clinical Psychology, University of Glasgow, UK
Edited by Sturmey and McMurran
Forensic Case Formulation
Edited by
Peter Sturmey and Mary McMurran
hb_9780470683958.indd 1
22/6/11 07:41:17
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
WILEY SERIES IN
FORENSIC CLINICAL PSYCHOLOGY Edited by
Clive R. Hollin School of Psychology, University of Leicester, UK and
Mary McMurran Institute of Mental Health, University of Nottingham, UK For other titles in this series please visit www.wiley.com/go/fcp
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION Edited by
Peter Sturmey Queens College and City University, New York, USA and
Mary McMurran Institute of Mental Health, University of Nottingham, UK
A John Wiley & Sons, Ltd., Publication
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
This edition first published 2011 © 2011 John Wiley & Sons, Ltd. Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley’s global Scientific, Technical and Medical business with Blackwell Publishing. Registered Office John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK Editorial Offices The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK 350 Main Street, Malden, MA 02148-5020, USA 9600 Garsington Road, Oxford, OX4 2DQ, UK For details of our global editorial offices, for customer services, and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell. The right of Mary McMurran and Peter Sturmey to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. Library of Congress Cataloging-in-Publication Data Forensic case formulation / edited by Peter Sturmey and Mary McMurran. p. cm. Includes index. ISBN 978-0-470-68395-8 (cloth) – ISBN 978-0-470-68394-1 (pbk.) 1. Correctional psychology. 2. Criminal psychology. 3. Criminals–Rehabilitation. I. Sturmey, Peter. II. McMurran, Mary. HV9276.F67 2011 614 .15–dc22 2011009359 A catalogue record for this book is available from the British Library. This book is published in the following electronic formats: ePDFs 9781119977025; Wiley Online Library 9781119977018; ePub 9781119976899 Set in 10/12pt Palatino by Aptara Inc., New Delhi, India. 1
2011
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
June 9, 2011
11:27
Printer Name: Yet to Come
CONTENTS About the Editors About the Contributors Series Editors’ Preface
vii ix xiii
Preface
xvii
PART I GENERAL ISSUES 1 Theoretical and Evidence-Based Approaches to Case Formulation Tracey D. Eells and Kenneth G. Lombart
1 3
2 Current Issues in Case Formulation Gregory H. Mumma
33
3 Does Case Formulation Make a Difference to Treatment Outcome? Ata Ghaderi
61
PART II VIOLENCE
81
4 Formulation of Violence Risk Using Evidence-Based Assessments: The Structured Professional Judgment Approach Stephen D. Hart and Caroline Logan
83
5 Cognitive Behavioral Approaches to Formulating Aggression and Violence Kevin Howells
107
6 Formulation of Serious Violent Offending Using Multiple Sequential Functional Analysis Aidan J.P. Hart, David M. Gresswell and Louise G. Braham
129
7 Forensic Case Formulation, Substance Abuse Disorders, and Anger Ellen Vedel and Paul M.G. Emmelkamp
153
P1: OTA/XYZ JWST073-FM
P2: ABC JWST073-McMurran
vi
June 9, 2011
11:27
Printer Name: Yet to Come
CONTENTS
PART III SEXUAL OFFENDING
173
8 Sexual Offenses Against Children James Vess and Tony Ward
175
9 Sexual Offenses Against Adults Stacey L. Shipley and Bruce A. Arrigo
195
PART IV SPECIFIC POPULATIONS
215
10 Forensic Case Formulation with Children and Adolescents Phil Rich
217
11 Formulating Offending Behavior with People with Mild Learning Disabilities William R. Lindsay
237
12 Case Formulation for Individuals with Personality Disorder Lawrence Jones
257
PART V CONCLUSION
281
13 Forensic Case Formulation: Emerging Issues Peter Sturmey and Mary McMurran
283
Index
305
P1: OTA/XYZ JWST073-ata
P2: ABC JWST073-McMurran
May 26, 2011
7:19
Printer Name: Yet to Come
ABOUT THE EDITORS Peter Sturmey is Professor of Psychology at Queens College and The Graduate Center, City University New York and a member of the Learning Processes and Behavior Analysis and Neuropsychology doctoral programs at City University of New York. He has published over 150 articles and 15 books on developmental disabilities. His current research focuses on applied behavior analysis and staff and parents training. Mary McMurran is Professor of Personality Disorder Research at the University of Nottingham’s Institute of Mental Health, United Kingdom. Her research interests include: social problem solving as a model of understanding and treating people with personality disorders; the assessment and treatment of alcohol-related aggression and violence; and understanding and enhancing offenders’ motivation to engage in therapy. She is a Fellow of the British Psychological Society, and recipient of the Division of Forensic Psychology’s Lifetime Achievement Award in 2005.
P1: OTA/XYZ JWST073-ata
P2: ABC JWST073-McMurran
May 26, 2011
7:19
Printer Name: Yet to Come
P1: OTA/XYZ JWST073-loc
P2: ABC JWST073-McMurran
June 9, 2011
11:24
Printer Name: Yet to Come
ABOUT THE CONTRIBUTORS Bruce A. Arrigo Professor of Criminology, Law, and Society in the Department of Criminal Justice and Criminology at the University of North Carolina – Charlotte. He holds affiliated appointments in the Departments of Psychology and Public Health Sciences, and in the Public Policy Program and the Center for Professional and Applied Ethics. He is a Fellow of the American Psychological Association and the Academy of Criminal Justice Sciences. Dr Arrigo’s numerous recognitions include the American Society of Criminology’s Critical Criminologist of the Year Award (2000), the Society for the Study of Social Problem’s Book of the Year Award in Crime and Delinquency (2005), the Academy of Criminal Justice Sciences’ Bruce Smith Sr. Distinguished Research Award (2007), and the First Citizens Bank Scholars Medal (2008). Dr Arrigo’s latest book is The Ethics of Total Confinement: A Critique of Madness, Citizenship and Social Justice (2011) Oxford University Press. Louise G. Braham Acting Lead Psychologist and Consultant Clinical and Forensic Psychologist, Mental Health Service, Rampton Hospital, Nottinghamshire Healthcare NHS Trust and Senior Clinical Tutor for the Trent Clinical Psychology Training Programme, University of Nottingham, UK. Louise has worked in Forensic Mental Health Services for over 15 years. She is a practising clinician, has published and is interested in violence, forensic case formulation, neuropsychological functioning and the experience and symptoms of psychosis related to risk. Tracy D. Eells Clinical Psychologist and Professor in the Department of Psychiatry and Behavioral Sciences at the University of Louisville in Louisville, Kentucky, USA. He earned a Ph.D. from the University of North Carolina, Chapel Hill. His primary area of scholarship is psychotherapy research, particularly expertise in case formulation. He is editor of the Handbook of Psychotherapy Case Formulation and is advisory editor for the journals Psychotherapy Research and Pragmatic Case Studies in Psychotherapy. Paul M.G. Emmelkamp Full professor of clinical psychology at the University of Amsterdam and senior consultant at the forensic psychiatry centre, De Waag. Over the years he has published widely on the etiology and treatment of anxiety disorders. He is involved in therapy-outcome studies on adults with work-related distress, substance abuse disorders, personality disorders, domestic violence, depression, and anxiety disorders, and on youth with ADHD, conduct disorder
P1: OTA/XYZ JWST073-loc
P2: ABC JWST073-McMurran
x
June 9, 2011
11:24
Printer Name: Yet to Come
ABOUT THE CONTRIBUTORS
and anxiety disorders. He has written and co-edited many books, and over 450 publications in peer reviewed journals or books. He is Editor of Clinical Psychology & Psychotherapy and of BMC-Psychiatry: Forensic Psychiatry. He has received a number of honors and awards, including a distinguished professorship (“Academy Professor”) by the Royal Academy of Arts and Sciences. Ata Ghaderi Clinical Psychologist, Licensed Psychotherapist, and Professor of Psychology at the Department of Psychology, Uppsala University, Sweden. His main field of research is prevention and treatment of eating disorders. His interests include philosophy of science in psychology in relation to assessment and diagnosis, psychotherapy research focusing on treatment matching and individualization, dissemination of evidence-based treatment, and basic research on perception and cognition. He is currently running several projects on the intensive treatment of obsessive-compulsive disorder, preventive parent training, treatment of chronic anorexia nervosa, obesity and body dismorphic disorder. David M. Gresswell Co-Director of the Trent Clinical Psychology Training Programme and works part-time as a Consultant Clinical Psychologist in NHS forensic services in Lincolnshire. Mark has worked in forensic mental health services for over 25 years in high and low security settings and in the community. He has a special interest in behavioral approaches and functional analysis. Aidan J.P. Hart Clinical Forensic Psychologist and Senior Lecturer in Clinical Psychology, University of Lincoln, UK. Aidan has worked in hospital and community settings with physically and sexually violent offenders. He has also worked in both inpatient and outpatient settings in the assessment and treatment of both early episode and severe and enduring mental health problems. He has an interest in behavioral analysis and in particular third wave behavioral approaches, such as Relation Frame Theory, and their application to understanding and furthering assessment and treatment processes. Stephen D. Hart Professor in the Department of Psychology at Simon Fraser University and Visiting Professor in the Faculty of Psychology at the University of Bergen. His work focuses on the assessment of violence risk and psychopathic personality disorder. He has co-authored more than 160 articles, chapters, and books, including several violence risk assessment guides. He is a recipient of the Career Achievement Award from the Society of Clinical Psychology (APA Division 12), the Saleem Shah Award for Early Career Research Excellence in Psychology and Law from the American Psychology-Law Society (APA Division 41) and the American Academy of Forensic Psychology, and the Distinguished Achievement Award from the Association of Threat Assessment Professionals. Kevin Howells Clinical and forensic psychologist. He was until recently Professor of Forensic and Clinical Psychology in the Institute of Mental Health and Division of Psychiatry at Nottingham University. He has worked as a clinician in the United Kingdom, Australia and the USA. He has published widely in the field of forensic clinical psychology and headed a research unit at Rampton Hospital investigating interventions for people with personality disorder and high risk. He has
P1: OTA/XYZ JWST073-loc
P2: ABC JWST073-McMurran
June 9, 2011
11:24
Printer Name: Yet to Come
ABOUT THE CONTRIBUTORS
xi
particular interests in cognitive behavioral treatments for offenders, anger and violence, therapeutic climates and readiness for treatment. Lawrence Jones Lead Psychologist and Consultant Clinical and Forensic Psychologist at the Peaks Unit, Rampton Hospital, Nottinghamshire Healthcare NHS Trust. He is a former chair of the division of forensic psychology in the British Psychological Society. He has worked in community, prison and secure hospital settings with offenders. Has published and is interested in planned environment based interventions, personality disordered offenders, sex offenders, forensic case formulation, motivational interventions and using offence paralleling behavior to inform intervention and risk management. William R. Lindsay Consultant Psychologist and Clinical Lead (Scotland) in Castlebeck, Darlington and Professor of Learning Disabilities and Forensic Psychology at the University of Abertay, Dundee. He is also Honorary Professor at Bangor University, Wales. He is a practicing clinician who has over 200 academic publications. Current clinical and research interests include sex offenders, pathways through learning disability forensic services, programmes for criminal behaviour, the assessment of personality and personality disorder, risk assessment and cognitive behaviour therapy. Caroline Logan Consultant Forensic Clinical Psychologist in Greater Manchester West Mental Health NHS Foundation Trust and an Honorary Research Fellow in the Department of Community Based Medicine at the University of Manchester. She has worked in forensic settings for many years, working directly with clients who are at risk to themselves and others and, in a consultancy role, with the multidisciplinary teams and local and national organizations that look after and manage them. Dr Logan has research interests in the areas of personality disorder, psychopathy, and risk, and a special interest in gender issues in offending. Kenneth G. Lombart Clinical Psychologist and Visiting Assistant Professor in the Department of Psychology at the University of Massachusetts Lowell. His primary research interest is in expertise in psychotherapy case formulation. Gregory H. Mumma Associate Professor of Psychology in the Clinical Ph.D. Program at Texas Tech University. His interests include behavioral and idiographic assessment, intra-individual construct validation, cognitive-behavioral case formulation, and clinical decision making for formulation-based, tailored treatment of complex and comorbid cases. His research focuses on person-specific quantitative methods to validate and test cognitive-behavioral case formulations using daily ratings and ecological momentary assessment. His research, as well as his part-time clinical practice, focuses on adults with comorbid mood and anxiety disorders. Phil Rich Former Clinical Director and the current Director of Clinical Program Development of the Stetson School, a 111-bed long-term residential treatment program for sexually reactive children and juvenile sexual offenders in Massachusetts. Phil holds a doctorate in applied behavioral and organizational studies and a master’s degree in social work, and is a licensed independent clinical social worker. He
P1: OTA/XYZ JWST073-loc
P2: ABC JWST073-McMurran
xii
June 9, 2011
11:24
Printer Name: Yet to Come
ABOUT THE CONTRIBUTORS
presents and trains nationally and internationally, and is the author of Understanding Juvenile Sexual Offenders: Assessment, Treatment, and Rehabilitation; Attachment and Sexual Offending: Understanding and Applying Attachment Theory to the Treatment of Juvenile Sexual Offenders; Juvenile Sexual Offenders: A Comprehensive Guide to Risk Evaluation; and the Stages of Accomplishment Workbooks for Sexually Abusive Youth. Stacey L. Shipley Dr Shipley has specialized in psychological services at the crossroads of psychology and the law. Her training has focused on both adults and adolescents. She is a licensed psychologist in three states and is the Director of Psychology at North Texas State Hospital, the only maximum-security forensic hospital in Texas. She specializes in forensic evaluation, treating maternal filicide offenders, and evaluation or treatment for individuals adjudicated by the courts as Not Guilty by Reason of Insanity or Incompetent to Stand Trial for violent offenses. She has published articles on psychopathy, wrote chapters on maternal filicide and serial rape and murder typologies, and co-authored Introduction to Forensic Psychology: Issues and Controversies in Crime and Justice (2nd edition) and the upcoming 3rd edition published by Academic Press. She was first author of The Female Homicide Offender: Serial Murder and the Case of Aileen Wuornos (2004) published by Prentice Hall. Ellen Vedel Cognitive behavior therapist and treatment manager at the Jellinek Addiction Treatment Centre in Amsterdam. As a senior researcher, she is currently involved in clinical trials testing integrated treatment protocols for substance abuse and Post traumatic Stress Disorder and for substance abuse and intimate partner violence. Together with Paul Emmelkamp she is the co-author of Evidence-based Treatments for Alcohol and Drug Abuse: A Practitioner’s Guide to Theory, Methods and Practice. Dr Jim Vess Senior Lecturer at Deakin University in Victoria, Australia and a member of the Clinical Forensic Research group of the Deakin Forensic Psychology Centre. He has over 25 years of clinical and research experience with forensic populations. After receiving his PhD in clinical psychology from Ohio State University in the United States, he then served in a variety of treatment, assessment, and supervisory roles at Atascadero State Hospital, the maximum security forensic psychiatric facility in California. He has been a Senior Lecturer at Victoria University of Wellington, New Zealand, where his research focus was primarily on risk assessment with violent and sexual offenders, as well as public policy dealing with high risk offenders. He has remained active in forensic practice, and has provided expert witness evidence in a variety of High Court and Court of Appeals cases involving high risk sexual offenders. Tony Ward Head of School and Professor of Clinical Psychology at Victoria University of Wellington, New Zealand. His research interests include cognition in offenders, rehabilitation and reintegration processes, and ethical issues in forensic psychology. He has over 285 academic publications and his most recent book is Desistance from Sex Offending: Alternatives to Throwing away the Keys (with Richard Laws; Guilford, 2011). He will be taking up a research chair in Clinical Forensic Mental health at Deakin University, Melbourne, Australia in June 2011.
P1: OTA/XYZ P2: ABC JWST073-seditor JWST073-McMurran
May 26, 2011
7:22
Printer Name: Yet to Come
SERIES EDITORS’ PREFACE ABOUT THE SERIES At the time of writing it is clear that we live in a time, certainly in the UK and other parts of Europe, if perhaps less so in areas of the world, when there is renewed enthusiasm for constructive approaches to working with offenders to prevent crime. What do we mean by this statement and what basis do we have for making it? First, by “constructive approaches to working with offenders” we mean bringing the use of effective methods and techniques of behaviour change into work with offenders. Indeed, this view might pass as a definition of forensic clinical psychology. Thus, our focus is the application of theory and research in order to develop practice aimed at bringing about a change in the offender’s functioning. The word constructive is important and can be set against approaches to behaviour change that seek to operate by destructive means. Such destructive approaches are typically based on the principles of deterrence and punishment, seeking to suppress the offender’s actions through fear and intimidation. A constructive approach, on the other hand, seeks to bring about changes in an offender’s functioning that will produce, say, enhanced possibilities of employment, greater levels of self-control, better family functioning, or increased awareness of the pain of victims.A constructive approach faces the criticism of being a “soft” response to the damage caused by offenders, neither inflicting pain and punishment nor delivering retribution. This point raises a serious question for those involved in working with offenders. Should advocates of constructive approaches oppose retribution as a goal of the criminal justice system as a process that is incompatible with treatment and rehabilitation? Alternatively, should constructive work with offenders take place within a system given to retribution? We believe that this issue merits serious debate.However, to return to our starting point, history shows that criminal justice systems are littered with many attempts at constructive work with offenders, not all of which have been successful. In raising the spectre of success, the second part of our opening sentence now merits attention: that is, “constructive approaches to working with offenders to prevent crime”. In order to achieve the goal of preventing crime, interventions must focus on the right targets for behaviour change. In addressing this crucial point, Andrews and Bonta (1994) have formulated the need principle:
P1: OTA/XYZ P2: ABC JWST073-seditor JWST073-McMurran
xiv
May 26, 2011
7:22
Printer Name: Yet to Come
SERIES EDITORS’ PREFACE Many offenders, especially high-risk offenders, have a variety of needs. They need places to live and work and/or they need to stop taking drugs. Some have poor self-esteem, chronic headaches or cavities in their teeth. These are all “needs”. The need principle draws our attention to the distinction between criminogenic and noncriminogenic needs. Criminogenic needs are a subset of an offender’s risk level. They are dynamic attributes of an offender that, when changed, are associated with changes in the probability of recidivism. Noncriminogenic needs are also dynamic and changeable, but these changes are not necessarily associated with the probability of recidivism. (p. 176)
Thus, successful work with offenders can be judged in terms of bringing about change in noncriminogenic need or in terms of bringing about change in criminogenic need. While the former is important and, indeed, may be a necessary precursor to offence-focused work, it is changing criminogenic need that, we argue, should be the touchstone in working with offenders. While, as noted above, the history of work with offenders is not replete with success, the research base developed since the early 1990s, particularly the meta¨ analyses (e.g. Losel, 1995), now strongly supports the position that effective work with offenders to prevent further offending is possible. The parameters of such evidence-based practice have become well established and widely disseminated under the banner of “What Works” (McGuire, 1995, 2008). It is important to state that we are not advocating that there is only one approach to preventing crime. Clearly there are many approaches, with different theoretical underpinnings, that can be applied. Nonetheless, a tangible momentum has grown in the wake of the “What Works” movement as academics, practitioners, and policy makers seek to capitalise on the possibilities that this research raises for preventing crime. The task that now faces many service agencies lies in turning the research into effective practice. Our aim in developing this Series in Forensic Clinical Psychology is to produce texts that review research and draw on clinical expertise to advance effective work with offenders. We are both committed to the ideal of evidence-based practice and we will encourage contributors to the Series to follow this approach. Thus, the books published in the Series will not be practice manuals or cook books: They will offer readers authoritative and critical information through which forensic clinical practice can develop. We are both enthusiastic about the contribution to effective practice that this Series can make and look forward to continuing to develop it in the years to come.
ABOUT THIS BOOK Although the context may be different, when it comes to practitioner skills in working with offenders, the same rules of good practice apply as they would with any other group. One of the bedrocks of practice lies in assessment and case formulation: The process of gathering information and then making sense of it, formulating it, within a given theoretical framework. Of course, formulation is independent of theory, so that meaningful formulations of, say, a given behaviour
P1: OTA/XYZ P2: ABC JWST073-seditor JWST073-McMurran
May 26, 2011
7:22
Printer Name: Yet to Come
SERIES EDITORS’ PREFACE
xv
can be made from different theoretical perspectives. The point of a case formulation is to guide practice based on the available case material, theoretical strictures, and extant empirical knowledge. In this book an impressive list of authors cover every angle concerning case formulation. The chapters range from discussion of the finer points of the theory and practice of case formulation, to applying these principles to specific offender and groups and to specific types of offences. There is a great deal to be taken at many levels from this book and we are pleased to see it as part of the growing body of work addition the Series represents. Clive Hollin Mary McMurran
REFERENCES Andrews, D.A. and Bonta, J. (1994) The Psychology of Criminal Conduct. Anderson, Cincinnati, OH. ¨ Losel, F. (1995) Increasing consensus in the evaluation of offender rehabilitation? Psychology, Crime, and Law, 2, 19–39. McGuire, J. (ed.) (1995) What Works: Reducing Reoffending. John Wiley & Sons, Chichester. McGuire, J. (2008) A review of effective interventions for reducing aggression and violence. Philosophical Transactions of the Royal Society, B, 363, 2577–97.
P1: OTA/XYZ P2: ABC JWST073-seditor JWST073-McMurran
May 26, 2011
7:22
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-preface JWST073-McMurran
May 26, 2011
7:21
Printer Name: Yet to Come
PREFACE Clinical case formulation is understood by forensic clinical psychologists and forensic psychiatrists to be key in designing appropriate and so potentially effective treatments for offenders. In forensic work, what is effective in treatment is usually taken to mean that an individual’s risk of reoffending is reduced. While treatments to ameliorate other problems are part of the work of forensic mental health professionals, they cannot ignore the expectation that their treatments should aim to reduce risk. This places an unusual burden upon this group of people in that they are to some degree responsible for their clients’ behavior and for any harm to others that this may cause. If case formulation is indeed the key to effective interventions, then it is imperative that it should be done well. Research into some of the basic issues in case formulation is lacking in the forensic literature and this lack urgently needs to be addressed. Fortunately there is some evidence about reliability, validity, and utility from clinical work in general, and we present this information here for forensic practitioners to draw upon. Additionally, there are forensic practitioners who have given a great deal of thought to the principles upon which forensic case formulation should rest. These ideas are also presented in this book. We are indebted to the authors who have made such excellent contributions to this volume. We hope that by drawing together this body of work we might create an impetus for further research in this important area. Peter Sturmey Mary McMurran December 2010
P1: OTA/XYZ P2: ABC JWST073-preface JWST073-McMurran
May 26, 2011
7:21
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
PART I
GENERAL ISSUES
8:16
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
Chapter 1
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION TRACY D. EELLS University of Louisville, USA
KENNETH G. LOMBART University of Massachusetts Lowell, USA
Our task in this chapter is to introduce the concept of case formulation. We begin by discussing the definition, functions and goals of case formulation, including why formulation is important. We continue by reviewing theoretical and evidentiary sources of information to guide the development of a formulation. Next, we summarize several structured case formulation models that have been developed to increase reliability and validity. Finally, we propose a general framework the therapist can use to structure a formulation and conclude with some practical tips.
WHAT IS A CASE FORMULATION? Our working definition of case formulation comes from a cross-theoretical perspective: “A psychotherapy case formulation is a hypothesis about the causes, precipitants, and maintaining influences of a person’s psychological, interpersonal and behavioral problems” (Eells, 2007, p. 4). A formulation involves inferences about predisposing vulnerabilities, a pathogenic learning history, biological or genetic factors, sociocultural influences, currently operating contingencies of reinforcement, conditioned stimulus–response relationships, or schemas, working models, and beliefs about the self, others, the future or the world. The aim of the formulation is to explain the individual’s problems and symptoms. The specifics of the formulation will vary depending on the theoretical orientation of the case formulator. As a hypothesis, a formulation is always subject to empirical test and to revision as new information becomes available. Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
4
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
A case formulation serves multiple functions (Eells, 2007). First, it provides a structure to organize information about a person and his or her problems. Clients produce enormous amounts of information in therapy, including verbal, behavioral, prosodic, gestural, affective, and interactional. Formulation facilitates the management of this information cascade. Second, formulation provides a blueprint guiding treatment. Its primary purpose is to help the therapist develop and implement a treatment plan that will lead to a successful outcome. The formulation therefore enables the therapist to anticipate future events, for example, therapy-interfering events, and to prepare for them. Third, a formulation serves as a gauge for measuring change. Indices to assess change may come from goals included in the formulation, from relief of problems identified in the formulation, or from the revision of an inferred explanatory mechanism that did not seem adequate when tested. Fourth, a formulation helps the therapist understand the patient and thereby exhibit greater empathy for the patient’s intrapsychic, interpersonal, cultural, and behavioral world. Kuyken, Padesky and Dudley (2009) offer another definition of case formulation, emphasizing its collaborative and resilience-building aspects. They define formulation as a “process whereby therapist and client work collaboratively first to describe and then to explain the issues a client presents in therapy. Its primary function is to guide therapy in order to relieve client distress and build client resilience” (p. 3). Using the metaphor of a crucible and focusing on cognitivebehavioral therapy (CBT), these authors emphasize that formulation integrates and synthesizes a client’s problems with CBT theory and research. Essential ingredients of a productive conceptualization are empirical collaboration between therapist and client, the development of the formulation over time from the descriptive level to an explanatory level, and the elicitation of both client strengths and problems. These authors also describe functions of a CBT case formulation. These include (1) synthesizing client experiences, relevant CBT theory and research; (2) normalizing and validating clients’ presenting issues; (3) promoting client engagement; (4) making complex and numerous problems more manageable for the client and therapist; (5) guiding the selection, focus, and sequence of interventions; (6) identifying strengths and suggesting ways to build resilience; (7) suggesting cost-efficient interventions; (8) anticipating and addressing problems in therapy; (9) helping the therapist understand nonresponse to therapy; and (10) facilitating high-quality supervision. Persons (2008) embeds her approach to formulation within a framework of clinical hypothesis testing. She emphasizes that the formulation is fundamentally a hypothesis that is constantly refined in the course of treatment. She views a complete formulation as one that ties the following elements together into a coherent whole: (1) the patient’s symptoms, disorders, and problems, (2) hypotheses about the mechanisms causing the disorders and problems, (3) precipitants of those disorders and problems, and (4) a statement of the origins of the mechanisms. Following similar lines, Tarrier and Calam (2002) define formulation as “the elicitation of appropriate information and the application and integration of a body of theoretical psychological knowledge to a specific clinical problem in order to understand the origins, development and maintenance of that problem. Its purpose is both to provide an accurate overview and explanation of the patient’s problems that is open to verification through hypothesis testing, and to arrive collaboratively with the
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
5
patient at a useful understanding of their problem that is meaningful to them” (pp. 311–12). The case formulation is then used to inform treatment or intervention by identifying key targets for change.
WHY FORMULATE? Multiple mental health care disciplines view case formulation as an essential clinical skill. A core competency for psychiatrists trained in the United States is the ability “to develop and document an integrative case formulation that includes neurobiological, phenomenological, psychological and sociocultural issues involved in diagnosis and management” (American Board of Psychiatry and Neurology, 2009, p. 1). Similarly, the American Psychological Association promotes evidence-based practice, which includes the application of “empirically supported principles of psychological assessment, case formulation, therapeutic relationship, and intervention” (APA Presidential Task Force on Evidence-Based Practice, 2006, p. 284). The British Psychological Society views formulation as a core skill (Division of Clinical Psychology, 2001, p. 2). Multiple authors support the importance of case formulation as a “lynchpin concept” (Bergner, 1998), the “first principle” underlying therapy (J. S. Beck, 1995) and the “heart of evidence-based practice” (Bieling and Kuyken, 2003). Formulation is a core skill for several reasons. First, and most importantly, formulation is where theory and empirical knowledge about psychotherapy, psychopathology, personality, development, culture, and neurobiology merge to inform the understanding and treatment of an individual, group, couple, or family. Formulation provides a structure to apply nomothetic knowledge to an idiographic context. Second, current nosologies are almost exclusively descriptive and symptomfocused. Thus, they provide no account of why a client has symptoms, what the origins of those symptoms are, and what triggers and maintains them. Major depressive disorder, one of the most commonly diagnosed disorders, is a case in point. According to the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (American Psychiatric Association, 1994, DSM-IV) to be diagnosed with this condition, one must meet five of nine criteria for two weeks, including depressed mood or loss of interest or pleasure. In addition, one must exhibit distress or impairment in one’s social or occupational functioning and meet other rule out criteria. The criteria say nothing about biochemical, psychological, behavioral, situational, or environmental factors that may be producing the depression. Formulation fills this explanatory gap between diagnosis and treatment. A third reason that formulation is essential is that diagnosis alone does not provide a sufficient guide to treatment selection. The same diagnosis might be treated with different types of empirically defensible treatments and interventions, creating the dilemma of which one to choose. Further, few psychotherapy outcome studies include diagnosis by treatment interactions and thus do not address the sensitivity and specificity of treatment for a specific diagnosis (Sturmey, 2008). A single treatment that is found effective for one diagnosis may also be effective for other diagnoses. Fourth, a case formulation approach tailors treatment to address individual circumstances. Empirically supported treatments (EST) do not provide guidance in
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
6
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
a number of situations (Persons, 2008). These include when the client has multiple disorders and problems, when multiple providers are treating the individual, when a situation arises that is not addressed by an EST, when no EST is available, when the client does not adhere to an EST, when establishing a collaborative therapeutic relationship proves problematic, and in cases of treatment failure. With regard to the latter point, as many as 40–60% of individuals do not respond to a first-line empirically supported treatment (Westen, Novotny and Thompson-Brenner, 2004).
THE GOALS OF FORMULATION If a formulation is to serve the above functions, it should meet at least five goals. First, a formulation should be accurate and fit the individual for whom it is constructed. The benefits of an accurate formulation have been demonstrated in a number of studies (Crits-Christoph, Cooper and Luborsky, 1988; Crits-Christoph et al., 2010; Silberschatz, 2005b). One way to assess accuracy of an individual formulation is to evaluate the patient’s response to a formulation-consistent intervention and to compare those responses to how the patient responds to formulation-inconsistent interventions. If the patient responds as the formulation predicts, one has evidence of its accuracy. Another way to assess accuracy is to share the formulation with the patient and get the patient’s opinion. Opinions vary as to whether and to what degree a formulation should be shared with a patient. CBT therapists tend to prefer sharing the formulation and see this as an important component of developing a collaborative relationship with the patient (Kuyken et al., 2009) More psychodynamically oriented therapists have expressed caution in sharing the formulation. Luborsky and Barrett (2007) advise sharing it in its component parts rather than as a whole. Curtis and Silberschatz (2007) advise deciding whether to share or not on the basis of what the formulation predicts the patient’s response will be. Ryle’s (1990), cognitive-dynamic model, on the other hand, includes sharing the formulation, composed as a letter from the therapist to the patient, as part of treatment. A second goal of formulation is that it have treatment utility (Hayes, Nelson and Jarrett, 1987). The formulation should contribute to the treatment beyond what would have been achieved in the absence of a formulation. One measure of utility is the contribution of the formulation to treatment outcome. There is little research in this area, and research that has been done has produced equivocal results (Bieling and Kuyken, 2003; Kuyken, 2006). Another index of treatment utility is the extent to which the formulation benefits the process or efficiency of the delivery of the therapy. Further, a formulation may have benefits for the therapist that filter indirectly to the patient and therapeutic process, for example by increasing the therapist’s confidence or improving his/her communication with the client. For example, Chadwick, Williams and Mackenzie (2003) found that while formulation-guided therapy did not predict alliance ratings among a group of psychotic patients, it was associated with improved therapist ratings of the therapeutic relationship. A third goal of formulation is that it should be parsimonious yet sufficiently comprehensive. Some problems and clients require relatively simple and circumscribed formulations whereas others need multifaceted and complex formulations,
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
7
especially when the client behaves in contradictory ways, meets criteria for multiple disorders, or has major problems in multiple spheres of functioning. The formulation should provide a structure to optimally and efficiently represent enough information about the patient to benefit treatment, but not more. A fourth goal of formulation is to strike the right balance between description and explanation. Research has shown that it is difficult to achieve good reliability when formulations are based on psychological constructs that are too distant from the experience and behavior of the patient (Seitz, 1966). On the other hand, if a formulation is to be genuinely explanatory, it must do more than summarize biographical information about a client. Notwithstanding this distinction, it is noteworthy that description and explanation can blur as one proposes an underlying mechanism. As Kazdin (2008, p. 12), wrote, “Depending on the detail, level of analysis, and sequence of moving from one to the other, description can become explanation” (p. 12). A final goal of formulation is that it should be evidence-based. The APA Task Force on Evidence-Based Practice in Psychology stated that evidence-based formulations apply the best research, knowledge, experience, and expertise to the task: What constitutes appropriate evidence in a case formulation? Various types of evidence may best be viewed in relative terms along a continuum. At the most clearly evidence-based end, one could imagine compelling outcomes from empirically supported treatments, well-demonstrated mechanisms underlying forms of psychopathology, powerfully predictive epidemiological data, or well-documented and replicated findings about basic psychological processes, for example, the age at which reliable autobiographical memories can be formed. At the other end of the continuum one might place a therapist’s hunches or intuitions. These might offer valuable insights that could be tested, but in themselves probably would not be described as evidence-based by most observers. Between these two end-points might be included data such as psychological test findings, rating scale results, a patient’s narrative of a relationship episode, a dream account, a thought record, a patient’s account of automatic thinking or an assertion by the client or therapist that a thought is a core belief. No consensus currently exists on what constitutes appropriate evidence for a case formulation. Therefore, our advice is that therapists create a plausible continuum and use their best judgment in evaluating evidence they gather as they formulate cases. If the above five goals of case formulation are met, the therapist is well on the way toward developing a productive tool to facilitate treatment. In the following section, we discuss two major sources of hypotheses about clients: theory and evidence.
THEORY AS A GUIDE TO FORMULATION Earlier we stated that the most important reason to formulate a case is because it provides an opportunity to apply theory and evidence to a specific case. In this section, we provide an overview of some primary sources of theory, illustrating the application of these sources to case formulation. We begin with four major theories underlying broad models of psychotherapy: psychodynamic, cognitive, behavioral, and humanistic.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
8
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Psychodynamic Theory Psychodynamic theory originates in the work of Freud and provides a rich source of inference for case formulation. Beginning with his early formulation that “hysterics suffer from reminiscences” (Breuer and Freud, 1955), Freud has contributed a multitude of ideas that have shaped our understanding of normal and abnormal psychology. Most prominently, these include the notion of psychic determinism and unconscious motivation. The former entails the assumption that all human thought has a specific cause, nothing is random or accidental. The latter is the idea that majority of mental activity is outside of awareness and is goal-directed or purposeful. Other ideas contributed by Freud are that of overdeterminism, the symbolic meaning of symptoms, symptom production as a compromise formation, ego defense mechanisms as stabilizers of the psyche, and the tripartite theory of the mind, that is, its division into id, ego, and superego. Messer and Wolitzky (2007) succinctly grouped contemporary psychodynamic theory, at least as practiced in North America, into three broad categories: the traditional Freudian drive/structural theory, object relations theory, and self-psychology. We will briefly describe each with a focus on what is formulated and why. The drive/structural theory proposes that human behavior is driven by intrapsychic conflict originating in sexual and aggressive drives that seek pleasure and avoid pain (the “pleasure principle”) but become thwarted when they confront obstacles such as fear, anxiety or guilt. The structural component of the drive model involves the tripartite division of the mind into the id, which is the repository of drives, the superego, which contains both our conscience and who we ideally would become (the “ego ideal”), and the ego, which mediates between the impulses of the id and the strictures of the superego. The ego utilizes defense mechanisms in an attempt to avoid anxiety and maintain psychic equilibrium. When these attempts fail, neurotic symptoms develop. These mental structures and specific defenses arise as the individual navigates through four psychosexual stages – oral, anal, phallic, and genital – each of which is associated with specific conflicts that if not resolved persist into adulthood. The key feature of a case formulation based on the Freudian drive/structural theory is an “emphasis on unconscious fantasy, the conflicts expressed in such fantasy, and the influence of such conflicts and fantasies on the patient’s behavior”, and further, the assumption that these conflicts originate in childhood (Messer and Wolitzky, 2007). Treatment focuses on helping patients appreciate the nature and pervasiveness of their unconsciously driven motives and the ways that they avoid awareness of them. The object relations perspective on psychodynamics focuses on mental representations of self and other and models of affect-laden transactions between the two. The approach tends to dichotomize self and other into “good” and “bad” components that are often viewed as compartmentalized and not integrated. Defense mechanisms such as projective identification, splitting, and role reversal are used frequently by practitioners of this perspective. Relationships constitute basic drives rather than instinct. Case formulations based on this perspective focus on this inability to integrate, the disavowal of rage toward attachment figures that are also loved and needed. The individual may project an image of self as “good” while projecting the “bad” onto others.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
9
The self-psychology (Kohut, 1971, 1977) perspective emphasizes the development and maintenance of a cohesive and coherent sense of self. Kohut viewed the self as the center of intention and experience, as the core of our being (GalatzerLevy, 2003, p. 479). Cohesion refers to a sense of the self as maintaining continuity across time and place. Temporal coherence is the experience of oneself as a person with sameness and history across time. Spatial coherence refers to the sense that various aspects of oneself are alive and share a common intention. Kohut’s primary tool for understanding others was through empathic connection and comprehension. He viewed empathy as the ability to understand another’s psychological experience, as a kind of vicarious introspection. Using this approach, he identified a number of disturbances in the development of self in his patients. For example, they seemed to experience “empty” depressions, in which life appeared colorless, alienating, pointless, and lacking in vitality. Others experienced traumatic states in which experiences could not be integrated into a coherent sense of self. Kohut also treated people subject to seemingly unexpected, situationally discrepant states of rage. Kohut explained these experiences in terms of caretakers’ failure to provide sufficient empathic responsivity to enable one to develop a cohesive sense of self. One of Kohut’s most distinctive concepts is that of the “selfobject”. He posited that the presence of others in one’s life is an essential prerequisite for mental wellbeing. A selfobject is an unconscious mental representation of a connection between self and other, as if the other is an extension of oneself. He identified two basic types of selfobject: idealized and mirroring. An idealized selfobject is revealed in the experience of feeling alive, vital and powerful through one’s connection to another whom one admires. As Messer and Wolitzky write, one with an idealized selfobject seems to be saying, “I admire you, therefore my sense of self and self-worth are enhanced by my vicarious participation in your strength and power.” A mirroring selfobject vitalizes the self through the sense of being affirmed by others to whom one feels connected. Messer and Wolitzky characterize the mirroring selfobject as, “You admire me, and therefore I feel affirmed as a person of worth.” Formulations from the self-psychology perspective emphasize explanations of disturbances in a cohesive sense of self due to failures of empathic responsiveness from caretakers. The nature of the patient’s transference to the therapist – as idealizing or mirroring – is an important component to understanding the patient. Practitioners of psychodynamic therapy can draw from any or all of these basic perspectives in drawing up a case formulation; however, according to Messer and Wolitzky (2007), who in turn draw from Rapaport and Gill (1959), a comprehensive contemporary psychodynamic case formulation should contain five components. First, it should address the patient’s major dynamic conflicts, for example, between wishes and the feared consequences of those wishes. Second, it should address those aspects of the patient’s personality involved in the conflicts, for example, the id, ego, supergo, or inferred selfobjects. Third, the formulation should address the antecedent and developmental events leading to the conflicts. For example, what were the crucial experiences in childhood that gave rise to the patient’s current concepts of self and others? Or, what were the episodes of failed empathic responsiveness on the part of caretakers that led to a disturbance in self cohesion? Fourth it should address the adaptive and maladaptive compromise formations that comprise the patient’s defensive and coping strategies. Which compromises
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
10
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
are relatively successful accommodations to the conflicts and which ones do not and lead to symptoms? Finally, the formulation should state the degree of conscious awareness of the conflicts.
Cognitive Theories Theories underlying contemporary cognitive therapies can be traced to the “cognitive revolution”, which took place in the mid-twentieth century as a response to what was increasingly perceived as the inadequacies of behavioristic, stimulusresponse models of learning that discounted the role of mentation and human agency (Mahoney, 1991). Borrowing terminology and concepts from information theory, computer science, and general systems theory, the interests of cognitive scientists turned toward “understanding and influencing the fundamental processes by which individual humans attend to, learn, remember, forget, transfer, adapt, relearn and otherwise engage with the challenges of life in development” (Mahoney, 1991, p. 75). As Bruner (1990) put it retrospectively, “that revolution was intended to bring ‘mind’ back into the human sciences after a long cold winter of objectivism” (p. 1). It was further intended “to establish meaning as the central concept of psychology – not stimuli and responses, not overtly observable behavior, not biological drives and their transformation, but meaning” (p. 2). Influential writings at the time included works by Bruner (e.g., Bruner, Goodnow, and Austin, 1956), Chomsky (1959), Festinger (1957), Kelly (1955), Postman (1951), and Simon and Newell (1958). As the cognitive revolution filtered into the social sciences and psychiatry, multiple theories of cognitive therapy took shape. More than 15 years ago, Kuehlwein and Rosen (1993) identified ten different models of cognitive therapy alone. As Nezu, Nezu and Cos (2007) pointed out, there is no single cognitive therapy, but rather a collection of therapies that share a common history and perspective. They hold in common not only their heritage within the cognitive revolution, but also the assumption that our appraisals of events are much more crucial to our mental well-being than are the events themselves. In this section we will review some of these theories and discuss their implications for formulation. In doing so, we recognize that most of these models also blend elements of behavior theory, which will be discussed later in the chapter. With regard to cognitive theories, we will emphasize Beck’s model since it is the most influential and has been subject to the most empirical scrutiny. Beck’s (1963) cognitive theory originated from observations of persistent thought patterns in depressed patients he interviewed. These individuals expressed views of themselves as inferior in areas of their lives that mattered to them. They viewed the world as depriving and saw the future as bleak. These observations led Beck to develop his now well-known “cognitive triad”, which is a framework he proposed to describe the automatic and systematically biased thinking of depressed patients. It was later expanded to describe a wide range of problems and psychological conditions. Automatic thoughts are brief, episodic, and often emotionally laden forms of thinking that occur unbidden and are often at the threshold of awareness. For example, one might think, “Writing this chapter is too hard. I’ll
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
11
never get it done,” which could be followed by a feeling of deflation or demoralization. Negative automatic thoughts are often erroneous, illogical, and unrealistic. Beck identified specific characteristic forms of thought distortion. Examples are arbitrary inferences, selective abstraction, overgeneralizations, catastrophizing, and personalization in which one erroneously explains events in terms of one’s own perceived shortcomings rather than considering other explanations (A.T. Beck, 1963; J.S. Beck, 1995). In addition to the cognitive triad and the notion of cognitive distortions, a third major characteristic of Beck’s cognitive theory is the idea of schemas. These refer to tacit, organized cognitive structures that influence perception and appraisal. The schemas give rise to beliefs about the self, world and future. At the most fundamental level are “core beliefs” (J.S. Beck, 1995), which are the most fundamental layer of beliefs and are assumed to develop in childhood and to be global, rigid and overgeneralized. In their negative form they tend to focus on beliefs of helplessness or unlovability. Between core beliefs and situationally specific automatic thoughts lie “intermediate beliefs”, which are rules, attitudes and assumptions that are more subject to revision and change than core beliefs but less so than automatic thoughts. The ideas reviewed above are relevant to cognitive case formulation in that formulation within a cognitive model entails identifying the client’s automatic thoughts, intermediate beliefs and core beliefs (J.S. Beck, 1995). Second, the assumption that characteristic patterns of thinking are specific to diagnostic categories suggests that implicit nomothetic explanatory mechanisms underlie diagnoses and can serve as templates for formulations (Persons, 2008). If the template fits the client, an empirically supported treatment may be suitable for the individual in question. Other cognitive theories of therapy have also developed since the cognitive revolution. These include those of Ellis (1994; 2000), Young (1990); Young et al. (2003), and Hayes and Strosahl (2004). A distinctive style of formulation can be identified from each of these approaches.
Behavioral Theories Behaviorism offers a rich theoretical source of ideas for case formulation. It represents a departure from the structuralism of the cognitive approach (Sturmey, 2008). The previous approaches all posit the existence of presumed cognitive structures that influence behavior, cognition and affect. Rather than viewing behavior as primary data, cognitive approaches see it as a derivative of unseen mental structures. One problem with structurally based explanations is that they may be based on circular reasoning. As Sturmey (2008) writes, “Cognitive psychologists use behavior to infer the presence of the unobservable structures . . . then use the unobservable structure to explain the observable behavior” (p. 9). Behaviorists have made three distinct contributions to the field of case formulation (Eells, 2007). First, consistent with their emphasis on observable behavior, they place primary emphasis on understanding and modifying symptoms. Using functional analysis, they examine the antecedents and consequences of symptoms
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
12
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
in order to change them. They also look at what events elicit and reinforce symptomatic behavior. From this standpoint, behaviorism is an inherently practical approach to helping people. Second, behaviorists emphasize the influence of the environment on behavior. A behavior analysis always examines what characteristics of an individual’s immediate surroundings may be affecting behavior. The presence of alcohol in the home of a person with alcohol dependence, for example, is likely to increase the chance of continued abuse. Removing the alcohol can play a major role in helping the person. This environmental emphasis is also less stigmatizing since it rejects the assumption that the source of problems is something inherent within the individual. A third contribution of behaviorism to case formulation is its emphasis on empirical assessment to test a formulation. Evidence of the accuracy of a behavioral assessment is readily determinable since behavioral tests can be run. For example, Wilder (2009) hypothesized that delusional behavior exhibited by an young male with schizophrenia could function either as attention-seeking behavior, as an attempt to escape work, or as self-stimulating activity. In developing his formulation, he devised separate behavioral tasks to determine which of these hypotheses is supported. Behavioral approaches can be broadly categorized as based on operant or respondent learning. In the following two sections, we will describe each, showing their application to formulation.
Operant Conditioning Operant conditioning models focus on the antecedents and consequences of behavior. For example, habit reversal is a technique based on operant learning that is intended to reverse problematic behavior such as trichotillomania and tics by identifying the antecedent and consequent reinforcers of these behaviors then changing them to eliminate the behavior. Case formulation from the operant conditioning perspective involves conducting a functional analysis of behavior. Skinner (1953, p. 35) defined functional analysis as follows: The external variables of which behavior is a function provide for what may be called a causal or functional analysis. We undertake to predict and control the behavior of the individual organism. This is our “dependent variable” – the effect for which we are to find the cause. Our “independent variables” – the causes of behavior – are the external conditions of which behavior is a function. Relations between the two – the “cause-and-effect relationships” in behavior – are the laws of a science. (p. 35)
Since Skinner, the term has been expanded to describe a wide range of interventions. Functional analysis is at the core of most behavioral case formulation approaches and some cognitive-behavioral approaches (Haynes and Williams, 2003; Nezu, Nezu and Cos et al., 2007; Persons, 2008). Functional analysis should take into account several aspects of operant conditioning. These include establishing operations (such as satiation or deprivation states), adaptive and maladaptive
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
13
shaping, adaptive and maladaptive extinction, modeling, chaining, avoidance and escape activity that may preempt positively reinforced activity, consequences of debilitating naturalistic schedules of reinforcement, punishers, and variability of behavioral repertoires (Ferster, 1973; Sturmey, 2008). As Sturmey (2008) notes, the operant conditioning framework provides a structure for case formulation since operant learning is involved in the acquisition and maintenance of many forms of maladaptive behavior. For example, a depressed individual may withdraw interpersonally, missing out on reinforcers that would counter depressive affect. In addition, others may avoid the depressed person, thus maintaining maladaptive avoidance and isolation. A case formulation based on operant conditioning should assess these possibilities and identify the contingences that may be maintaining the problematic behavior. It could also include hypotheses about why current contingencies do not support adaptive behavior and why contingencies that once supported independence are no longer present. Clinicians could also consider schedules of reinforcement operable in the client’s life. A variable ratio schedule should result in higher rates of responding than a fixed ratio schedule. Consequently, a clinician observing high rates of behavior might infer the presence of variable ratio schedules. Conversely, low rates of desired behavior may indicate reinforcement schedules for these behaviors that are weak or perhaps punishing. The clinician can evaluate not only the rate and frequency of reinforcement but also how immediately the consequences occur and whether they comprise primary or secondary reinforcement. The clinician can also assess whether behaviors are or are not under stimulus control. For example, a chronically anxious client may lack stimulus control of relaxation (Sturmey, 2008). Interventions can be planned accordingly, for example, teaching the client to take deep breaths, think pleasant thoughts, or engage in imagery while present in anxiety arousing environments. As a final example, behavioral chaining can be analyzed and treatment plans developed to help the client learn alternative behavior (Koerner, 2007).
Respondent Conditioning In contrast to operant behavior, which is controlled by its consequences, respondent behavior is elicited by its antecedents. The classic example is that of Pavlov’s dogs who were trained to salivate at the sound of a bell. This was accomplished by pairing the presentation of meat, which elicited salivation, with the presentation of the bell. When the pairings occurred enough times, the bell alone could elicit salivation. The meat is considered to be an unconditioned stimulus (US) and salivation an unconditioned response (UR). The bell came to serve as a conditioned stimulus (CS) that could elicit what is now considered a conditioned response (CR), that is, the salivation. Respondent behavior is said to be rooted in responses that are naturally occurring as a result of our evolutionary past. Examples of unconditioned responses are fear at the site of a genuine threat to life, hunger when food is present, startling in response to a loud sound, and recoiling from a bitter smell. These responses share the characteristic of being unlearned. They can all, however, be brought under the control of other stimuli through pairings, such as the bell with the meat in the case of Pavlov’s dogs. For example, repeated exposure to gunfire (UCS) in war settings setting off a startle response can create an exaggerated startle
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
14
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
response in a veteran such that the sound of a car door closing (CS) elicits a startle response (CR), as well as fear associated with the war experience. Respondent conditioning has been associated with many psychological disorders, including post-traumatic stress disorder, phobias, and obsessive-compulsive disorder. There are several principles of classical conditioning that one can use to understand how psychological disorders develop, are maintained and may be treated (Persons, 2008; Sturmey, 2008). These, in turn, can be incorporated into a case formulation. One principle is that the greater number of pairings of a CS and a UCS, the more likely the CS is to elicit a CR. For example, the more often one experiences a spontaneous panic attack (a UCS that elicits a UCR of fear) while at a restaurant (CS), the more likely visiting a restaurant may elicit a panic attack (which is now a CR). Another principle is that when a CS occurs repeatedly in the absence of a UCS the CS exerts less and less control over the CR. This is the principle underlying the behavioral technique of flooding, which has been used to treat phobias and other anxiety disorders. Flooding involves repeated exposure to a CS (e.g., plastic spiders, heights, public speaking) until it is no longer able to elicit a CR (fear). A third principle is that counterconditioning, or elimination of a CR, occurs when one pairs a CS to a UCS that elicits a new response that is incompatible with the old one. This is the principle that underlies Wolpe’s systematic desensitization technique for treating phobias and anxiety. Wolpe held that one cannot simultaneously experience relaxation and fear. In systematic desensitization, one first teaches the patient relaxation exercises. Then, when the patient is relaxed, he or she is exposed to increasing levels of anxiety arousing experiences until those experiences no longer elicit anxiety. Case formulation from the standpoint of respondent conditioning has several components. First, the therapist must identify events that serve as the UCS, CS, US and CR. Second, the therapist should be alert to how these can be affected by other factors. Third, the clinician should consider the relationship between stimulus and response pairings. For example Bouton (2002) summarized evidence that extinction of a CS–CR pairing does not eliminate a link to the UCS, but rather establishes alternate, benign associations to the CS. If true, extinction is rarely permanent, an important consideration in treatment planning. Fourth, the therapist should inquire closely into the patient’s actual experience when symptomatic behavior occurs. Presumed exposure to a CR, for example, may not be what it appears to be. To illustrate, Behar and Borkovec (2006) propose that generalized anxiety disorder (GAD) persists despite the patient’s repeated exposure to anxiety arousing events due to compensatory mechanisms aimed at psychologically avoiding the CR. For example, GAD patients tend to worry or ruminate rather than immerse themselves experientially in the threatening situation. Were they to do so repeatedly, the consequent exposure would theoretically lead to extinction.
Humanistic Theory Humanistic theory emerged in the 1950s as an alternative to the determinism of the psychodynamic and behavioral approaches current at the time. In contrast to the
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
15
view that humans are the inevitable product of their reinforcement history and environment or of their unconscious minds, the humanistic framework sees humans as self-actualizing and goal-directed. The task of therapy is to provide a nondirective, empathic and supportive environment in which the client can recapture his self-actualization tendency. From this standpoint, formulation or “psychological diagnosis” was de-emphasized and viewed as potentially detrimental to the therapeutic process (Rogers, 1951). As Rogers (1951) wrote,
the very process of psychological diagnosis places the locus of evaluation so definitely in the expert that it may increase any dependent tendencies in the client, and cause him to feel that the responsibility for understanding and improving his situation lies in the hands of another. (p. 223)
In addition, to the extent that the client comes to see the therapist as the only person who can really understand him, there is “a degree of loss of personhood” (p. 224). A second objection to formulation from the humanistic point of view is based on social and philosophical grounds: “When the locus of evaluation is seen as residing in the expert, it would appear that the long-range social implications are in the direction of the social control of the many by the few” (p. 224). Notwithstanding these objections, a distinct theory of personality emerged from the humanistic standpoint that can be formulated. Rogers posited that human nature is driven by one master motive: the self-actualizing tendency, which is an inherent drive to survive, grow and improve. Further, we all live in a subjective world through which we assess what is consistent or inconsistent with selfactualization. The self emerges from experience, and develops positively when met with unconditional positive regard from others. When it is not, incongruence develops as an individual no longer grows in a manner consistent with the self-actualizing tendency. The self as experienced is incongruent with the real or genuine self. The task of therapy, therefore, is to facilitate greater congruence. When collaboratively developed, formulation can potentially facilitate such a process. Other theories identified within the humanistic tradition have been developed by Maslow (1987), Kelly (1955), Perls, Hefferline and Goodman (1965) and more recently, by Greenberg (2002) and Bohart and Tallman (1999), among others. It is noteworthy that contemporary proponents of the humanistic school are more accepting of formulation as a useful tool in therapy, although the emphasis tends to be on formulating moment-by-moment experiences rather than developing a global case formulation (Greenberg and Goldman, 2007). As noted elsewhere (Eells, 2007), the primary contributions of humanistic psychology to formulation include its emphasis on the client as a person instead of a disorder, the focus on the here-and-now aspect of the human encounter rather than an intellectualized “formulation”, and its view of the client and therapist as equal collaborators. An additional contribution of the humanistic approach is its emphasis on humans as capable of self-determination and free choice.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
16
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Eclectic Approaches Before leaving this section on theoretical contributions to case formulation, we note that a number of case formulation approaches and theories about the development of psychological disorders blend two or more of the approaches we have described. One example based on animal research is Mowrer’s (1960) theory of the development and maintenance of fear. He posits that respondent conditioning establishes fear and operant conditioning maintains it through negative reinforcement of avoidance responses. This theory is the basis of modern exposure-based treatments of phobias and other anxiety disorders. Wachtel (1977) gives other examples of how common disorders can be viewed compatibly within both the behavioral and psychodynamic perspectives. The combination of cognitive and behavioral approaches is also characteristic of several structured case formulation approaches.
EVIDENCE AS A GUIDE FOR FORMULATION As noted above, the APA Task Force on Evidence-Based Practice recommends that systematic case formulations be based on empirically supported principles. In this section we describe five sources of evidence that can guide case formulation. One draws from the patient, one from the psychometric tradition, and three from the base of empirical knowledge within psychology.
The Patient as a Guide When discussing definitions of case formulation we emphasized their hypothetical nature. That is, a formulation should be considered a hypothesis to be revised as indicated and warranted. It must be tested against the patient’s response to interventions based on it. Evidence from the patient can include (1) direct feedback when the formulation is presented by the therapist, (2) narratives the patient tells that either confirm or disconfirm the hypothesis, (3) dreams or fantasies the patient reveals in therapy, (4) changes in the patient’s symptoms based on interventions consistent with the formulation, and (5) autobiographical information the patient discloses. Although the patient is a crucial source of information to refine and revise the formulation, the therapist should attempt to understand the material in the context of the scientific evidence base in psychology.
Psychometric Applications Psychometric data can provide useful information for case formulation. Studies have shown that structured interviews, personality inventories, and brief selfrated and clinician-rated measures provide incremental validity regarding diagnosis, assessment of psychopathology and personality, and prediction of behavior, although the contribution to case formulation validity itself is unexplored
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
17
(Garb, 2003). The use of symptom rating scales is recommended by a number of case formulation experts (Kuyken, Padesky and Dudley, 2009; Persons, 2008). These provide a time efficient, reliable and valid way of assessing the range of problems, current level of general distress, red flag issues (e.g., dangerousness), and social and adaptive functioning (A.T. Beck et al., 1988; A.T. Beck et al., 1961; Derogatis, 1983; Halstead, Leach and Rust, 2008; Lambert and Finch, 1999). Further, comprehensive personality tests such as the Minnesota Multiphasic Personality Inventory or the Personality Assessment Inventory can provide useful information for case formulation that allows the therapist to compare the patient’s responses against a standardization sample. Interview-based measures can also be helpful, for example the Structured Clinical Interview for DSM Disorders (SCID) (First et al., 1995; Spitzer et al., 1992).
Psychotherapy Process and Outcome Research Psychotherapy models investigated in efficacy studies contain implicit mechanisms of change and, thus, implicit case formulations. Since these implicit formulations are linked to outcome data, they can be useful starting points for individual formulations. Persons (2008) recommends that these implicit case formulations within empirically supported treatments serve as default nomothetic formulations that are then tailored for individual patients. One should be cautioned, however, that little is known about these presumed mechanisms. Kazdin (2007) has observed that although cognitive-behavior therapy is effective for depression, evidence suggests that symptom change occurs before a change in cognition, which runs counter to the model’s assumption that a change in cognition will lead to a change in symptoms. Improving our understanding of the processes involved in helping individuals with specific problems and diagnoses will be important for case formulation. As Kazdin (2008, p. 152) wrote, Evidence-based mechanisms of change could prove to be even more interesting or important than EBTs [evidence-based treatments]. We might be able to use multiple interventions to activate similar mechanisms once we know the mechanisms of change and learn how to optimize their use.
Psychopathology Research Research on psychopathological processes is also relevant for case formulation. The more we understand the predictors of psychopathology and the mechanisms that underlie, precipitate and maintain these conditions, the better we can plan treatment for them. One example is the role of rumination in depression (NolenHoeksema, Wisco and Lyubomirsky, 2008). Rumination as a thinking process is characterized by a perseverative, passive, and nonproductive fixation on symptoms of distress and the possible causes and consequences of the distress, but without any active attempt at problem solving. Nolen-Hokesema and colleagues have demonstrated that rumination exacerbates depression, enhances negative
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
18
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
thinking, impairs problem solving, erodes social support and interrupts instrumental behavior. Rumination predicts the onset of depression, may contribute to its course, and may also contribute to disorders such as anxiety, post-traumatic stress disorder, binge-eating, binge-drinking, self-harm, and maladaptive grief reactions. These researchers have also investigated methods to combat rumination, such as distraction and increasing awareness of its nonproductive and negative function. This research can inform case formulation and treatment planning. It helps the therapist recognize the seductive but deceptive nature of rumination as a phenomenon that gives the appearance of solving problems when in reality it is a problem in itself. Other examples include research on anxiety (Mineka and Zinbarg, 2006), on adverse effect of repressive coping on subjective well-being (DeNeve and Cooper, 1998), and on the function of psychotic symptoms (Freeman, Bentall and Garety, 2008).
Epidemiology Epidemiology is the study of “how disease is distributed in populations and of the factors that influence or determine its distribution” (Gordis, 1990, p. 3). It includes study of the causes of disease, including mental disorders, and associated risk factors, the extent of disease in a population, and the natural history and prognosis of disease. Unlike psychotherapy, which primarily focuses on the individual, epidemiology focuses on entire populations. Epidemiology can be helpful in case formulation in a number of ways. First, epidemiological information can sensitize the clinician to how psychological conditions are predicted by factors such as low socioeconomic status, general disease status, and neighborhood safety. This knowledge can help the clinician gain insights into the individual’s condition, assess prognosis, and plan interventions. Second, epidemiological information helps the clinician understand what is normative in a community. Deviations from this norm inform case formulation. Third, epidemiology can help the therapist form prognoses. Knowledge of the natural course of disorders such as depression (Kessler and Wang, 2009; Wells et al., 1992) or alcoholism (Vaillant, 1995), for example, helps a therapist predict risk and shape treatment. Fourth, epidemiological information can help the clinician predict comorbidity. Knowing that alcohol abuse commonly co-occurs with social anxiety (Randall et al., 2008), for example, should lead the therapist to thoroughly assess substance abuse in the socially anxious individual. Fifth, base rate information can help predict sources of problems. A patient with borderline personality disorder may claim to be a victim of ritualistic abuse, but even our imperfect knowledge of the prevalence of such activity can help the therapist put such claims into a probabilistic context (Frankfurter, 2006). In addition, knowledge of differences among psychological disorders related to age of onset, gender, ethnicity, and region facilitates the development of explanatory mechanisms. Tarrier and Calam (2002) noted that causal inferences in case formulation are more credible when based on epidemiological data relevant to base rates associated with the development of a disorder rather than the patient’s retrospective recall of life events. The latter form of inference risks tautology and is subject to error in retrospective recall. Sixth,
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
19
epidemiological data can help the clinician assess risk factors a patient faces. For example, knowledge of the relative risk factors for suicide attempts and suicide gestures can inform a case formulation and treatment plan (Nock and Kessler, 2006). Seventh, epidemiology data can help with treatment planning and motivation. Explaining risk of heart disease and diets, for example, can be part of treatment for obesity. Epidemiologically derived knowledge of the benefits of exercise, combined with a clinicians’ skill in developing behavioral plans, can combine to treat obesity. The seven sources of evidence just reviewed provide a broad knowledge base that can be paired with theoretical models. Together, they form the basis for a comprehensive formulation. Additionally, several structured case formulation models have been developed. These case formulation models can be used in developing formulations for individual clients. In the following section, we describe several of these structured case formulation models.
STRUCTURED SYSTEMATIC CASE FORMULATION MODELS Several decades ago, psychotherapy researchers and clinicians began developing systematic, structured methods of psychotherapy case formulation. They emerged as part of the need to develop systematic manuals to study psychotherapy research outcomes, as well as to facilitate clinical work. As more of these methods were developed, a major concern was that they be both reliable and valid. Reliability refers to the extent to which independent clinicians can develop similar formulations based on the same case material. Validity refers to the extent that the resulting formulations predicted events in therapy. Initial efforts to measure reliability were not encouraging (Seitz, 1966) as it appeared that therapists tended to focus on different aspects of case material, to make inferences that went too far beyond the available supporting data, and presented the formulation in formats that were difficult to compare. The newer structured case formulation methods produced much more reliable formulations owing to a number of features they shared. First, they structured the formulation by identifying preset categories of information necessary for the formulation. These include categories such as a problem list, core beliefs, schemas of self and other, relationship schemas, defense or coping styles, strengths/assets, and precipitants. Second, they involved relatively low-level inferences, often by linking inferences directly to case material such as therapy transcripts. There was no effort to infer “deep” psychological structures; rather, all inferences could be traced to biographical information or other statements or narratives provided by the client. Third, the process for case formulation was well-defined and structured. Finally, the therapists producing the formulations underwent training in the method. In the following section, we review some of these structured case formulation methods.
Core Conflictual Relationship Theme Based on the psychodynamic concept of therapeutic transference (Freud, 1958a, 1958b; Luborsky et al., 1991), the Core Conflictual Relationship Theme (CCRT) was
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
20
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
developed by Luborsky (1977), and is the earliest and most researched relationshipbased structured formulation model. The CCRT assumes that early interpersonal experiences predict later interpersonal relationship patterns. When early interpersonal experiences are traumatic, they serve as maladaptive interpersonal templates that harm the individual in later life. The CCRT is identified primarily by focusing on the person’s relationship narratives in therapy. From these narratives, the clinician identifies the client’s most common interpersonal wishes, the expected responses of others to those wishes, and, in turn, the responses of the self to the expected responses from others. The most frequent of these wishes and responses comprise the CCRT. A relatively simple and basic case formulation method, the CCRT is reliable and has convergent validity with similar, interpersonally focused methods (Luborsky and Barrett, 2007). It has been linked to therapy outcome and to symptom onset in therapy sessions; further, CCRTs tend to remain consistent longitudinally, across different relationships and throughout a course of therapy. They have also been associated with specific diagnoses and defense styles (Luborsky and Barrett, 2007).
Role Relationship Models Configuration The Role Relationship Model’s Configuration (RRMC) method expands upon the CCRT by, among other changes, positing a set of CCRTs formed into a configuration of wishes, fears, and compromises to those wishes and fears, and by adding inferences about the individual’s concepts of self and others (Horowitz, 2005, 1991b). The theoretical basis of the RRMC is person schemas theory (Horowitz, 1991a), which seeks to integrate elements of psychodynamic and cognitive theory. Person schemas theory assumes that an individual’s maladaptive interpersonal behavior patterns, including emotions, perceptions, memory, and actions in interpersonal situations, are organized by mental representations of the self, others, and the self with others. Like the CCRT, the RRMC has demonstrated good reliability and convergent validity.
Control Mastery Theory and the Plan Formulation Method of Case Formulation With roots in both psychodynamic and cognitive theory, Weiss’ control mastery theory (1993; Weiss and Sampson, 1986) begins with the assumption that humans have evolved to need stable attachments to others, a reliable conception of reality, and safety (Silberschatz, 2005a). From this starting point, Weiss asserts that psychopathology stems from “pathogenic beliefs” originating in traumatic childhood experiences. These beliefs are unconscious, powerful, emotion-laden, threatening, and emotionally distressing. They organize perception in close relationships throughout a person’s life and function to preserve stable relationships, but can also damage one’s personal development. Burdened by these pathogenic beliefs, individuals develop an adaptive and usually unconscious “plan” to disconfirm
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
21
their own pathogenic belief. The plan organizes behavior, including the choice to enter therapy. The goal of therapy is to facilitate the patient’s plan; therefore, it unfolds as a series of tests the patient engages in to determine whether the pathogenic beliefs can be safely abandoned. The Plan Formulation Method (Curtis and Silberschatz, 2005) is the case formulation model developed for therapy based on control mastery theory. Since planning treatment is highly individualistic, the development of an idiographic case formulation is essential. The formulation has the following components and steps: (1) identify traumas the patient has experienced; (2) infer the resulting pathogenic beliefs; (3) identify the “potential behaviors, affects, attitudes or capacities” (Curtis and Silberschatz, 2005, p. 89) the patient would like to adopt, in other words, the patient’s goals; (4) predict the “tests” the patient will employ in therapy to disconfirm pathogenic beliefs; and (5) identify the insights or knowledge to be acquired during therapy that will help the patient achieve his or her goal. The Plan Formulation Method has been demonstrated to have excellent reliability and predicts both process and outcome events in therapy (Silberschatz, 2005b).
Beck’s Cognitive Case Formulation Method Beck (1995) developed a basic formulation approach for cognitive therapy. The formulation links automatic thoughts to deeper-level beliefs and the experiences that led to their development. The therapist first identifies automatic thoughts and their associated emotions and behavior and then links these thoughts to compensatory strategies, intermediate beliefs such as assumptions and rules, and core beliefs. The core beliefs are traced to experiences that contributed to their development and maintenance. When the formulation is complete, the therapist has mapped out past experiences that led to core beliefs, the resulting intermediate beliefs, and the compensatory strategies that developed in response to automatic thoughts that are associated with specific situations, emotions, meanings, and behavior.
Persons’ Cognitive-Behavioral Formulation Persons and colleagues developed a cognitive case formulation approach emphasizing hypothesis testing (Persons, 1989, 2008; Persons and Tompkins, 2007). After assessment information is gathered, the therapist generates a comprehensive list of the client’s problems from which a multi-axial DSM diagnosis is assigned and an anchoring diagnosis is selected. The anchoring diagnosis is used to develop a nomothetic formulation which serves as a template of the psychological mechanisms hypothesized to be at work. The nomothetic formulation is derived from formulations that are implicit in empirically supported treatments or derives from cognitive and emotional theory. The nomothetic template is then individualized to account for client-specific details, including items on the problem list. The clinician hypothesizes mechanisms about how those problems are maintained, infers the origin of the mechanisms, and the precipitants that trigger the mechanisms
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
22
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
causing the problems. This information is then used to develop a comprehensive treatment plan.
Collaborative Cognitive Case Conceptualization Kuyken, Padesky and Dudley (2009) developed a distinct approach to cognitive case conceptualization, emphasizing collaborative empiricism and building on the client’s strengths. Collaborative empiricism involves “integrating the client’s experience with appropriate theory and research in an unfolding process of generating and testing hypotheses” (p. 27). Emphasizing client’s strengths incorporates resilience into a treatment plan, thus enhancing chances of a lasting recovery. These authors describe three levels of conceptualization: Descriptive, cross-sectional, and explanatory. The descriptive level involves eliciting and characterizing the client’s presenting issues in cognitive and behavioral terms and in the context of relevant cognitive-behavioral theory (CBT) and research. The goal is to connect the client’s experiences with the descriptive language of CBT theory. The cross-sectional level of conceptualization focuses on understanding the triggers and maintenance factors of a client’s problems. The primary task is to use cognitive and behavioral mechanisms to explain the situations in which the triggers arise and the factors operating to maintain the problems. The explanatory level of conceptualization seeks to understand predisposing and protective factors. Developmental history is used to understand and contextualize the current problems.
Haynes’ Functional Analytic Clinical Case Models Haynes’ Functional Analytic Clinical Case Models (FACCMs) approach is an elaborated functional analysis of behavior problems (Haynes, Leisen and Blaine, 1997; Haynes and Williams, 2003). The method produces an individualized behavioral treatment plan based on the clinician’s judgments about specific problems that have been identified. It considers the impact of situational factors, events that trigger and maintain problems, and the behavioral skills with which a client enters treatment. More specifically, the FACCM approach involves the clinician’s analysis of the relative importance, interrelationships and effects of behavior problems and goals. It includes inferences about causal mechanisms and their clinical utility, and an assessment of how causal mechanisms operate and are related to problem behaviors. The clinician also assesses moderating variables and estimates their impact. A diagram is then produced that depicts the problems, the inferred causes, mediating variables, and the interrelationships among them. The diagram includes numerical estimates of the impact of the problems, estimates of how modifiable the causal variables are, and estimates of the likelihood that interventions under consideration will have an impact. The FACCM guides the therapist in determining which problems and causal variables to target in treatment and which interventions may have the greatest effect.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
23
A Problem-Solving Perspective for CBT Case Formulation Nezu’s problem-solving model of case formulation (Nezu and Nezu, 1993; Nezu and Nezu, 1989; Nezu et al., 2007; Nezu, Nezu and Lombardo, 2004) is also based on functional analysis and is similar in other respects to Hayne’s FACCM approach. It is distinctive in its major emphasis on goal analysis and goal setting. In goal setting the clinician first identifies ultimate outcome goals (Nezu et al., 2007). These are the primary goals the therapy aims to achieve and reflect the reason for the therapy in the first place. Ultimate outcome goals may include relieving depression, improving a marital relationship or eliminating a phobia. They may be contrasted with instrumental outcomes, which are goals that serve as instruments for the attainment of the ultimate outcomes. For example, increasing self esteem may help relief depression. Instrumental outcomes may also serve as instruments for the attainment of other instrumental outcomes that eventually lead to an ultimate outcome. For example, improving coping skills can lead to increased self efficacy that, in turn, leads to reduced depression. The problem-solving approach to case formulation involves a systematic analysis that leads to the identification of ultimate outcomes, the instrumental outcomes that help the client achieve the ultimate outcome, and the relationships among them.
Emotion-Focused Therapy Case Formulation Emotion-focused therapy (EFT), developed by Leslie Greenberg (2002), has roots in the humanistic experiential tradition, and also in modern emotion theory and affective neuroscience (Greenberg and Goldman, 2007). Unlike the other case formulation methods described, it focuses on the moment-to-moment experiences unfolding in therapy and the attendant emotion, with a goal of strengthening the self. It does not involve developing a global case formulation of a client. “In EFT, formulations are never performed a priori (i.e., based on early assessment) as we do not attempt to establish what is dysfunctional or presume to know what will be most salient or important for the client” (Greenberg and Goldman, 2007, p. 380). The major means of formulation is “process diagnosis”, whereby the focus is on how people are currently experiencing their problem and whether they are doing so in an adaptive manner aimed at resolution, or not. Formulations are developed and redeveloped continually in a collaborative fashion with clients. The case formulation aspect of the therapy involves “identifying the client’s core pain and using that as a guide to the development of a focus on underlying determinants generating the presenting concerns” (Greenberg and Goldman, 2007, p. 384). Presenting problems are viewed as reflections of “underlying emotion-schematic processing difficulties” (p. 384). Put another way, the approach attends primarily to diagnosing clients’ manner of cognitive-affective processing rather than diagnosing clients per se. The therapist aims to identify markers of current emotional concerns and tasks to help resolve these concerns. Markers are client statements or behaviors that signify problems in need of attention as possible determinants of the presenting problem. These markers guide intervention, rather than an explicit case
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
24
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
formulation. Emotion-focused therapy, based on this approach to formulation, has been demonstrated to be efficacious for those with major depression (Goldman, Greenberg and Angus, 2006; Greenberg and Watson, 1998).
A GENERAL FRAMEWORK FOR FORMULATION Thus far we have defined formulation and made a case for its importance. We reviewed a range of theories and a variety of empirical sources that inform formulation. We then reviewed several structured case formulation models that can also be used as a basis for developing a case formulation. With all this theory, these sources of empirical evidence, and these formulation models available, how should one choose among them? The clinician committed to a single theoretical orientation could answer this question straightforwardly simply by disregarding the approaches that do not fit his or her orientation. We do not recommend this approach since we believe that each theory, source of information and formulation model has something to offer. For this reason, we recommend initial consideration of several models, recognizing that there is overlap among a number of them. Several of the structured models and theories, for example, share the concept of a cognitive schema that predicts behavioral tendencies. In addition, different theories tend to focus on different aspects of functioning. Behavioral models focus on symptom production and maintenance whereas psychodynamic models tend to emphasize personality organization, internal conflict, and the quality of interpersonal relationships. Cognitive models emphasize relatively accessible thought processes and how they shape behavior. These features need not be incompatible. In order to facilitate the choice among theories, models, and empirical sources, we recommend beginning with a general case formulation framework. Several are available in the literature (Eells, Kendjelic and Lucas, 1998; Meier, 2003; Mellsop and Banzato, 2006; Porzelius, 2002; Sperry et al., 1992). In addition, some methods designed for specific theoretical approaches are adaptable to a general model (e.g., Persons, 2008). All these methods view case formulation as lying between data gathering and formally providing treatment. Further, psychological problems are viewed within a diathesis-stress framework in which a mechanism is proposed that reflects a vulnerability on the part of the individual toward the development of problems and precipitants are proposed that trigger symptoms or episodes of distress. We propose the general model depicted in Figure 1.1, which due to space limitations we can only describe in a cursory fashion. As shown, the case formulation process is embedded in a general therapy model. Formulation itself occurs after information gathering and prior to providing treatment, although in actual practice one moves more fluidly among these stages. The model begins with gathering information because case formulation requires inputs. Information gathering can include a standard intake interview in which the clinician learns the presenting complaint, the history of the complaint, past history of mental health problems, medical history, the current living situation, developmental and social history, and related information (Morrison, 1993). In addition to gathering these biographical
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
25
Monitor, Test and Assess
Gather Information
Create Problem List
Formulate
Diagnose
Treat
Develop Explanatory Hypothesis
Terminate
Plan Treatment
Sources: Theory Evidence Components: Precipitants Origins Resources Obstacles
Figure 1.1 General case formulation mode.
details the clinician will likely want to know the client’s appraisals of the events described (Eells and Lombart, 2004). Other sources of information may be psychological testing results, symptom measures, medical records, and records from previous episodes of psychological treatment as well as information from family members. All this information serves as input to help develop the formulation. The general case formulation model itself has four major components. First is the identification of a set of problems to work on. These may or may not be the initial problems the client presents. Eliciting and collaboratively agreeing on the problems to focus on is a critical task since it is the problems themselves that are formulated. These goals should be specific, measurable, achievable, realistic, and timely. The second step is diagnosis. Despite controversies regarding the value of psychological diagnosis, we conclude that it is essential for a least three practical reasons. First, many treatment protocols are designed for individuals meeting specific diagnostic criteria. Knowledge of diagnosis, therefore, helps the clinician select treatment and, since treatment models contain implicit formulations and are linked to diagnoses, diagnosis can provide an initial lead on developing an explanatory hypothesis. Second, diagnosis facilitates communication among mental health professionals. If the client is obtaining concurrent services from others, such
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
26
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
as psychopharmacological treatment from a psychiatrist, providing that individual with diagnostic information can facilitate communication and consequently treatment. Third, diagnosis is often a practical necessity in order to bill and collect for one’s services. The third step in the general formulation model, developing the explanatory hypothesis, is the most crucial. It is the step in which the theoretical and evidentiary sources described earlier in this chapter come to bear on a specific individual Multiple explanatory hypotheses can often be proposed for a specific set of problems. There may not be a single correct explanation. Rather, the power of the explanatory hypothesis is evidenced primarily in its practical application. Nevertheless, we suggest that a high-quality explanatory hypothesis be adequately comprehensive in explaining the items on the problem list, be sufficiently elaborated and complex in linking together multiple facets of the individual’s functioning, be coherent in the sense of being internally consistent, be precise in the use of language, and be the product of systematic approach to formulation. Although the explanatory hypothesis could have multiple and varying components depending on the specific model one is following, we suggest that four are of primary importance. First, consider precipitants. These are events, stressors, experiences, or appraisals that trigger the onset of symptoms or the hypothesized mechanism that leads to symptoms. Second, provide an account of the origins of the proposed mechanism. This can include a hypothesized learning history that led to the individual’s vulnerability to the problems. Alternatively it can include traumas or empathic failures that hurt the person, genetic or other biological vulnerabilities, or contributing cultural factors. Third, consider the individual’s personal resources or strengths. These can be used to marshal hope, motivation, and leverage to recover. Examples of resources include unimpaired areas of functioning, premorbid functioning, intelligence, inferred level of psychosocial development, social support, capacity for pleasure, and sense of humor or irony. The final component we suggest for all formulations is a listing of obstacles that may impair a successful treatment outcome. These can be quite varied. Examples may include primitive or image distorting defense mechanisms, dichotomous thinking patterns, low capacity for the tolerance of ambiguity, poor social skills, financial problems, poor housing or living in a crime-ridden neighborhood, or lack of social support. The final step in the general case formulation model is that of treatment planning. It is also a critical step since it provides the link from the explanatory hypothesis to treatment implementation. Without a well thought out treatment plan, formulation is little more than an intellectual exercise. Regardless of its specific details, the treatment plan should flow directly and logically from the prior formulations steps and it should be sufficiently well elaborated and sequenced. One approach, as discussed earlier, is to begin with ultimate aims or goals for the treatment, then list process or instrumental goals that, if accomplished, should lead to the desired ultimate outcome. Once the formulation is developed, it should be testing in treatment and revised as necessary. Note the feedback loops in Figure 1.1. These depict the process of regular monitoring of outcome, or testing and revising the formulation, and of constantly assessing progress or the lack thereof. The final step, as shown, is termination.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
27
PRACTICAL TIPS FOR CASE FORMULATION We conclude this chapter with some practical tips to consider in formulation. First, write down the case formulation be written down and reference it prior to each session, at least the early sessions. In our experience, writing down the formulation facilitates a well thought out and comprehensive product. Often sketching a diagram rather than preparing a narrative helps to depict relationships among components of the formulation. Referencing the formulation prior to the session brings it back in memory and facilitates therapist consistency from session to session. Second, formulate a case using more than one theoretical approach or structured model. Viewing a client from multiple angles facilitates a flexible therapeutic approach and helps the therapist see the strengths and weakness of each formulation. Third, devise specific tests of your formulation. The best test of a formulation is how well it contributes to treatment outcome. Consider interventions that test the validity of a formulation and predict what response should be expected if the formulation is valid or if it is not. Fourth, be aware of biases in reasoning. Researchers have documented multiple judgment errors that individuals are prone toward (Ruscio, 2007). Clinicians are not immune to these errors and should be aware of them. Fifth, keep in mind that case formulation is a tool to help guide your treatment planning. It needs to work for you not the other way around. That is to say, one need not rigidly adhere to a formulation regardless of what transpires in therapy. Rather, consider the formulation as a map guiding empathic and effective interventions. It is a map that will change as the terrain of therapy changes. Finally, we recommend sharing the formulation with the client and getting feedback. Ideally, the formulation should be developed, tested, and revised collaboratively.
REFERENCES American Board of Psychiatry and Neurology (2009) Psychiatry and Neurology Core Competencies – Version 4.1. Retrieved February 20, 2010, from http://www.abpn.com/ downloads/core_comp_outlines/core_psych_neuro_v4.1.pdf American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders (4th edn). Washington, DC: American Psychiatric Press. APA Presidential Task Force on Evidence-Based Practice (2006) Evidence-based practice in psychology. American Psychologist, 61(4), 271–85. Beck, A.T. (1963) Thinking and depression: I. Idiosyncratic content and cognitive distortions. Archives of General Psychiatry, 9(4), 324–33. Beck, A.T., Ward, C., Mendelson, M. et al. (1961) An inventory for measuring depression. Archives of General Psychiatry, 4, 561–71. Beck, A.T., Epstein, N., Brown, G. and Steer, R.A. (1988) An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56(6), 893–7. Beck, J.S. (1995) Cognitive Therapy: Basics and Beyond. New York: Guilford Press. Behar, E. and Borkovec, T.D. (2006) The nature and treatment of generalized anxiety disorder. In B.O. Rothbaum (ed.), Pathological Anxiety: Emotional Processing in Etiology and Treatment (pp. 181–96). New York: Guilford Press. Bergner, R.M. (1998) Characteristics of optimal clinical case formulations: The linchpin concept. American Journal of Psychotherapy, 52, 287–300.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
28
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Bieling, P.J. and Kuyken, W. (2003) Is cognitive case formulation science or science fiction? Clinical Psychology: Science and Practice, 10, 52–69. Bohart, A.C. and Tallman, K. (1999) How Clients Make Therapy Work: The Process of Active Self-healing. Washington, DC, US: American Psychological Association. Bouton, M.E. (2002) Context, ambiguity, and unlearning: Sources of relapse after behavioral extinction. Biological Psychiatry, 52(10), 976–86. Breuer, J. and Freud, S. (1955) On the psychical mechanism of hysterical phenomena: Preliminary communication. In J. Strachey (ed.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 2, pp. 1–17). London: Hogarth Press. (Original work published in 1893.) Bruner, J.S. (1990) Acts of Meaning. Cambridge, MA, US: Harvard University Press. Bruner, J.S., Goodnow, J.J. and Austin, G.A. (1956) A Study of Thinking. Oxford: John Wiley & Sons, Ltd. Chadwick, P., Williams, C. and Mackenzie, J. (2003) Impact of case formulation in cognitive behaviour therapy for psychosis. Behaviour, Research and Therapy, 41(6), 671–80. Chomsky, N. (1959) A review of B.F. Skinner’s Verbal Behavior. Language, 35(1), 26–58. Crits-Christoph, P., Cooper, A. and Luborsky, L. (1988) The accuracy of therapists’ interpretations and the outcome of dynamic psychotherapy. Journal of Consulting and Clinical Psychology, 56(4), 490–5. Crits-Christoph, P., Gibbons, M.B.C., Temes, C.M. et al. (2010) Interpersonal accuracy of interventions and the outcome of cognitive and interpersonal therapies for depression. Journal of Consulting and Clinical Psychology, 78, 420–8. Curtis, J.T. and Silberschatz, G. (2005) The assessment of pathogenic beliefs. In G. Silberschatz (ed.), Transformative Relationships: The Control-Mastery Theory of Psychotherapy (pp. 69–92). New York, NY, US: Routledge. Curtis, J.T. and Silberschatz, G. (2007) Plan formulation method. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 198–220). New York: Guilford Press. DeNeve, K.M. and Cooper, H. (1998) The happy personality: A meta-analysis of 137 personality traits and subjective well-being. Psychological Bulletin, 124(2), 197–229. Derogatis, L.R. (1983) SCL-90-R Administration, Scoring, and Procedures Manual II (2nd edn). Towson, MD: Clinical Psychometric Research. Division of Clinical Psychology (2001) The Core Purpose and Philosophy of the Profession. Leicester, England: The British Psychological Society. Eells, T.D. (2007) Psychotherapy case formulation: History and current status. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 3–32). New York: Guilford Press. Eells, T.D., Kendjelic, E.M. and Lucas, C.P. (1998) What’s in a case formulation? Development and use of a content coding manual. Journal of Psychotherapy Practice and Research, 7(2), 144–53. Eells, T.D. and Lombart, K.G. (2004) Case formulation: Determining the focus in brief dynamic psychotherapy. In D. P. Charman (ed.), Core Processes in Brief Psychodynamic Psychotherapy (pp. 119–44). Mahwah, New Jersey: Lawrence Erlbaum Associates. Ellis, A. (1994) Reason and Emotion in Psychotherapy (revised and updated). Secaucus, NJ: Birch Lane. Ellis, A. (2000) Rational emotive behavior therapy (REBT). In Encyclopedia of Psychology, Vol. 7 (pp. 7–9): Washington, DC, US: American Psychological Association; New York, NY, US: Oxford University Press. Ferster, C.B. (1973) A functional analysis of depression. American Psychologist, 28, 857–70. Festinger, L. (1957) A Theory of Cognitive Dissonance. Stanford: Stanford University Press. First, M.B., Spitzer, R.L., Gibbon, M. and Williams, J.B.W. (1995) The Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II): I. Description. Journal of Personality Disorders, 9(2), 83–91. Frankfurter, D. (2006) Evil Incarnate: Rumors of Demonic Conspiracy and Satanic Abuse in History. Princeton, NJ, US: Princeton University Press. Freeman, D., Bentall, R. and Garety, P. (eds) (2008) Persecutory Delusions: Assessment, Theoryand Treatment. Oxford: Oxford University Press.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
29
Freud, S. (1958a) The dynamics of the transference (J. Strachey, trans.). In J. Strachey (ed.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 12, pp. 99–108). London: Hogarth Press. (Original work published 1912.) Freud, S. (1958b) Recommendations to physicians practising psycho-analysis (J. Strachey, trans.). In J. Strachey (ed.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 12, pp. 111–20). London: Hogarth Press. (Original work published 1912.) Galatzer-Levy, R.M. (2003) Self psychology. In A. Tasman, J. Kay and J.A. Lieberman (eds), Psychiatry (2nd edn, Vol. 1, pp. 478–82). Sussex: John Wiley & Sons, Ltd. Garb, H.N. (2003) Incremental validity and the assessment of psychopathology in adults. Psychological Assessment, 15(4), 508–20. Goldman, R.N., Greenberg, L.S. and Angus, L. (2006) The effects of adding emotion-focused interventions to the client-centered relationship conditions in the treatment of depression. Psychotherapy Research, 16(5), 536–46. Gordis, L. (1990) Epidemiology. Philadelphia, PA: W.B. Saunders Company. Greenberg, L.S. (2002) Emotion-Focused Therapy: Coaching Clients to Work through their Feelings. Washington, DC, US: American Psychological Association. Greenberg, L.S. and Goldman, R. (2007) Case formulation in emotion-focused therapy. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 379–411). Guilford Press: New York. Greenberg, L.S. and Watson, J. (1998) Experiential therapy of depression: Differential effects of client-centered relationship conditions and process experiential interventions. Psychotherapy Research, 8(2), 210–24. Halstead, J.E., Leach, C. and Rust, J. (2008) The development of a brief distress measure for the evaluation of psychotherapy and counseling (sPaCE). Psychotherapy Research, 17(6), 656–72. Hayes, S.C., Nelson, R.O. and Jarrett, R.B. (1987) The treatment utility of assessment: A functional approach to evaluating assessment quality. American Psychologist, 42, 963–74. Hayes, S.C. and Strosahl, K.D. (eds) (2004) A Practical Guide to Acceptance and Commitment Therapy. New York: Springer. Haynes, S.N., Leisen, M.B. and Blaine, D.D. (1997) Design of individualized behavioral treatment programs using functional analytic clinical case models. Psychological Assessment, 9(4), 334–8. Haynes, S.N. and Williams, A.E. (2003) Case formulation and design of behavioral treatment programs: Matching treatment mechanisms to causal variables for behavior problems. European Journal of Psychological Assessment, 19(3), 164–74. Horowitz, M.J. (1991a) Person schemas. In M.J. Horowitz (ed.), Person Schemas and Maladaptive Interpersonal Patterns (pp. 13–31): Chicago, IL, US: University of Chicago Press. Horowitz, M.J. (ed.) (1991b) Person Schemas and Maladaptive Interpersonal Patterns. Chicago, IL: University of Chicago Press. Horowitz, M.J. (2005) Understanding Psychotherapy Change: A Practical Guide to Configurational Analysis. Washington, DC, US: American Psychological Association. Kazdin, A.E. (2007) Mediators and mechanisms of change in psychotherapy research. In Annual Review of Clinical Psychology, 3, 1–27. Kazdin, A.E. (2008) Evidence-based treatment and practice: New opportunities to bridge clinical research and practice, enhance the knowledge base, and improve patient care. American Psychologist, 63(3), 146–59. Kelly, G.A. (1955) The Psychology of Personal Constructs. Vol. 1. A Theory of Personality. Vol. 2. Clinical Diagnosis and Psychotherapy. Oxford: Norton. Kessler, R.C. and Wang, P.S. (2009) Epidemiology of depression. In Handbook of Depression (2nd edn, pp. 5–22). New York, NY, US: Guilford Press. Koerner, K. (2007) Case formulation in dialectical behavior therapy for borderline personality disorder. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn). New York: Guilford Press. Kohut, H. (1971) Analysis of the Self. New York: International Universities Press. Kohut, H. (1977) Restoration of the Self. New York: International Universities Press.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
30
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Kuehlwein, K.T. and Rosen, H. (1993) Cognitive Therapies in Action: Evolving Innovative Practice. San Francisco, CA, US: Jossey-Bass. Kuyken, W. (2006) Evidence-based case formulation: Is the emperor clothed? In N. Tarrier (ed.), Case Formulation in Cognitive Behaviour Therapy? The Treatment of Challenging and Complex Cases (pp. 12–35). New York, NY: Routledge/Taylor & Francis Group. Kuyken, W., Padesky, C.A. and Dudley, R. (2009) Collaborative Case Conceptualization: Working Effectively with Clients in Cognitive-Behavioral Therapy. New York, NY: Guilford Press. Lambert, M.J. and Finch, A.E. (1999) The Outcome Questionnaire. In M.E. Maruish (ed.) The Use of Psychological Testing for Treatment Planning and Outcomes Assessment (2nd edn) (pp. 831–69). Mahwah, NJ: Lawrence Erlbaum Associates. Linehan, M.M. (1993) Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York: Guilford. Luborsky, L. (1977) Measuring a pervasive psychic structure in psychotherapy: The core conflictual relationship theme. In N. Freedman and S. Grand (eds), Communicative Structures and Psychic Structures (pp. 367–95). New York: Plenum Press. Luborsky, L. and Barrett, M.S. (2007) The core conflictual relationship theme: A basic case formulation method. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 105–35). New York: Guilford Press. Luborsky, L., Crits-Christoph, P., Friedman, S.H. et al. (1991) Freud’s transference template compared with the Core Conflictual Relationship Theme (CCRT). In M.J. Horowitz (ed.), Person Schemas and Maladaptive Interpersonal Patterns (pp. 167–95). Chicago: University of Chicago Press. Mahoney, M.J. (1991) Human Change Processes: The Scientific Foundations of Psychotherapy. New York, NY, US: Basic Books. Maslow, A.H. (1987) Motivation and Personality (3rd edn). New York, NY: Harper & Row. Meier, S.T. (2003) Bridging Case Conceptualization, Assessment, and Intervention. Thousand Oaks, CA: Sage. Mellsop, G.W. and Banzato, C.E.M. (2006) A concise conceptualization of formulation. Academic Psychiatry, 30, 424–5. Messer, S.B. and Wolitzky, D.L. (2007) The traditional psychoanalytic aproach to case formulation. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 67–104). New York, NY: Guilford Press. Mineka, S. and Zinbarg, R. (2006) A contemporary learning theory perspective on the etiology of anxiety disorders: It’s not what you thought it was. American Psychologist, 61(1), 10–26. Morrison, J. (1993) The First Interview. New York: Guilford Press. Mowrer, O.H. (1960) Learning Theory and Behaviour. Hoboken, NJ, US: John Wiley & Sons Inc. Newell, A., Shaw, J.C. and Simon, H.A. (1958) Elements of a theory of human problem solving. Psychological Review, 65(3), 151–66. Nezu, A.M. and Nezu, C.M. (eds) (1989) Clinical Decision Making in Behavior Therapy: A Problem-Solving Perspective. Champaign, IL: Research Press. Nezu, A.M. and Nezu, C.M. (1993) Identifying and selecting target problems for clinical interventions: A problem-solving model. Psychological Assessment, 5(3), 254–63. Nezu, A.M., Nezu, C.M. and Cos, T.A. (2007) Case formulation for the behavioral and cognitive therapies. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 349–78). New York: Guilford. Nezu, A.M., Nezu, C.M. and Lombardo, E.R. (2004) Cognitive-Behavioral Case Formulation and Treatment Design: A Problem-Solving Approach. New York: Springer Publishing Co. Nock, M.K. and Kessler, R.C. (2006) Prevalence of and risk factors for suicide attempts versus suicide gestures: Analysis of the National Comorbidity Survey. Journal of Abnormal Psychology, 115(3), 616–23. Nolen-Hoeksema, S., Wisco, B.E. and Lyubomirsky, S. (2008) Rethinking rumination. Perspectives on Psychological Science, 3(5), 400–24. Perls, F., Hefferline, R.F. and Goodman, P. (1965) Gestalt Therapy. Oxford, England: Dell.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
May 28, 2011
8:16
Printer Name: Yet to Come
THEORETICAL AND EVIDENCE-BASED APPROACHES TO CASE FORMULATION
31
Persons, J.B. (1989) Cognitive Therapy in Practice: A Case Formulation Approach. New York: W.W. Norton. Persons, J.B. (2008) The Case Formulation Approach to Cognitive-Behavior Therapy. New York, NY: Guilford. Persons, J.B. and Tompkins, M.A. (2007) Cognitive-behavioral case formulation. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 290–316). New York: Guilford. Porzelius, L.K. (2002) Overview. In M. Hersen and L.K. Porzelius (eds), Diagnosis, Conceptualization, and Treatment Planning for Adults: A Step-by-Step Guide (pp. 3–12). Mahwah, NJ, US: Lawrence Erlbaum. Postman, L. (1951) Toward a general theory of cognition. In J. Roherer and M. Sherif (eds), Social Psychology at the Crossroads: The University of Oklahoma Lectures in Social Psychology (pp. 242–72). Oxford: Harper. Randall, C.L., Book, S.W., Carrigan, M.H. and Thomas, S.E. (2008) Treatment of co-occurring alcoholism and social anxiety disorder. In S.H. Stewart and P. Conrod (eds), Anxiety and Substance Use Disorders: The Vicious Cycle of Comorbidity (pp. 139–55): New York, NY, US: Springer Science + Business Media. Rapaport, D. and Gill, M.M. (1959) The points of view and assumptions of metapsychology. International Journal of Psychoanalysis, 40, 153–61. Rogers, C.R. (1951) Client-Centered Therapy, Its Current Practice, Implications, and Theory. Boston: Houghton Mifflin. Ruscio, J. (2007) The clinician as subject: Practitioners are prone to the same judgment errors as everyone else. In S.O. Lilienfeld and W.T. O’Donohue (eds), The Great Ideas of Clinical Science: Seventeen Principles that Every Mental Health Professional Should Understand (pp. 29-47). New York: Routledge/Taylor & Francis. Ryle, A. (1990) Cognitive Analytic Therapy: Active Participation in Change. Chichester: John Wiley & Sons, Ltd. Seitz, P F. (1966) The consensus problem in psychoanalytic research. In L. Gottschalk and L. Auerbach (eds), Methods of Research and Psychotherapy (pp. 209–225). New York: Appleton, Century, Crofts. Silberschatz, G. (2005a) The control-master theory. In G. Silberschatz (ed.), Transformative Relationships: The Control-Mastery Theory of Psychotherapy (pp. 3–24). New York, NY, US: Routledge. Silberschatz, G. (2005b) An overview of research on control-mastery theory. In G. Silberschatz (ed.), Transformative Relationships: The Control-Mastery Theory of Psychotherapy (pp. 189–218). New York, NY, US: Routledge. Simon, H.A. and Newell, A. (1958) Heuristic problem solving: The next advance in operations research. Operations Research, 6, 1–10. Skinner, B.F. (1953) Science and Human Behavior. New York, NY: The Free Press. Sperry, L., Gudeman, J.E., Blackwell, B. and Faulkner, L.R. (1992) Psychiatric Case Formulations. Washington, DC: American Psychiatric Press. Spitzer, R.L., Williams, J.B., Gibbon, M. and First, M.B. (1992) The Structured Clinical Interview for DSM-III-R (SCID): I. History, rationale, and description. Archives of General Psychiatry, 49(8), 624–9. Sturmey, P. (2008) Behavioral Case Formulation and Intervention: A Functional Analytic Approach. Chichester: Wiley-Blackwell. Tarrier, N. and Calam, R. (2002) New developments in cognitive-behavioural case formulation. Epidemiological, systemic and social context: An integrative approach. Behavioural and Cognitive Psychotherapy, 30(3), 311–28. Vaillant, G.E. (1995) The Natural History of Alcoholism Revisited. Cambridge, MA, US: Harvard University Press. Wachtel, P.L. (1977) Psychoanalysis and Behavior Therapy. New York: Basic Books. Weiss, J. (1993) How Psychotherapy Works: Process and Technique. New York: Guilford. Weiss, J. and Sampson, H. (1986) The Psychoanalytic Process: Theory, Clinical Observation, and Empirical Research. New York: Guilford.
P1: OTA/XYZ P2: ABC JWST073-01 JWST073-McMurran
32
May 28, 2011
8:16
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Wells, K.B., Burnam, M.A., Rogers, W. et al. (1992) The course of depression in adult outpatients: Results from the Medical Outcomes Study. Archives of General Psychiatry, 49(10), 788–94. Westen, D., Novotny, C.M. and Thompson-Brenner, H. (2004) The empirical status of empirically supported psychotherapies: Assumptions, findings, and reporting in controlled clinical trials. Psychological Bulletin, 130(4), 631–63. Wilder, D.A. (2009) A behavior analytic formulation of a case of psychosis. In P. Sturmey (ed.), Clinical Case Formulation: Varieties of Approaches (pp. 107–18). Chichester, UK: WileyBlackwell. Young, J.E. (1990) Cognitive Therapy for Personality Disorders: A Schema-Focused Approach. Sarasota, Florida: Professional Resource Exchange, Inc. Young, J.E., Klosko, J.S. and Weishaar, M.E. (2003) Schema Therapy: A Practitioner’s Guide. New York, NY, US: Guilford.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
Chapter 2
CURRENT ISSUES IN CASE FORMULATION GREGORY H. MUMMA Texas Tech University, USA
CHAPTER GOALS The first three sections of this chapter describe psychometric issues relevant to the scientific status of clinical case formulation (CF). These sections focus on content validity, reliability, and additional validity issues in CF. There is a particularly strong emphasis on validity issues due to their relative neglect in the CF literature. The next section briefly covers certain treatment validity or utility issues that supplement the main treatment validity focus of Chapter 3 by Ata Ghaderi. These issues include the effects of sharing the CF with the patient on the therapeutic relationship and treatment outcome. A brief final section of this chapter focuses on aspects of clinical training and expert-novice differences in CF. Due to the breadth of coverage combined with length limitations, this chapter does not attempt to comprehensively review the literature in these areas, but instead will highlight some of the major theoretical, methodological, and clinical-practice issues in each of these areas and, when helpful, discuss either representative or particularly noteworthy theoretical, methodological, or empirical contributions. A few terminology issues: I will use the terms “construct” or “variable” to describe behaviors or constellations of behaviors grouped together because of theoretical or clinical relevance (e.g., depressive symptoms). A “target” or “outcome” variable, construct, or behavior refers to symptoms, dysfunction, or behaviors that are characteristic of a disorder or problem, and which may be targeted for intervention. “Causal variable” refers to both triggering and maintaining variables, as well as other variables that may influence the topography of the target behavior (Haynes, 1992; Haynes and O’Brien, 2000). Such variables may be considered “functional causal” variables if they have a functional relationship to a target variable. Briefly, this involves statistical and temporal relationships (e.g., the parent’s attention follows the child’s aggression which it functions to positively reinforce). See Haynes and O’Brien (2000) for a more detailed discussion of these issues. Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
34
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
This chapter focuses on CFs from a number of different theoretical perspectives including behavioral, cognitive, and psychodynamic. Empirical research within each of these theoretical frameworks has made important contributions that may have important scientific implications for CF within other frameworks and for developing areas such as forensic case formulation.
SCIENTIFIC BASES I: CF COMPONENTS AND CONTENT VALIDITY Construct Explication and Content Validity in CF The components of a CF are relevant to the issue of content validity as well as construct explication. These issues are important because reliability and other types of validity issues are tied to this. Generally, construct explication is a largely theorydriven delineation of the facets and elements of a construct as well as its relationship to other constructs (Nunnally and Bernstein, 1994; Shadish, Cook and Campbell, 2002). For CF, construct explication involves a theoretically driven delineation of both the content and structure of the CF within that framework. This may include the components that should comprise the CF, including both outcome variables, such as depression, or functioning in school or work, and causal variables, such a environmental or cognitive triggers. Additional CF components are the relationships between the causal and outcome variables, including functional, correlational, or temporal relationships. Construct explication provides general guidelines on a theoretical level – it provides a structure for the development of a CF. This is often done as a manual. For example, Persons (1989, 2008) described the components that should be included in a cognitive-behavioral CF: Problem list, developmental origins, precipitants or triggers, and mechanisms causally linking these. These “guidelines” may be general, as in the components of a cognitive-behavioral CF generally, or may be specific to certain disorders or types of problems, such as a CF template for generalized anxiety disorder or post-traumatic stress disorder (Nezu, Nezu and Lombardo, 2004; Needleman, 1999; Wells, 2006). Of course, on either level, different investigators within the same theoretical framework may explicate different structures, components, or relationships between them for a CF. For example, for cognitive-behavioral case formulation we can contrast the approaches of J.S. Beck (1995) versus Persons. Likewise, for behavioral CF we can contrast the Clinical Pathogenesis Map of Nezu and colleagues versus the Functional Analytic Clinical Case Model of Haynes and O’Brien (1990, 2000).
Content Validity Content validity of a CF refers to the extent to which the relevant target and causal variables, their components (facets and elements), and their relationships are represented or included in the CF for a particular person (cf. Haynes and O’Brien, 2000). Thus, the content validity of a CF will vary between clinicians and cases. That is,
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
35
even when using the same approach on the same case, the content validity of one CF may be quite strong whereas another may be weak, as when two clinicians independently develop a CF of the same person. Content validity includes two main components – relevance and coverage (or representativeness), which together determine the fit of the hypothesized CF to the data for a particular person. Content validity involves the empirical evaluation of these two components. The content validity of a CF is constrained by construct explication of the CF variables and their facets as well as description of the procedures in the CF manual. However, content validity is empirically evaluated for specific cases, just as the content validity of an educational test is evaluated by comparing it to the content or subject matter it purports to measure. Analogous to behavioral observation, in which a manual may carefully define and exemplify the coding criteria for various target behaviors (e.g., “aggression” involves physical contact with apparent intent to injure or cause pain or distress), content validity in CF refers to whether the CF accurately captures the frequency and severity of the most relevant target behaviors for that particular individual. This would include the particular facets of aggression relevant for that person in contexts that adequately represent those important in his or her life. In forensic CF, relevant target behaviors may be defined legally, or by the patient or client, such as parent or teacher for a child’s aggressive behavior. The content validity of a CF may also include other potentially relevant target behaviors, such as the child’s anger and depression, as well as constructs/behaviors that may function as triggering or maintaining variables. Some of these components may be suggested in a CF template for that problem or disorder, based on aggregate-level psychopathology research. However, the content validity of a CF covers the problems or issues and causal variables of greatest relevance for that particular person. For example, the likelihood of partner aggression for a particular person may be related to highly specific and idiosyncratic external situational triggers (e.g., the partner says something in an angry tone) combined with recent stressors (e.g., the person was “chewed out” by his/her boss) and internal states (angry and depressed, has been drinking, interprets the partner’s communication as insulting). It is important to note that results of aggregate-level research may find that some of these variables are only weakly related to the target. For example, having been “chewed out” by the boss may be relevant in only 5% of cases, but may be strongly related for that particular person. Thus, content validity of a CF needs to be evaluated, at least in part, on a case-by-case level. The main issues are the relevance of the variable/construct for that person and the representativeness or coverage of the problems/issues and relevant causal variables (Haynes and O’Brien, 2000; O’Brien, Oemig and Northern, 2010). But what are the specific criteria for relevance and coverage? Given the absence of a gold standard, issues involving convergence of ratings between, say, the patient and the clinician may play a central role (Kuyken, 2006).
Content Validity in Psychodynamic CF This often involves evaluating recurring maladaptive relationship patterns such as the Core Conflictual Relationship Theme (CCRT: Luborsky and Crits-Christoph,
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
36
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
1998) or Configurational Analysis (Horowitz and Eells, 2007). The CCRT attempts to capture recurring relationship patterns consisting of a wish, response of other, and response of self as reported by the patient for particular interactions with others. Configurational analysis involves developing a set of four role relationship models involving dreaded, desired, adaptive, and problematic compromises. Role relationship models are diagrams that organize relationship schema and core interactional patterns between the patient and others (Horowitz and Eells, 2007). Content validity issues include whether the CCRT or Configurational Analysis captures patterns or issues occurring in the most important relationships in a person’s life, such as a spouse or partner, in domains where dysfunction is particularly problematic for that person, such as relationships with a supervisor or possibly coworkers, or a relationship with an adolescent child, and in the therapeutic relationship – the transference (Fried, Crits-Christoph and Luborsky, 1998). To evaluate content validity, clinical judges rate the relevance of items from the target patient’s formulation and compare these to ratings of items developed for other cases, which is referred to as the “method of mismatched cases” (Curtis et al., 1994; Levine and Luborsky, 1981). A potential content validity issue in psychodynamic formulations such as the CCRT or Configurational Analysis is the extent to which the CF includes problems or issue relevant to the patient’s specific problems/dysfunction or general life functioning. Whereas depression, worry, anger, or aggression would likely be specifically and clearly targeted as a problem in a cognitive-behavioral or behavioral CF, this is not necessarily the case with certain types of psychodynamic formulations that focus primarily on core maladaptive relationship patterns.
Content Validity in Cognitive-Behavioral and Behavioral CFs For cognitive-behavioral and behavioral CFs, content validity is important for the causal variables as well as the target behaviors (Haynes and O’Brien, 2000). Omission of potentially important causal variables means that important functional or causal relationships may be missed in the CF. A common approach to increase the content validity of the causal variables is to start the CF with a template formed through aggregate-level research for the relevant diagnosis: the “nomothetic formulation of the anchoring diagnosis” (Persons and Tompkins, 2007, p. 301) or for that type of behavioral problem (Haynes, Kaholokula and Nelson, 1999; Haynes and O’Brien, 2000). Unfortunately, evaluations of the content validity of cognitivebehavioral CFs are relatively scant and have generally occurred when evaluating reliability, discriminant validity, the effects of training, or expert-novice differences (see below). One approach used in cognitive-behavioral CF and cognitiveinterpersonal approaches, similar to the method of mismatched cases (Levine and Luborsky, 1981), involves rating the relevance of cognitive items, such as thoughts and beliefs, taken from clinical material for that person versus items taken from other cases (Mumma, 2004; Mumma and Mooney, 2007a; Muran, Samstag and Segal, 1998; Muran et al., 2001; Muran and Segal, 1992).
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
37
Summary and Comments Content validity, which includes coverage and relevance, is a matter of degree and needs to be evaluated on a case-by-case basis. The CF’s focus on the individual permits potentially high content validity because the assessment can be tailored to the unique situation and idiosyncratic life circumstances of a particular person. The incorporation of idiographic assessment can result in a fine-grained, highly relevant CF that covers the problems/issues and causal variables of greatest importance to that individual. To evaluate content validity, the method of mismatched cases – comparing formulation components for the target patient to other cases – has been used across a variety of theoretical frameworks. It is the methodology of choice at this time for research and clinical training or practice.
SCIENTIFIC BASES II: RELIABILITY A second major scientific challenge for CF is the extent to which the CF or its components can be reliably identified for a specific case or cases. The type of reliability that has been considered most important and been evaluated most frequently is interrater, or more aptly, interformulator reliability. However, several additional forms of reliability are potentially relevant and will be discussed briefly.
Interrater Reliability A CF is developed or generated by the clinician, so the term “interformulator” reliability is probably more apt than “interrater” because there is far more involved in CF than making ratings. Interrater reliability is critically important in behavioral observation and coding and in behavioral ratings (e.g., clinician ratings). Generally, two or more raters/clinicians view the same material, which is often videotaped. For behavioral observation, observers are frequently trained to criterion prior to making observations. Reliability may be increased, and observer drift decreased, by increased operationalization of criteria and using examples, but having too many codes can decrease reliability. Interrater reliability for behavioral observation is readily calculated using percentage agreement between the coders, or a chance-corrected agreement statistic such as Kappa. Interrater reliability for numeric ratings is often calculated using the intraclass correlation coefficient, which can be used to estimate the reliability of a single, “average” judge or any number of judges.
Interformulator Reliability This is somewhat more complicated because it involves CF components, such as target behaviors, relevant causal variables and their interrelationships, that are generated by each clinician. These components must then be somehow coded or rated to study interformulator reliability.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
38
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Cognitive-Behavioral CFs Two studies have been completed by Persons and one by Kuyken and colleagues which address this topic. Persons, Mooney and Padesky (1995) studied 46 clinicians who had just completed the morning portion of a workshop on cognitivebehavioral CF using Persons’ (1989) approach. Clinicians were able to reasonably well identify most problems and issues on two cases (67% and 100% agreement). However, the reliability of identification of underlying mechanisms even when presented in a multiple-choice questionnaire was substantially lower (mean intraclass correlation coefficient = .46, range = .07 to .70). In a replication study, Persons and Bertagnolli (1999) found similar results with reliability of identification of overt problems “excellent” (>90%) for four problems, and “moderate” (>67%) for three more problems for two cases (p. 26). However, reliability of ratings of strength of belief in underlying schemas presented on a list was more problematic for an average clinician (mean intraclass correlation = .37, range = .13 to .66). Specifying the context for the ratings (e.g., “at work” or “with boyfriend”) did not improve reliability. Two points are worth noting. First, the reliability of problem identification was calculated only for those problems defined by the expert clinicians and would probably be lower if a list of all problems generated by the trainee clinicians had been rated. Second, interrater reliability for ratings of core beliefs or schema were made from a predetermined list of beliefs in the self, other, and world domains rather than beliefs generated by the trainee clinicians, a procedure that may over-estimate reliability in routine clinical practice (Aston, 2009). Kuyken and colleagues (2005) increased structure in the cognitive-behavioral CF procedure by having clinicians use the Case Conceptualization Diagram developed by J. Beck (1995). Adequate reliability (>60% agreement) was found for just two of the seven items of childhood data (e.g., “father critical”) identified by J. Beck in her benchmark formulation, on two of three core beliefs (e.g., “I’m incompetent”), and on one of four compensatory strategies (e.g., “avoidance”, “self-harm”). This study assessed interformulator reliability because the clinicians’ responses about childhood data, core beliefs, etc. were open-ended and coded from what was written on the Case Conceptualization Diagram. This response format likely provides a more challenging and realistic test of reliability than providing clinicians with a list of beliefs or mechanisms, as was done in the two Persons’ studies.
Comments and Issues for Cognitive-Behavioral CFs There is reasonably strong evidence that, after a rather brief period of training, clinicians can learn to identify most of the overt problems for a case with reasonable interrater reliability. These problems serve as a starting point for a CF. However, results of brief training sessions relevant to the variables that are triggering and maintaining the target behaviors are less encouraging. There are two points worth considering here. First, in contrast to more behavioral or functional analytic approaches to CF, Persons’ cognitive-behavioral CF approach includes few methods for testing the formulation once developed. This limits the feedback available to the clinician. Second, the amount of training provided in these workshops is minimal compared to the extent of practicum training received by students in doctoral
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
39
scientist-practitioner training programs. Cognitive-behavioral CF is a complex skill that probably can not be learned very well in half a day or a weekend!
Psychodynamic CFs Interformulator reliability evaluation for two types of psychodynamic CFs, the Plan Formulation Method and the CCRT, are summarized next. In the Plan Formulation Method, based on material from clinical interviews, the clinician formulates the goals, obstructions, tests, and insights expected from the patient during therapy (Curtis et al., 1994). Interformulator reliabilities for each of these components ranged from .47 to .72 for a single or “average” experienced clinician (psychologists with >5 years of CF experience) but were somewhat lower (.41 to .50) for graduate students. For the CCRT, a series of reliability studies using data from relationship episodes described during therapy, have found agreement (weighted kappa) on the wish, negative response of other, and negative response of self components between. 61 to .81 (Crits-Christoph, Cooper and Luborsky, 1988; Luborsky and Diguer, 1998; Barber et al., 1995; Lefebvre et al., 1996; Popp et al., 1996). Kappas in this range indicate good to excellent (>.75) chance-corrected reliability (Fleiss, 1981). Most interrater reliability studies for the CCRT use the standard category clusters, as opposed to the standard categories, a methodology that may increase the reliability due to aggregation. Studies investigating the reliability of the CCRT components generated idiographically require an additional set of raters who compare the tailored CCRT component responses across formulators. This methodology seems to be used relatively infrequently. An example is a preliminary study of a single patient using idiographic CCRT components (Levine and Luborsky, 1981) that found poor reliabilities for two of the three CCRT components.1
Comments and Issues The relatively large number of studies of interrater reliability in psychodynamic CFs is laudable, yet a number of limitations to the CCRT reliability studies should be noted. First, in these research studies relationship episode transcripts were selected from psychotherapy transcripts prior to development of the CF. This probably increases reliability compared to a clinical context in which clinicians may differ in the material selected for formulating the CCRT. Indeed, Zander and colleagues (1995a, b) found significantly higher coder agreement for preselected relationship episodes versus not. Second, calculating the interrater agreement for each of the three components of the CCRT (wish, response of other, response of self) is procedurally necessary, but probably overestimates the reliability of the entire CF which is a conjunction of the reliabilities of each part. Finally, as mentioned above, 1 For example, averaging across two agreement judges, student judges agreed that theme elements were
from the same case 12 times versus from a different case 9.5 times for the Wish component of the CCRT, but were 9.5 (same) versus 20.5 (other) for the Response of Other and 12 (same) versus 16 (other) for the Response of Self.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
40
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
for some psychodynamic approaches, including the CCRT, a clear identification of the problems or symptoms is not necessarily made in the CF.
Internal Consistency Reliability This basic type of reliability is typically evaluated for multi-item measures. Scores on items intended to measure the same construct should be intercorrelated because they all purport to measure the same construct. A reasonably high, average interitem correlation combined with a sufficient number of items will yield a satisfactory coefficient alpha – a widely used measure of the internal consistency reliability of the entire scale. Multiple items or operationalizations are necessary to evaluate this type of reliability, which is closely related to convergent validation on the item level (Burns and Haynes, 2006). Interestingly, this type of reliability is rarely investigated in CF, apparently because many investigators appear satisfied with single-item operationalization of the target constructs or behaviors in the CF. However, there are several exceptions. For example, Curtis et al. (1988; 1994) provided the relevance ratings of items from each of the four components of the Plan Formulation (goals, obstructions, insights, tests) but they did not compute the internal consistency reliability for items developed for the target patient. For the CCRT, internal consistency is inherent to formulating central relationship themes. That is, to infer a core conflictual relationship theme, it must be present over a relatively large number of relationship episodes for that patient. In fact, pervasiveness, the percent of relationship episodes evidencing the particular wish, response-of-other, or response-of-self component of the CCRT (Crits-Christoph and Luborsky, 1998; Wilczek et al., 2004), is basically a measure of internal consistency reliability. Although cognitive-behavioral CFs may include multiple examples of thoughts and beliefs for a schema (e.g., Beck, 1995), they have rarely been evaluated for internal consistency or item-level convergent validity. However, methods to do so are available when data is collected repeatedly over time. For example, using verbatim statements from the patient, Mumma (2004) and Mumma and Mooney (2007a,b) developed multiple items tapping into each cognitive schema, stressful event, and distress construct in the CF for that person. The internal consistency of each construct was evaluated with a form of confirmatory factor analysis appropriate for daily ratings made by the patient on each item.
Test-Retest Reliability and Stability Test-retest reliability generally involves re-administration of a test after a relatively brief (e.g., two-week) interval and comparing the scores from the two administrations. A parallel form of the test may be used to reduce recall effects. An appropriate test-retest interval is related to the presumed temporal stability of the targeted construct (e.g., trait versus state anxiety or anger; Spielberger et al., 1999). For a CF, test-retest reliability would be evaluated by re-collecting the relevant clinical data following a relatively brief period and reformulating the case. To properly control for recall effects by the clinician, the reformulation should be done by a different
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
41
clinician than the first. Thus, interformulator reliability places a ceiling on testretest reliability for CFs. Additionally, reactivity – whereby the patient reacts to, and may change from the collection of data for the first CF – is a potential issue, one probably difficult to control. Finally, the interval between evaluations is important – estimation of test-retest reliability assumes the underlying trait, construct, or process is stable from time 1 to time 2. The longer the interval the more likely the retest evaluates both reliability and stability. Evaluation of test-retest reliability of CFs is rarely done. Despite the paucity of CF test-retest evaluations, several studies have evaluated the stability of psychodynamic CFs, such as the CCRT (Barber et al., 1995; Luborsky et al., 1998). For example, Luborsky and colleagues (1998) evaluated the stability of each of the three components of the CCRT (wish, response other, response of self)when rated from relationship episodes prior to therapy versus sessions 3 and 5. Agreement ranged from .77 to 1.00.
SCIENTIFIC BASES III: VALIDITY OF CF AND ITS COMPONENTS General Issues Pertaining to Validity of CFs Few scientists would disagree that validation is an essential process for developing scientifically useful measures of a construct or variable. Furthermore, few clinicians would use a measure with a patient if there was no reasonable evidence for its validity. Yet, with several exceptions, validity is an issue that has received relatively little attention in the case formulation literature, even within theoretical frameworks, such as cognitive behavioral, that have typically presented a strong empirical basis for treatment efficacy (Bieling and Kuyken, 2003; Kuyken, 2006; Kuyken, Padesky and Dudley, 2009). There are several types or forms of validity and each has a meaning or application to CF that may differ somewhat from its conceptualization, evaluation, or use with traditional tests or measures. As a general principle for this discussion consider that “Technically speaking, it is not the measurement tool itself that is valid, but rather it is the inferences (or interpretations) that we make from that tool that require evidence of validity” (Michael and Li, 2010, p. 350). Although the meaning of this statement is relatively clear for interpreting a person’s score on a standardized measure, say, of intelligence or depression, in terms of criterion or construct validity (Furr and Bacharach, 2008; Nunnally and Bernstein, 1994; Shadish, Cook and Campbell, 2002), its meaning when applied to CF seems less clear. Next, after discussing several general issues relevant to validity in CF, issues specific to a number of types of validity issues in CF are described.
Incremental Validity The CF must yield explanations, predictions, or intervention implications that justify the cost of the additional complexity and effort over and above standardized assessment using nomothetic models of psychopathology, such as a disorder-level theory. The criteria for evaluating the validity of a CF for a particular case are
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
42
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
analogous to the incremental validity criteria necessary for the development of a new standardized measure of a nomothetic construct (cf. Haynes and Lench, 2003).
Individual and Aggregate Levels Validity issues in CF are complicated by the fact that the formulations are focused on the individual – each CF is an idiographic theory of the person-situation. Validity is relevant on this individual level – how valid is the formulation of this particular individual? Specifically, how accurate are predictions of behavior for a particular individual? Does the CF adequately cover the various facets of the target problem and the relevant causal variables? Does the CF for this person improve treatment planning and outcome for him/her? Validity issues are also relevant on the aggregate level. Here, concerns focus on the validity of the approach and its theoretical framework. For example, does this CF approach improve prediction (e.g., of suicide attempts) or treatment outcomes over and above that of standardized protocols?
Idiographic Assessment The use of idiographic or individualized assessment has important implications for validity on both the aggregate and individual levels. Comparisons of formulationbased versus manualized or standardized treatment, for example, have almost exclusively measured nomothetic constructs – idiographic measurement has generally not been used. However, measurement of individually relevant and specific problems, relationships, including functional relationships, and intervention targets is an important aspect of CF on the individual level within both behavioral and cognitive frameworks. Indeed, evaluating the validity of a CF for a particular individual without including individually-relevant (idiographic) problems, goals, mechanisms, and outcomes is probably inadequate (Haynes and O’Brien, 2000; Nelson-Gray, 1996; Tarrier and Calam, 2002).
Approach- or Theory-Specific Aspects of Validity As mentioned, certain validity issues are relevant generally and apply to CFs regardless of the particular approach within which it was developed. Alternatively, the particular issues relevant to other kinds of validity may vary with the theoretical orientation within which the CF is developed. For example, the validity of stimulus generalization and functional response classes is relevant to behavioral approaches to CF (Sturmey, 2008); the validity of measures of negative self-referent cognitive schema is relevant in a cognitive-behavioral CF for depression (J. Beck, 1995; Persons, Davidson and Tompkins, 2001); and the validity of central maladaptive relationship patterns is relevant to the CCRT or Configurational Analysis (Horowitz, 1997; Luborsky and Crits-Christoph, 1998). This concern is further complicated by the increasing number of approaches to CF within a particular theoretical framework. For example, there are different additional validity issues relevant to different approaches to behavioral case formulation, such as the basic learning approaches of Wolpe, Turkat and Follette (Sturmey, 2008), the Clinical
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
43
Pathogenesis Map (Nezu, Nezu and Lombardo, 2004) and the Functional Analytic Clinical Case Model (Haynes and O’Brien, 1990, 2000).
The Complexity of Validity in CF Validity issues in CF have received somewhat scant attention compared to reliability concerns. This may, in part, be due to the complexity of these concerns. Some different types of validity issues relevant to CF are summarized in Table 2.1. Notice that for each type of validity, relevant issues can be defined on the individual and aggregate level. Also, there are several distinguishable facets for a number of these types of validity. For example, as discussed above, content validity involves facets or components of coverage and relevance. Given space limitations, we will focus on several of the validity issues that may be particularly relevant for forensic CF.
Predictive Validity Within traditional psychometric frameworks, criterion validity is typically broken down as to whether the measure is predicting a criterion at the present time (concurrent validity) or in the future (predictive validity). For example, scores on a standardized achievement test may concurrently correlate with students’ grades, as determined by other measures, in a class taken when the test was administered; or the test may predict future performance, such as a grades as a graduate student in that discipline. However, for validity issues pertinent to CF, I will use the term criterion validity to compare the CF skills of novice or trainee clinicians to experts, where the latter is the criterion. This is discussed briefly later. An additional meaning of criterion validity – expected differences in the CFs of patients from a target group versus another population – is covered under discriminant validity issues (see Table 2.1). Predictive validity in CF refers to how well the CF predicts the behavior of the client, when this behavior is empirically measured as opposed to reported during an interview. The CF should enable prediction of behavior with greater specificity and precision than would be obtained from aggregate-level studies. For example, research using between-persons designs indicates a number of variables that are predictive of suicide attempts or urges, including prior suicide attempts, substance abuse, family history of suicide or psychiatric hospitalization, etc. (Qin, Agerbo and Mortensen, 2002). The risk or probability of a suicide attempt can be estimated from a person’s scores on these predictors using between-persons logistic regression. However, the CF may specify additional conditions or situations under which suicide attempts/urges are more probable for a particular person. Perhaps a client finds interactions with a verbally abusive mother in law particularly distressing, such that a recent interaction substantially increases the likelihood of suicidal urges. Another person may be more likely to make an attempt when she is feeling overwhelmed, has thoughts that she is worthless at work, and has just had an argument with her teenager. Thus, the CF can increase the level of specificity and the situational embeddedness of triggering variables so that they are fine
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
44
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Table 2.1 Validity issues in case formulation. Validity type
Level
Component/ facet/focus
Content
Individual
Coverage Relevance
Aggregate
Ecological
Coverage
Individual Aggregate
Criterion
Both
Predictive
Individual
Assessment
Aggregate
Construct Convergent
Individual
Item level (multi-item measurement)
Aggregate
Item level
Both
Variable level
Both
Formulation
Individual
Across types of CFs
Validity issues Are important target problems/issues covered in CF? Are important causal or concomitant variables included? Are target and causal variables/ constructs relevant to that person? Are the facets and elements of each variable selected the most relevant for that person? Are standardized measures used valid for this population in this context? Does the disorder level model include empirically based causal variables? Are the variables operationalized and measured so as to maximize “real life” functioning and issues? Are standardized measures tapping into real-life, functional issues for this population (i.e., not just diagnostic criteria)? How do CF skills of novices or trainees compare to experts? How well does the CF predict empirically measured relationships between variables within that person? Does a CF approach improve ability to predict the likelihood or severity of a target behavior compared to prediction from nomothetic aggregate studies? Do items converge to measure a variable or construct for that person? Which items are better measures of the targeted variable/ construct for that person? How well do certain items measure the target construct for individuals with that problem or disorder? Do scores on different measures of a variable or construct converge (agree), especially when different methods are used? Do different types of raters/ formulators agree/converge on CF? Do different types of CFs developed by independent formulators converge on certain characteristics of the case?
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
45
Table 2.1 (Continued ) Validity type
Level
Component/ facet/focus
Discriminant
Individual
Item level
Aggregate
Item level
Aggregate
Variable level
Both
CF level: Between individuals CF level: Between groups
Both
Intra-individual Tx Response
Both
Both
Within-person generalizability
Treatment
Individual
Prediction of tx response Comparison to other txs
Aggregate
Between groups: RCT
Within group
Both
Sharing CF with pt
Validity issues Do items intended to measure a construct for that person not measure another construct? Do items intended to measure a construct not load significantly on factors for other constructs? Does the average score on a standardized measure differ between those with target problem/disorder versus those with other problems/disorders? Does the CF or aspects of the CF for this person correctly discriminate between CFs for other individuals? (Method of mis-matched cases) Do CFs or aspects of the CF for individuals with this problem/disorder differ, on average, from CFs for other populations? When using the CF as a (complex) outcome measure, do components change in predicted ways in response to tx? Does the CF or aspects of the CF generalize across data obtained from different situations or contexts? Does the CF predict the pt’s response to a specific intervention or type of tx, or to an event occurring within session? Are CF-based interventions for this person more helpful/useful than no treatment, standardized tx, or previous tx? Does CF-based tx result in better outcomes than standardized tx for this problem, disorder, or population? (See Chapter 3 by Ata Ghaderi.) Within a sample receiving tx, do pts receiving interventions consistent with the CF have better (sub)outcomes than those receiving inconsistent interventions? Does sharing the CF with the pt impact outcome or the therapeutic alliance?
Note: Tx = Treatment (or intervention). RCT = Randomized controlled clinical trial. Pt = Patient. CF = Case formulation.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
46
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
tuned to a particular individual. Situational embeddedness can be external, such as the specific people with whom the person is interacting, or internal, such as the person’s thoughts or beliefs. Such highly specific variables may not predict enough of the proportion of variance in the outcome measure, suicide attempts/urges in this example, in aggregate-level, nomothetic research, to be clinically or statistically significant. Also, the number of individuals for whom the situation is relevant may be so small as to not justify measuring it for all individuals. However, such idiosyncratic external and internal conditions or stimuli can be assessed idiographically and might have strong predictive power within a particular individual (Haynes, Mumma and Pinson, 2009; Mumma, 2001; Mumma and Mooney, 2007b). Note that within a functional analytic approach to behavioral CF, such prediction is inherent to specifying discriminative stimuli, establishing operations, or contingencies relevant for a particular individual.
Predictive Validity in Behavioral CF An important aspect of predictive validity in CF is that the level of a variable (e.g., anger) and the relationship between variables (e.g., anger above a certain level when in a particular situation) may be functionally related to a target problem (e.g., they trigger physical aggression). For example, based on data from a clinical interview with a couple, the clinician may hypothesize that the probability of aggression from Bill toward his partner increases greatly once his anger passes a particular level, but only when there has been a recent argument and Bill feels stressed from his job. Aggression may be measured with a number of items including impulses or action tendencies (“felt like hitting”), automatic thoughts (image of hitting in the past), or verbal aggression (e.g., insults, swearing). Predictive validity for this nonlinear CF prediction can be evaluated using data collected by both Bill and his partner (cf., Haynes, 1992). Given the strong emphasis on assessing functional relationships and on collecting data through behavioral observation, the predictive validity of many behavioral CFs may be quite good (Cipani and Golden, 2007).
Predictive Validity in Cognitive-Behavioral CF Cognitive schemas are given a central role in Beck’s theories of depression and anxiety (Clark and Beck, 1999, 2010). The specificity of cognitive content to certain disorders and types of distress has been the focus of a number of between-subjects, aggregate studies (Clark and Beck, 1999; Smith and Mumma, 2008). Although a number of investigators have urged careful empirical study of the relationship between cognitions and behavior/symptoms hypothesized in CFs (Bieling and Kuyken, 2003; Kuyken, 2006), relatively little empirical work has been done to evaluate hypothesized relationships between cognitions and distress on the intraindividual level. Mumma and colleagues (Mumma, 2004; Mumma and Mooney, 2007a, b) have addressed the dearth of work on predictive validity in CBCF by using daily ratings of items to test hypothesized relationships in CBCFs including thoughts/beliefs and stressful events predicting type and severity of distress (depression, anxiety). Data are analyzed using one or both of two approaches:
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
47
Dynamic time series regression is a form of multiple regression analysis that models serial dependence in the data while evaluating lagged relationships between the predictors and outcome variable.2 Confirmatory dynamic factor analysis is a form of structural equation modeling applied to multivariate times series data that evaluates (a) the concurrent and lagged relationships between item scores and constructs, (b) the lagged relationships within each construct, and (c) the concurrent and lagged relationships between predictor and outcome constructs. Further refinements of this approach have used an ecological momentary assessment paradigm involving multiple ratings per day (Riggins and Mumma, 2008) and could use observations and ratings by a partner.
Predictive Validity in Psychodynamic Case Formulations Three types of PD formulations focus on core maladaptive relationship patterns – the CCRT (Luborsky and Crits-Christoph, 1998), Configurational Analysis (Horowitz, 1997; Horowitz and Eells, 2007), both of which were briefly described above, and the Cyclical Maladaptive Pattern (Levenson and Strupp, 2007). . The latter involves idiosyncratic “vicious cycle(s)” of maladaptive interactions involving inflexible, negative, and self-defeating behavior of the patient, expectations of others’ reactions, and the report of others’ behaviors (p. 171). Although several studies have examined the relationship between, say, the CCRT for the client’s interactions with the therapist versus others (e.g., Fried, Crits-Christoph and Luborsky, 1992, 1998), client self-report of relationship episodes and the CCRT method were used for both. To meet the definition of predictive validity suggested in this chapter, CF hypotheses about relationship themes need to be validated against behaviors measured by methods other than interview-based client self-report. For example, predictive validity could be evaluated within sessions using the client’s response to therapist behaviors (cf. Curtis and Silberschatz, 2007) or to interactions with a family member. These interactions could be videotaped, coded, and evaluated for consistency with the CF predictions. Alternatively, extrasession events could be used, again provided behavior is measured using methods other than the patient’s interview-based report of these events. Plausible methods include direct observation of extrasession interactions, ratings by significant others, or ecological momentary assessment.
Construct Validity Part I: General Issues and Convergent Validity A CF, as an idiographic theory of the person-situation, consists of a number of variables or constructs, both target and causal. Thus, construct validity issues would seem relevant. Construct validity is a complex notion (Burns and Haynes, 2006; Nunnally and Bernstein, 1994; Shadish et al., 2002) that includes the following 2 Serial
dependency is present when the error or residual component of today’s depression score, the dependent variable, is correlated with yesterday’s error score on the dependent variable. An example of a lagged relationship would be yesterday’s score on a measure of hopelessness predicting today’s score on a measure of depression.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
48
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
issues: A construct (a) consists of multiple facets so that a single operationalization or item typically provides insufficient coverage and carries unwanted variance associated with other constructs (mono-operational bias) while each facet may include a number of response modes (Haynes and O’Brien, 2000); (b) is best measured using at least two different measurement methods (e.g., self report, observer rating) so as not to confound methods and target construct variance (Campbell and Fiske, 1959: Eid, Lischetzke and Nussbeck, 2006); (c) should be conceptually/theoretically defined in terms of its expected relationships to other constructs; and (d) these predicted relationships should be tested empirically using well-validated measures. Finally, if a nomothetic construct does not adequately cover the specificity, relevance, or uniqueness of a variable for a particular person, methods are available to evaluate the psychometrics of an idiographic measure. The two most important components of construct validity are convergent and discriminant validity (Campbell and Fiske, 1959). Each of these can be evaluated on item, measure (Burns and Haynes, 2006), and CF levels (see Table 2.1).
Convergent Validity on the Item Level This evaluates the extent to which different items tap into the same construct. For between-persons, aggregate level research, this is readily evaluated using factor loadings initially from exploratory then from confirmatory factor analyses. Items with higher factor loadings are better measures of (i.e., share more variance with) the construct. A recent model developed by Haynes, Mumma and Pinson (2009) permits the importance (loading) of an item to vary across individuals. For example, the item “My face feels flushed” as a measure of anger may have a high loading (i.e., is a good measure of anger) for one individual but a relatively low loading for another. On the individual level, convergent validity of items can be evaluated with repeated ratings on the items, such as with daily ratings or ecological momentary assessment. Intra-individual item loadings can be evaluated with P-technique3 – or, preferably, confirmatory dynamic factor analysis (see Mumma, 2004; Mumma and Mooney, 2007a for examples in cognitive-behavioral CF). The items may measure a nomothetic or an idiographic construct or variable.
Convergent Validity on the Measure Level This is probably most familiar to readers – to what extent do different measures of a construct or variable converge or agree in their scores, particularly if different methods are used (Campbell and Fiske, 1959)?
3 P-technique
factor analysis (Cattell, 1978) uses the same methods as the more familiar R-technique factor analysis except the rows of the data matrix are the item scores from repeated observations within the same individual instead of scores from different individuals. Unlike dynamic factor analysis, however, P-technique factor analysis does not model serial dependency either in the form of autocorrelated errors/residuals or lagged relationships between factor scores. Essentially, the rows (occasions) can be randomly shuffled and the results will not be affected.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
49
Interformulator Convergent Validity This involves having different raters, or types of raters or formulators rate, evaluate or develop the CF or its components. This issue overlaps somewhat with interformulator reliability but is broader in that it encompasses convergence across different types of raters who may vary in perspective, extent of training, etc. (cf. Burns and Haynes, 2006). For example, in developing self-scenarios and interpersonal scenarios, Muran and colleagues (Muran and Segal, 1992; Muran et al., 1998, 2001) had the scenarios and their components rated for relevance by the formulator, patient, interviewer, and observer.
Convergent Validity across Different Types of Formulations In a far too rare evaluation of convergent validity between formulation approaches, Perry et al. (1989) report the similarities between components of the CCRT, an early version of the Plan Formulation Method, and the Idiographic Conflict Formulation based on a single case. Each of the three formulations was developed independently by different teams using the videotape of a single session. The authors report “significant similarity” (p. 318) between items generated when several components shared by these three methods were compared.4
Construct Validity Part II: Discriminant Validity Discriminant validity, the second major component of construct validity, was originally conceptualized in terms of whether two measures targeting two different constructs have empirically distinct scores, even when the two measures share the same method (such as self-report; Campbell and Fiske, 1959). Types of discriminant validity relevant to CF are generally analogous to the types of convergent validity discussed above.
Item Level Again, a distinction between aggregate and individual level concerns can be made. For between-person, aggregate level research, item-level discriminant validity is present if an item intended to measure construct A does not also load on construct B (i.e., is not a double-loading or complex item). This structure can be tested with confirmatory factor analysis (Anderson and Gerbing, 1988). On the individual level, discriminant validity has been evaluated in several ways. First, Mumma and colleagues (Mumma, 2004; Mumma and Mooney, 2007a) used confirmatory dynamic factor analysis on daily ratings by a patient to evaluate whether each item loaded on the target construct (e.g., depression) but not other factors (e.g., anxiety). Second, investigators from a number of different theoretical frameworks have a used the method of mismatched cases (Levine and Luborsky, 1981) to evaluate 4 Unfortunately, the authors report only that these comparisons were made with paired t-tests and were
statistically significant “<.05.” More explicit quantitative results are not reported.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
50
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
whether relevance ratings of items developed for the target person are higher than relevance ratings for items taken from other cases (Curtis et al., 1994; Curtis and Silberschatz, 2007; Mumma, 2004; Mumma and Smith, 2001; Muran and Segal, 1992; Muran et al. 2001)
Aggregate Level: Disorder-Related Discriminant Validity A basic aspect of discriminant validity for CFs is that the formulations should be expected to differ on theoretically and clinically meaningful parameters for persons with different disorders or types of target problems. Of course, a major purpose of the CF is to include individuating information, but disorder- or problem-level differences may represent an important form of validation. Several studies of psychodynamic CFs have addressed disorder-related discriminant validity. For example, studies of the CCRT have indicated systematic differences in one or more of its three components between patients with borderline personality disorder versus other personality disorders (Drapeau and Perry, 2009), bulimia patients versus nonpatients (Benninghoven et al., 2003), child molesters ¨ versus nonmolesters (Drapeau, DeRoten and Korner, 2004), and as a function of severity of alexithymia and chronic fatigue (Vanheule et al., 2007; Vandenbergen et al., 2009). Although the research on the CCRT in this area is generally more substantial than for other types of CFs, the above results should be replicated before considered reliable differences. Although quite a number of cognitive-behavioral CF studies have compared groups with various disorders or types of distress (e.g., depression, anxiety) for cognitive content specificity, I am not aware of any studies directly comparing cognitive-behavioral CFs of individuals with different disorders. This is complicated by the increasing use of disorder-specific templates for cognitive-behavioral CFs (e.g., Boschen and Oei, 2008; Tarrier, 2006).
Changes in CF Associated with Treatment Response Another form of discriminant validity relevant to CFs involves expected changes in the CF in response to intervention. This issue differs from treatment validity, which examines the effects of CF on treatment response. Specifically, for treatment validity, the independent variable is type of treatment (formulation versus manualized) and the dependent variable is various outcome measures of distress or functioning. For this type of discriminant validity issue, however, the dependent or outcome variable is the CF and the predictor is theoretically relevant treatment or degree of success in treatment implementation. Studies examining this type of discriminant validity include tests of three theoretically relevant predictions for the CCRT: (a) psychotherapy changes the nature of the three CCRT components (wishes, response of other, response of self) from more negative to more positive (Wilczek et al., 2004); (b) the pervasiveness of the negative components of the CCRT across various relationship episodes for patients should decrease in response to psychotherapy, indicating greater flexibility or less rigidity in negative relationship theme components (e.g., fewer responses of self involving depression, disappointment, or anger) (Crits-Christoph and Luborsky, 1998;
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
51
Wilczek et al., 2004); and (c) the degree of change in either of the above should be related to the degree of symptom reduction (Crits-Christoph and Luborsky, 1998).
Within-Patient Convergent-Discriminant Validity – Generalizability across Facets Does the CF or aspects of the CF generalize across data obtained from different situations or contexts for a particular person or for individuals within a sample? For behavioral CF, situational specificity is an important issue in a functional analysis. The presence/absence or severity of a target behavior in one context but not another may yield important information about functionally relevant controlling variables and contingencies (Sturmey, 2008). Thus, this issue would be expected to be addressed explicitly during the behavioral assessment process when developing the behavioral CF. For psychodynamic CFs, studies evaluating this convergent/discriminant validity issue have included the generalizability of the three major components of the CCRT across the following facets: relationship episode narratives focused on the therapist versus significant others, different types of others (e.g., parents versus peers for adolescents), different stages of therapy, relationship episodes from therapy transcripts versus dreams, and relationship episodes obtained from therapy sessions versus a research-based interview used to extract multiple relationship episodes (the Relationship Anecdotes Paradigm) (Beretta, et al., 2007; Popp et al., 1996; Barber et al., 1995; Fried et al., 1992, 1998; Tishby, Raitchick and Shefler, 2007). Finally, items constructed for the same component of Horowitz’s Role Relationship Model Configuration (Desired, Dreaded, Adaptive Compromise, vs. Problematic Compromise) were rated as significantly more similar than items developed for different components, different objects (e.g., spouse versus therapist), or both (Eells et al., 1995).
General Comments and Issues on Construct Validity in CF Surprisingly, with the exception of certain psychodynamic CFs, issues of construct validity in CF have received rather limited attention conceptually, methodologically, and empirically. For example, two recent comprehensive guides to a widely used psychodynamic and cognitive-behavioral CF (Luborsky and Crits-Christoph, 1998; Persons, 2008) do not include “validity” or any of the types of validity discussed above as an entry in the index. Despite the central role of validity issues in psychometrics, with some exceptions such concerns have generally been skirted for CFs. Forensic CF may involve target behaviors that are relatively rare or infrequent or are sufficiently severe that intervention options may be limited by legal and dispositional actions (e.g., incarceration). To ethically and pragmatically enable treatment of some target behaviors, and to increase the potential relevance and treatment utility of behavioral or cognitive-behavioral CFs, the clinician may need to focus on the idiographic construct validity of the target behavior. For example, other manifestations of the target behavior may need to be included that are more
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
52
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
frequently and readily accessible and are less extreme (Haynes and O’Brien, 2000). For violent behavior, this may include minor aggressive behaviors, impulses or urges for violent behavior (action tendencies), thoughts of violence, anger, or even physiological arousal (e.g., increased heart rate). Proxy measures involving verbal reports of action tendencies may be particularly useful (e.g., “I felt like hitting”). All of these may be considered facets or elements of the target construct. Within a behavioral perspective, they would be conceptualized as a functional response class, and evidence for their functional equivalence would need to be obtained (Haynes and O’Brien, 1990, 2000). Of course, the specific behaviors, facets, or elements will vary from person to person (Haynes et al., 2009). Within each person, though, their relationship with each other could be empirically evaluated in terms of which behaviors co-occur with others. Treatment efficacy might be measured using those elements that have high centrality or loadings on that individual’s construct of violence or aggression, but which are subthreshold to actual violent acts (e.g., verbal reports of action tendencies or automatic thoughts) and present in the situations of greatest relevance for that person.
SCIENTIFIC BASES IV: THE CASE FORMULATION AND TREATMENT Issues of treatment validity or treatment utility of the CF as evaluated using experimental designs (randomized controlled clinical trials) are covered in Chapter 3 by Ata Ghaderi. Three additional issues relevant to the CF and treatment are discussed in this section.
Prediction of an Individual’s Response to a Specific Intervention or Therapy Event A CF might help the clinician predict the client’s response to a specific intervention or type of treatment. This is analogous to predictive validity (discussed above) except here it is a prediction of client behavior in response to an intervention. This form of treatment validity could be further divided into predicting the person’s response to events occurring within a treatment session (e.g., in response to the therapist’s intervention) versus response to an intervention made outside of the session. This type of treatment validity is central to behavioral CF – using the functional assessment, the clinician predicts the client’s response to different types of interventions and then selects that intervention most likely to achieve the treatment goals. Using the CF to predict client response to an intervention outside of the session is central to behavioral intervention for children, such as when working with parents or teachers on changing contingencies for the child’s behavior (McMahon, 1999). Such predictions may even be tested using single-subject experimental designs – such as an ABAB reversal design (Kazdin, 2003). Sommerfeld and colleagues (2008) provide an example of this type of treatment validity using a psychodynamic CF. The CCRT assessed prior to treatment was used to predict the
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
53
patient’s responses and his/her expectations for the therapist’s responses when alliance ruptures occurred during the session.
Effect of Sharing the CF with the Patient This issue has been studied infrequently. Two studies (Chadwick, Williams and Mackenzie, 2003; Pain, Chadwick and Abba, 2008) examined the effects of feedback about their cognitive-behavioral CFs on the therapeutic alliance and measures of delusional beliefs and other distress in adults with psychotic disorders. Both studies found mixed results, with some individuals reporting benefits and others problems. Limitations of these studies include the patient population and providing feedback during a single session (cf. Kuyken et al., 2009). Working within a psychodynamic/interpersonal orientation (cognitive analytic therapy) and using a multiple baseline design, Evans and Parry (1996) found that sharing a written CF with each of four patients had no impact on an idiographic measure of problems/issues or on the therapeutic alliance. As Aston (2009) stated regarding the effect of sharing the CF with the client: “there is still a distinct lack of clarity as to what the client finds the most useful and the most unhelpful” (p. 67.)
Are Therapy Interventions Congruent with the CF More Beneficial? This issue pertains to treatment validity but does not use a randomized controlled, clinical trial format (see Table 2.1). The few studies of this type explore whether within a sample of individuals receiving treatment, naturally occurring (i.e., nonmanipulated) interventions that are more versus less congruent with the CF effect outcome or suboutcome. There are several good examples of this type of treatment validity evaluation for psychodynamic CFs. Using a single-group pre-post design, Crits-Christoph, Cooper and Luborsky (1998) reported that therapists’ interpretations that were more accurate – more congruent with the wish and response components of the CCRT – predicted positive treatment outcome (residual gain) both by itself and incrementally over prediction from the therapeutic alliance. In an intensive study of two cases, Messer, Tishby and Spillman (1992) showed that interventions that were consistent with the Plan Formulation (Curtis et al., 1994) predicted progress as opposed to stagnation during that session and that this relationship was strongest during the early and middle phases of psychodynamic treatment.
Behavioral and Cognitive-Behavioral CF Behavioral intervention studies that use a reversal or multiple baseline design explicitly address this issue by evaluating differential response after the intervention. Studies that compare different types of interventions for different types of reinforcers of the target behavior provide important validation for functional relationships hypothesized in the behavioral CF. Studies of self-injurious behavior by Iwata and collegues (1994) and studies of school refusal by Kearney and Silverman
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
54
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
(1999) exemplify this type of study. Although difficult to find analogous studies within a cognitive CF approach, the relatively few studies examining the effects of matching the type of therapy to functionally relevant characteristics of the patient are relevant (e.g., Nelson-Gray et al., 1989).
TRAINING AND COMPETENCY ISSUES Criterion Validity: Using Experts’ CFs to Evaluate Training in CF Skills One approach to criterion validity uses the CF of an expert(s) as the criterion to which the CFs of other clinicians is compared (Table 2.1). Although relatively few studies have been done in this area, there are some encouraging results. In studies of cognitive-behavioral CF reliability described above, Persons and Bertagnolli (1999) and Persons and colleagues (2001) found that clinicians who received brief training in cognitive-behavioral CF were relatively good at identifying problems identified by the expert clinician, but substantially poorer at identifying inferred mechanisms, such as cognitive beliefs or schema. Kuyken and colleagues (2005) found similar results comparing cognitive-behavioral CFs of trainees with those of experts using J. Beck’s (1995) approach to cognitive-behavioral CF. However, the absence of a pretest or of a control group in these studies leaves unclear the extent to which the cognitive-behavioral CF skills of the trainees were due to the brief training. In contrast, in a randomized experimental study of trainees from several disciplines, Kendjelic and Eells (2007) found that clinicians who received brief generic training in four basic components of CF (symptoms/problems, predisposing variables, precipitating stressors, and inferred mechanism) produced formulations higher in overall quality, comprehensiveness, and elaboration of inferred mechanism, as well as greater complexity and precision of language. As noted by the authors, although these results were encouraging, there was a lot of room for further improvement. ¨ Horowitz and Moller (2009) trained 24 psychodynamic and 17 cognitive behavioral therapists to use the Role Relationship Model Configuration to conceptualize a current patient “well known” to the clinician. Although most of the therapists rated the approach as helpful, an evaluation of CF quality was not undertaken. Although the above studies evaluating brief training in CF skills are helpful, CF is a high-level skill which is likely to require a systematic program of supervised training experiences involving multiple cases over a period of time (Kuyken et al., 2009). Studies evaluating the effects of more prolonged types of training in CF skills are needed.
Level of Training and Expert/Novice Differences in CF In a study mentioned above, Kuyken and colleagues (2005) found that the quality of a cognitive-behavioral CF using J. Beck’s (1995) Case Conceptualization Diagram completed after a half-day training workshop, was associated with level of training and credentialing (prequalified, qualified, qualified and accredited). Eells, Lombart and Kendjelic (2005) found expert–novice differences between clinicians
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
55
using psychodynamic or cognitive behavioral frameworks on measures of CF comprehensiveness, degree of elaboration of the formulation and treatment plan, precision of language, complexity of the formulation, fit of the CF to the treatment plan, and use of a systematic approach to formulating cases. Likewise, Mumma and Mooney (2007a) found differences between an expert versus novice clinician in the complexity, elaboration, and predictive validity of the cognitive-behavioral CF of a single case, with the expert’s formulation explaining approximately twice the variance in daily ratings of distress than the novice’s CF. The development of coding systems evaluating multiple dimensions of CF quality (e.g., Eells et al., 2005), should enhance further empirical efforts to study differences between expert, novice, and experienced clinicians.
SUMMARY AND CONCLUSIONS The recent increase of interest in CF and the recognition of CF as a high-level but core skill for effective psychological treatment, particularly for complex or comorbid cases, is an exciting process that has the potential to improve the quality of interventions offered to clients by individualizing or tailoring treatment to the specific needs of a particular person. However, along with the development of more manuals describing how to develop CFs, both researchers and clinicians need in increase their efforts to evaluate the scientific status of the resulting CFs, as well as the validity of the various approaches to CF within each theoretical framework. This chapter has described some of the more important psychometric issues relevant to CF, including several types of reliability and validity. Issues of content validity, predictive validity, construct validity, and treatment validity are particularly important, and each of these types of validity has a number of different forms or issues relevant to CF and its components (Table 2.1). Not every form or issue needs to be addressed for CFs focusing on different problems or disorders, but it behooves us as scientists and practitioners to know the status of reliability and especially validity studies relevant to the CFs used for tailored, formulation-based treatments for the clients we consult to and treat. For CF, an essential tension exists between structure and flexibility. Structure is provided by manuals describing how to develop CFs, templates for CFs for particular disorders, typically based on nomothetic causal models for that problem or disorder, standardized measures of target problems and of potentially relevant causal variables, semi-structured or structured interviews, behavioral observation procedures, etc. Ideally, this structure functions to increase the accuracy of the CF and the clinical judgments resulting from it not only by increasing the content and potentially ecological validity of the CF, but also by reducing judgmental and inferential biases that are present in human thought and from which the clinician is not immune (Garb, 1998; Haynes and O’Brien, 2000). On the other hand, flexibility is necessary to explore and idiographically measure the unique and idiosyncratic aspects of the case, including variables that may not adequately represented in standardized measures or models of psychopathology and functional relationships that are present intra-individually. Flexibility is needed to address the range of problems and causal mechanisms particularly relevant for a complex case and to
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
56
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
design a tailored treatment plan that addresses these problems and issues in a systematic way while building on the strengths of a particular client.
REFERENCES Anderson, J.C. and Gerbing, D.W. (1988) Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 3, 411–23. Aston, R. (2009) A literature review exploring the efficacy of case formulations in clinical practice. What are the themes and pertinent issues? Cognitive Behavior Therapist, 2, 63–74. Barber, J.P., Luborsky, L., Crits-Christoph, P. and Diguer, L. (1995) A comparison of core conflictual relationship themes before psychotherapy and during early sessions. Journal of Consulting and Clinical Psychology, 63, 145–8. Beck, J. S. (1995) Cognitive Therapy: Basics and Beyond. New York: Guilford. Benninghoven, D., Schneider, H., Strack, M., Reich, G., and Cierpka, M. (2003) Family representations in relationship episodes of patiens with a diagnosis of bulimia nervosa. Psychology and Psychotherapy: Theory, Research, and Practice, 76, 323–36. Beretta, V., Despland, J.N., Drapeau, M., et al. (2007) Are relationship patterns with significant others reenacted with the therapist? A study of transference reactions. Journal of Nervous and Mental Diseases, 195, 443–50. Bieling, P.J and Kuyken, W. (2003) Is cognitive case formulation science or science fiction? Clinical Psychology: Science and Practice, 10, 52–69. Boschen, M.J. and Oei, T.P.S. (2008) A cognitive behavioral case formulation framework for treatment planning in anxiety disorders. Depression and Anxiety, 25, 811–23. Burns, G.L. and Haynes, S.N. (2006) Clinical psychology: Construct validation with multiple sources of information and multiple settings. In M. Eid and E. Diener (eds), Handbook of Multimethod Measurement in Psychology (pp. 401–18). Washington, DC: American Psychological Association. Campbell, D. and Fiske, D. (1959) Convergent and discriminant validation by the multitraitmultimethod matrix. Psychological Bulletin, 56, 81–105. Cattell, R.B. (1978) The Scientific Use of Factor Analysis in Behavioral and Life Sciences. New York: Plenum. Chadwick, P., Williams, C., and Mackenzie, J. (2003) Impact of case formulation in cognitive behaviour therapy for psychosis. Behaviour, Research and Therapy, 41, 671–80. Cipani, E. and Golden, J.A. (2007) Differentiating behavioral and traditional case formulations for children with severe behavioral and emotional problems. International Journal of Behavioral Consultation and Therapy, 3, 537–45. Clark, D.A. and Beck, A.T. (1999) Scientific Foundations of Cognitive Theory and Therapy of Depression. New York: John Wiley & Sons, Inc. Clark, D.A. and Beck, A.T. (2010) Cognitive Therapy of Anxiety Disorders: Science and Practice. New York: Guilford. Crits-Christoph, P., Cooper, A. and Luborsky, L. (1998) The measurement of accuracy of interpretations. In L. Luborsky and P. Crits-Christoph (eds), Understanding Transference: The Core Conflictual Relationship Theme Method. (2nd edn, pp. 197–212). Washington, D.C.: American Psychological Association. Crits-Christoph, P. and Luborsky, L. (1998) Changes in CCRT pervasiveness during psychotherapy. In L. Luborsky and P. Crits-Christoph (eds), Understanding Transference: The Core Conflictual Relationship Theme Method. (2nd edn, pp. 151–64). Washington, D.C.: American Psychological Association. Curtis, J.T. and Silberschatz, G. (2007) The Plan Formulation Method. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 198-220). New York: Guilford. Curtis, J.T., Silberschatz, G., Sampson, H. et al. (1988) Developing reliable psychodynamic case formulations: An illustration of the plan diagnosis method. Psychotherapy, 25, 256–65. Curtis, J.T., Silberschatz, G., Sampson, H. and Weiss, J. (1994) The Plan Formulation Method. Psychotherapy Research, 4, 197–207.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
57
Drapeau, M. and Perry, J.C. (2009) The Core Conflictual Relationship Themes (CCRT) in borderline personality disorder. Journal of Personality Disorders, 23, 425–31. ¨ Drapeau, M., de Roten, Y. and Korner, A.C. (2004) An exploratory study of child molesters’ relationship patterns using the Core Relationship Theme Method. Journal of Interpersonal Violence, 19, 264–75. Eells, T.D., Horowitz, M. J., Singer, J. et al. (1995) The Role-Relationships Models method: A comparison of independently derived case formulations. Psychotherapy Research, 5, 154–68. Eells, T.D., Lombart, K.G. and Kendjelic, E.M. (2005) The quality of psychotherapy case formulations: a comparison of expert, experienced, and novice cognitive-behavioral and psychodynamic therapists. Journal of Consulting and Clinical Psychology, 73, 579–89. Eid, M., Lischetzke, T. and Nussbeck, F.W. (2006) Structural equation models for multitraitmultimethod data. In M. Eid and E. Diener (eds), Multimethod Measurement in Psychology (pp. 283–300). Washington, DC: American Psychological Association. Evans, J. and Parry, G. (1996) The impact of reformulation in cognitive-analytic therapy with difficult-to-help clients. Clinical Psychology and Psychotherapy, 3, 109–17. Fleiss, J. (1981) Statistical Methods for Rates and Proportions (2nd edn). New York: John Wiley & Sons, Inc. Fried, D., Crits-Christoph, P. and Luborsky (1992) The first empirical demonstration of transference in psychotherapy. Journal of Nervous and Mental Diseases, 180, 326–31. Fried, D., Crits-Christoph, P. and Luborsky (1998) The parallel of the CCRT for the therapist with the CCRT for other people. In L. Luborsky and P. Crits-Christoph (eds), Understanding Transference: The Core Conflictual Relationship Theme Method (2nd edn, pp. 175–96). Washington, D.C.: American Psychological Association. Furr, M.R. and Bacharach, V.R. (2008) Psychometrics: An Introduction. Thousand Oaks, CA: Sage. Garb, H.N. (1998) Studying the Clinician: Judgment Research and Psychological Assessment. Washington, DC: American Psychological Association. Haynes, S.N. (1992) Models of Causality in Psychopathology: Toward Dynamic, Synthetic, and Nonlinear Models of Behavior Disorders. New York: Macmillan. Haynes, S.N., Kaholokula, J.K. and Nelson, K. (1999) The idiographic application of nomothetic, empirically based treatments. Clinical Psychology: Science and Practice, 6, 456–61. Haynes, S.N. and Lench, H.C. (2003) Incremental validity of new clinical assessment measures. Psychological Assessment, 15, 456–66. Haynes, S.N., Mumma, G.H. and Pinson, C. (2009) Idiographic assessment: Conceptual and psychometric foundations of individualized behavioral assessment. Clinical Psychology Review, 29, 179–91. Haynes, S.N. and O’Brien, W.H. (1990) Functional analysis in behavior therapy. Clinical Psychology Review, 10, 649–68. Haynes, S.N. and O’Brien, W.H. (2000) Principles and Practice of Behavioral Assessment. New York: Plenum. Horowitz, M.J. (1997) Formulation as a Basis for Planning Psychotherapy Treatment. Washington, D.C.: American Psychiatric Press. Horowitz, M.J. and Eells, T.D. (2007) Configurational analysis: States of mind, person schemas, and the control of ideas and affect. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 136–63). New York: Guilford. ¨ Horowitz, M.J. and Moller, B. (2009) Formulating transference in cognitive and dynamic psychotherapies using role relationship models. Journal of Psychiatric Practice, 15, 25–33. Iwata, B.A., Pace, G.M., Dorsey, M.F., Zarcone, J.R., Vollmer, B. and Smith, J. (1994) The function of self-injurious behavior: An experimental-epidemiological analysis. Journal of Applied Behavior Analysis, 27, 215–40. Kazdin, A.E. (2003) Research Design in Clinical Psychology (4th edn). Boston: Allyn & Bacon. Kearney, C.A. and Silverman, W.K. (1999) Functionally based prescriptive and nonprescriptive treatment for children and adolescents with school refusal behavior. Behavior Therapy, 30, 673–95.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
58
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Kendjelic, E.M. and Eells, T.D. (2007) Generic psychotherapy case formulation training improves formulation quality. Psychotherapy: Theory, Research, Practice, Training, 44, 66–77. Kuyken, W. (2006) Evidence-based case formulation. In N. Tarrier (ed.), Case Formulation in Cognitive Behaviour Therapy (pp. 12–35). New York: Routledge. Kuyken, W. Fothergill, C.D., Musa, M. and Chadwick, P. (2005) The reliability and quality of cognitive case formulation. Behaviour, Research and Therapy, 43, 1187–1201. Kuyken, W., Padesky, C.A. and Dudley, R. (2009) Collaborative Case Conceptualization: Working Effectively with Clients in Cognitive-Behavioral Therapy. New York: Guilford. Lefebvre, R., Diguer, L., Morissette, E. et al. (1996) The borderline and their Core Conflictual Relationship Themes. As reported in Luborsky, L. and Crits-Christoph, Understanding Transference: The Core Conflictual Relationship Theme Method. Washington, D.C.: American Psychological Association. Levenson, H. and Strupp, H.H. (2007) Cyclical maladaptive patterns: Case formulation in time-limited dynamic psychotherapy. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 164–97). New York: Guilford. Levine, F.J. and Luborsky, L. (1981) The Core Conflictual Relationship Theme: A demonstration of reliable clinical inference by the method of mismatched cases. In S. Tutman, C. Kaye, and M. Zimmerman (eds), Object and Self: A Developmental Approach (pp. 501–20). Madison, CT: International Universities Press. Luborsky, L. and Crits-Christoph, P. (1998) Understanding Transference: The Core Conflictual Relationship Theme Method (2nd edn). Washington, D.C.: American Psychological Association. Luborsky, L. and Diguer, L. (1998) The reliability of the CCRT measure: Results from eight samples. In L. Luborsky and P. Crits-Christoph (eds), Understanding Transference: The Core Conflictual Relationship Theme Method (pp. 97–108). New York: Basic Books. Luborsky, L., Luborsky, E., Diguer, L. et al. (1998) Stability of the CCRT from age 3 to 5. In L. Luborsky and P. Crits-Christoph (eds), Understanding Transference: The Core Conflictual Relationship Theme Method (pp. 233–52). New York: Basic Books. McMahon, R.J. (1999) Parent training. In Russ and Ollendick (eds), Handbook of Psychotherapies with Children and Families (pp. 153–80). New York: Kluwer. Messer, S., Tishby, O. and Spillman, A. (1992) Taking context seriously in psychotherapy research: Relating therapist interventions to patient progress in brief psychodynamic therapy. Journal of Consulting and Clinical Psychology, 60, 678–88. Michael, P.G. and Li, S.T. (2010) Psychometrics. In J. Thomas and M. Hersen (eds), Handbook of Clinical Psychology Competencies (pp. 343–65). New York: Springer. Mumma, G.H. (2001) Increasing accuracy in clinical decision making: Towards an integration of nomothetic-aggregate and intraindividual-idiographic approaches. The Behavior Therapist, 24, 77–94. Mumma, G.H. (2004) Validation of idiosyncratic cognitive schema in cognitive case formulations: An intra-individual idiographic approach. Psychological Assessment, 16, 211–30. Mumma, G.H. and Mooney, S.R. (2007a) Comparing the validity of alternative cognitive case formulations: A latent variable, multivariate time series approach. Cognitive Therapy and Research, 31, 451–81. Mumma, G.H. and Mooney, S.R. (2007b) Incremental validity of cognitions in a clinical case formulation: An Intraindividual test in a case example. Journal of Psychopathology and Behavioral Assessment, 29, 17–28. Mumma, G.H. and Smith, J.L. (2001) Cognitive-behavioral interpersonal scenarios: Interformulator reliability and convergent validity. Journal of Psychopathology and Behavioral Assessment, 23, 203–21. Muran, J.C. Samstag, L.W. and Segal, Z.V. (1998) Interpersonal scenarios: An idiographic measure of self-schemas. Psychotherapy Research, 8, 321–33. Muran, J.C., Samstag, L.W., Ventur, E.D. et al. (2001) A cognitive-interpersonal case study of the self. Journal of Clinical Psychology, 57, 307–30. Muran, J.C. and Segal, Z.V. (1992) The development of an idiographic measure of selfschemas: An illustration of the construction and use of self-scenarios. Psychotherapy, 29, 524–35.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
May 28, 2011
9:33
Printer Name: Yet to Come
CURRENT ISSUES IN CASE FORMULATION
59
Needleman, L.D. (1999) Cognitive Case Conceptualization. Mahwah, NJ: Erlbaum. Nelson-Gray, R.O. (1996) Treatment outcome measures: Nomothetic or idiographic? Clinical Psychology: Science and Practice, 3, 164–7. Nelson-Gray, R.O., Herbert, J.D., Herbert, D.L. et al. (1989) Effectiveness of matched, mismatched, and package treatments of depression. Journal of Behavior Therapy and Experimental Psychiatry, 20, 281–94. Nezu, A.M., Nezu, C.M. and Lombardo, E. (2004) Cognitive-Behavioral Case Formulation and Treatment Design: A Problem Solving Approach. New York: Springer. Nunnally, J.C. and Bernstein, I.H. (1994) Psychometric Theory (3rd edn). New York: McGrawHill. O’Brien, W.H., Oemig, C.K. and Northern, J.J. (2010) Behavioral assessment with adults. In J. Thomas and M. Hersen (eds), Handbook of Clinical Psychology Competencies (pp. 343–65). New York: Springer. Pain, C.M., Chadwick, P. and Abba, N. (2008) Clients’ experience of case formulation in cognitive behavior therapy for psychosis. British Journal of Clinical Psychology, 47, 127–38. Perry, J.C., Luborsky, L., Silberschatz, G. and Popp, C. (1989) An examination of three methods of psychodynamic formulation based on the same videotaped interview. Psychiatry, 52, 302–23. Persons, J.B. (1989) Cognitive Therapy in Practice: A Case Formulation Approach. New York: Norton. Persons, J.B. (2008) The Case Formulation Approach to Cognitive-Behavior Therapy. New York: Guilford. Persons, J.B. and Bertagnolli, A. (1999) Interrater reliability of cognitive-behavioral case formulations of depression: A replication. Cognitive Therapy and Research, 23, 271–83. Persons, J.B., Davidson, J. and Tompkins, M.A. (2001) Essential Components of CognitiveBehavior Therapy for Depression. Washington, D.C.: American Psychological Association. Persons, J.B., Mooney, K.A. and Padesky, C.A. (1995) Interrater reliability of cognitivebehavioral case formulations. Cognitive Therapy and Research, 19, 21–34. Persons, J.B. and Tompkins, M.A. (2007) Cognitive-behavioral case formulation. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (2nd edn, pp. 290–316). New York: Guilford. Popp, C.A., Diguer, L., Luborsky, et al. (1996) Repetitive relationship themes in waking narratives and dreams. Journal of Consulting and Clinical Psychology, 64, 1073–8. Qin, P., Agerbo, E. and Mortensen, P. (2002) Suicide risk in relation to family history of completed suicide and psychiatric disorders: A nested case-control study based on longitudinal registers. Lancet, 360, 1126–30. Riggins, J. and Mumma, G.H. (2008, November) Using EMA to test cognitive-behavioral case formulation hypotheses. Presented at the 42nd Annual Convention of the Association of Behavioral and Cognitive Therapies, Orlando, FL. Shadish, W.R., Cook, T.D. and Campbell, D.T. (2002) Experimental and Quasi-Experimental Designs for Generalized Causal Inference. New York: Houghton Mifflin. Smith, P.N. and Mumma, G.H. (2008) A multi-wave web-based evaluation of cognitive content specificity for depression, anxiety, and anger. Cognitive Therapy and Research, 32, 50–65. Sommerfeld, E., Orach, I., Zim, S. and Mikulincer, M. (2008) An in-session exploration of ruptures in working alliance and their associations with clients’ core conflictual relationship themes, alliance-related discourse, and clients’ postsession evaluations. Psychotherapy Research, 18, 377–88. Spielberger, C.D., Syndman, S.J., Owen, A.E. and Marsh, B.J. (1999) Measuring anxiety and anger with the State-Trait Anxiety Inventory (STAI) and the State-Trait Anger Expression Inventory (STAXI). In M.E. Maurish (ed.), The Use of Psychological Testing for Treatment Planning and Outcome Assessment (2nd edn, pp. 993–1021). Mahwah: Erlbaum. Sturmey, P. (2008) Behavioral Case Formulation and Intervention. New York: John Wiley & Sons, Inc. Tarrier, N. (ed.) (2006) Case Formulation in Cognitive Behavior Therapy: The Treatment of Challenging and Complex Cases. New York: Routledge.
P1: OTA/XYZ P2: ABC JWST073-02 JWST073-McMurran
60
May 28, 2011
9:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Tarrier, N. and Calam, R. (2002) New developments in cognitive-behavioural case formulation. Epidemiological, systemic, and social context: An integrative approach. Behavioural and Cognitive Psychotherapy, 30, 311–28. Tishby, O., Raitchick, I. and Shefler, G. (2007) Changes in interpersonal conflicts among adolescents during psychodynamic therapy. Psychotherapy Research, 17, 301–9. Vandenbergen, J., Vanheule, S., Rosseel, Y. et al. (2009) Unexplained chronic fatigue and core conflictual relationship themes: A study in a chronically fatigued population. Psychology and Psychotherapy: Theory, Research, and Practice, 82, 31–42. Vanheule, S., Vandenbergen, J., Desmet, M. et al. (2007) Alexithymia and core conflictual relationship themes: A study in a chronically fatigued primary care population. International Journal of Psychiatry in Medicine, 37, 87–98. Wells, A. (2006) Cognitive therapy case formulation in anxiety disorders. In N. Tarrier, Case Formulation in Cognitive Behavior Therapy: The Treatment of Challenging and Complex Cases, (pp. 52–80). London: Routledge. Wilczek, A., Weinryb, R.M., Barber, J.P., Gustavsson, J.P. and Asberg, M. (2004) Change in the core conflictual relationship theme after long-term dynamic psychotherapy. Psychotherapy Research, 14, 107–25. Zander, B., Strack, M., Cierpka, M. et al. (1995a) Coder agreement using the german editions of Luborsky’s CCRT method in videotaped or transcribed RAP interviews. Psychotherapy Research, 5 (2), 231–6. Zander, B., Strack, M., Cierpka, M. et al. (1995b) Different reliabilities at the episode level and that of the final CCRT – a rejoinder to Luborsky and Diguer. Psychotherapy Research, 5, 242–4.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
Chapter 3
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME? ATA GHADERI Uppsala University, Sweden
Case formulation is a key skill for clinicians from many clinical disciplines (Sturmey, 2008). Many clinicians believe that treatment should be tailored to the need of the individual patient, instead of applying a standardized treatment protocol addressing the main concerns and problems of the patient. This is a logical assumption, and, despite the complexity when it comes to the application of case formulation-based treatments, it is in line with linear thinking about cause and effect, and analogous to the diagnosis-treatment thinking within the field of medicine. As human beings, we are complex and we respond in many different ways to environmental stimuli, depending on the conditions and demands in different contexts and our experiences, rules, and expectations. Therefore, understanding and influencing people might necessitate some level of individualization in terms of case formulation. Eells (2007) suggested that case formulation “lies at an intersection of diagnosis and treatment, theory and practice, science and art, and etiology and description” (p. 4). To approach a scientific answer to the question “Does case formulation make a difference to treatment outcome?”, we need to decide which theory of science we are using as the basis for our observations. In addition, before we review and critically evaluate the empirical evidence for the effect of case formulation on treatment outcome, we need to decide what truth criterion we will apply. Data will also guide us to revise and improve our theories and direct us regarding how we should obtain more data. It is beyond the scope of this chapter to review and discuss different theories of science and their implications for psychotherapy research and how each theoretical view might influence our interpretation of data. Nevertheless, I will adopt an underlying assumption of a need to balance what is common for all of us and what makes us unique as individuals: I explicitly assume that a functional approach to behavior, health, and psychopathology is superior
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
62
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
to a topographic view for prediction and influence. Also, I will take into account practical variables, such as the ease of use of case formulation, its treatment utility, its ease of integration with manual-based treatments, and the risks for therapist drift. Therapist drift might be defined as focusing on irrelevant variables instead of key tasks in therapy mainly due to our cognitive distortions, emotional reactions or use of safety behaviors (Waller, 2009).
CASE FORMULATION: NOT A SINGLE APPROACH There are many approaches to case formulation. In most cases, they relate to a particular theory. For example, approaches to case formulation can be based on emotions-focused therapy, interpersonal psychotherapy, psychoanalysis or different schools of thought within psychodynamic tradition. I will not review the evidence of reliability, validity and treatment utility of case formulation within psychodynamic psychotherapies and psychoanalysis, although some studies in these fields provide some support for the use of case formulation. Moderate to good reliability (.64 to .90) in case formulations has been found in several studies, although the estimates tend to plunge when they are translated to the level of individual therapists, as it would be in routine clinical practice (Eells, 2007). In addition, empirical research on the validity and treatment utility of case formulation within these traditions is scarce. With the exception of one study (Crits-Cristoph, Cooper and Luborsky, 1988), almost all of the data within these schools of psychotherapy are based on case studies and studies with very small samples (Eells, 2007). Data from these studies are analyzed at the individual level and the conclusions from each case are then aggregated to reach a final conclusion, instead of aggregating data, and then analyzing the data. This might be analogous to a more subjective review of literature in a field, versus a meta-analysis. The methodology of analyzing and then aggregating might also explain some significant part of the positive outcomes, comparing to aggregating and then analyzing data. As is the case with meta-analysis, the quality and validity of conclusions based on aggregated data are dependent on the quality of included studies (garbage in – garbage out principle). Unfortunately, the included case studies are almost never well-controlled experimental single subject design studies, and conclusions from a large number of uncontrolled case studies will not automatically result in a robust and methodologically sound outcome. Another important problem is that some leading psychoanalytically-oriented researchers assume that formal methods of case formulation are not useful. Rather, they embrace the narrative tradition to case formulation (Messer and Wolitzky, 2007) which focuses on the ways in which people make and use stories to interpret the world, and to understand themselves and others, and reflects a constructionist movement in psychotherapy that is distinct from narrow cognitive models of personality and psychopathology. Messer and Wolitzky (2007) assume that formal methods would not directly inform us about the treatment utility or validity of case formulation as conducted in everyday clinical practice. If this impression of psychoanalytic case formulation is correct, then the generalization of research findings from research to practice would be highly problematic. Consequently, the preliminary positive findings from
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
63
psychoanalytic and psychodynamic studies of case formulation need to be interpreted with caution. In this chapter, I will focus on behavioral, cognitive, and cognitive-behavioral case formulation. Much of what can be said about case formulation within these approaches is also relevant to case formulation in dialectic behavior therapy.
DIFFICULTIES AND BARRIERS IN RESEARCH ON CASE FORMULATION Case formulation is a tool to organize information about variables that cause and maintain a person’s emotional, cognitive, behavioral, interpersonal and, to some extent, physiological problems. As such, it comprises a large number of observations, primarily made and reported by the patients, as well as several tentative hypotheses made by the therapist, preferably in collaboration with the patient. Flexibility and the search for information that helps to develop a comprehensive and useful explanation of current problems is a defining characteristic of case formulation. Too strict formalization of the procedures of arriving at a case formulation in accordance with the underlying theory might be at odds with the function of case formulation. Case formulation is subject to constant modification with emergence of new information from observations, tests and experiments, therefore, it is very difficult to study and evaluate. Nevertheless, empirical research is needed and the utility and effectiveness of case formulation has been indirectly investigated in a few psychotherapy studies.
RESEARCH ON CASE FORMULATION: EMPIRICAL FINDINGS Case Formulation in the Treatment of Depression One of the earliest studies of case formulation, in terms of matching function/ difficulties and intervention, focused on the treatment of depression (McKnight et al., 1984). Using an alternating treatments design combined with multiple baseline design, McKnight et al. (1984) treated nine depressed women, three of whom had primarily problems in social skills, three with problems with irrational cognitions, and three with both type of problems. Before the intervention, all the patients were diagnosed with major affective disorders with a score of 70 or above on the depression subscale of the Minnesota Multiphasic Personality Inventory (MMPI: Hathaway, 1946), and Research Diagnostic Criteria based on the Schedule for Affective Disorders and Schizophrenia Interview (Endicott and Spitzer, 1978). All the patients received four sessions of social skills training and four sessions of cognitive therapy, but in an alternating fashion with multiple baselines. Those with primarily deficits in social skills benefited more from social skills training, both in terms of skills acquisition and depression, while those with problems mainly related to irrational cognitions improved more from cognitive therapy. Those who, at assessment, had both social skills deficit and irrational cognitions showed
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
64
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
equivalent improvement in depression with the two types of treatment. Furthermore, social skills training in this group resulted in greater improvement in their social skills and cognitive therapy led to better outcome in terms of decrement in their irrational cognitions (McKnight et al., 1984). The specific effect of social skills training for the three patients with deficits and difficulties in social skills can be seen in Figure 3.1. Social skills training produced better results (lower scores, p = .003), according to the Lubin Depression Adjective Checklist (Lubin, 1967), in 11 out of 12 measurement occasions compared to cognitive treatment. As can be seen in Figure 3.1, the first session for patients 1 and 3 was social skills training. All three patients received social skills training at the last session because
28
BASELINE SUBJECT 1
TREATMENT
COG
14
TOTAL SCORE : DACL
SS 0 28
SUBJECT 2 COG
14
0 28
SS
SUBJECT 3
COG 14
SS
0 1
2
3
4
1
2
3
4
5
6
7
8
Weeks Figure 3.1 The effect of social skills training (SS) and cognitive treatment (COG) in a multiple baseline design with alternating treatments across three patients with difficulties in social skills on a weekly measure of Lubin Depression Adjective Checklist. Source: Figure reproduced by kind permission from Elsevier through Copyright Clearance Center: Lic. No. 2457660664660.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
65
of ethical reasons, as their primary area of difficulty was considered to be a lack of social skills. Social skills reduced depression more than cognitive therapy across the study and the multiple baseline shows that the first session of social skills training produced an immediate amelioration in depression for all three patients (McKnight et al., 1984). Similarly, depressed patients with primary irrational cognition, but with adequate social skills, significantly improved in cognition and depression after receiving cognitive therapy compared to the unrelated treatment of social skills. The mixed group is also an interesting case. They were initially assessed as suffering from difficulties in both social skills and irrational cognitions. Their level of depression decreased significantly and they improved in social skills and cognitions after receiving the corresponding treatments. McKnight and colleagues concluded that “The results are consistent with Craighead’s (1980) and Liberman’s (1981) assertion that treatment effectiveness for depression would be maximized when treatment is related to target behaviors assessed to be problematic for the individual” (McKnight et al., 1984, p. 332). This is an important study, despite its small sample size and drastic exclusion rates (9 participants remained out of a group of 72 women who responded to community notices about the study, of which 45 passed the screening stage one of the study). It clearly illustrates not only the power of matching, but also the treatment utility of assessment (Hayes, Nelson and Jarrett, 1987). The outcome of this study is in line with previous research within other areas (e.g., social anxiety), showing that assessment and treatment of specific target behaviors is important in ameliorating psychiatric disorders or psychological problems. This is a very important point that I will return to when I describe the use of case conceptualization and functional analysis within developmental disabilities. Functional analysis, which can be considered one way of conceptualizing the case at hand, is the foundation of behavioral treatment in developmental disabilities. Similarly, Trower et al. (1987) showed, in a randomized trial of treatment for social failure, that patients whose assessment indicated a lack of social skills improved more if they received social skills training compared to systematic desensitization. On the other hand, those who were socially anxious according to the assessment protocol improved equally after receiving social skills training or systematic desensitization. Some methodological strengths and specific features of the study by McKnight et al. (1984) are worth mentioning. Interestingly, the therapists were not very experienced psychotherapists. They were graduate students finishing their second and third year of graduate training in clinical psychology, with experience in assessing and treating depressed patients. It should be mentioned, though, that they had received intensive training in the treatment strategies utilized in the study. A specific strength of the design of the study was that the therapists were never informed which patients were in which group. Neither were they informed that they all had three types of patient, one from each group (social skills deficit, irrational cognitions, and a mixed group). A good and relatively unambiguous outcome, in the context of a blind assessment and treatment procedure and moderately skilled therapists, might be interpreted in this study as a sign of the power of treatment matching. Furthermore, the inclusion of a mixed group of patients, showing both social skills problems and irrational cognitions, helped to mitigate the major shortcoming of the design (i.e., treatment interference and carry-over effects due
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
66
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
to alternating treatments design). The outcome of the control group showed that the two treatments were equally effective. Consequently, it can be concluded with a greater level of confidence that the good outcome was related to the matching procedure (i.e., the relation between assessment and treatment). If one of the treatments had been clearly more superior to the other one, it would have been difficult to determine whether the outcome for each group of patients was due to a more potent treatment or a more important link between assessment and treatment (McKnight et al., 1984). A more recent study investigating the effect of matching in the field of depression showed similar promising results (Strauman et al., 2006). In this fairly small randomized study (N = 45), a new therapy called Self-System Therapy (SST) that focused on improving socialization toward pursuing promotion goals was compared to cognitive therapy. Patients were recruited via adverts in local television news broadcasts and in newspapers, at a university psychiatric clinic and women’s health clinics, and referral from a university counseling center (Strauman et al., 2006). A total of 65 out of 110 individuals who made an initial inquiry about the study were excluded due to various exclusion criteria (e.g., history of mania or psychosis, concurrent treatment of depression, current substance abuse, etc.). The majority of the patients met the DSM-V criteria for Major Depressive or Dysthymic Disorder. Treatment was provided by nine therapists with different levels of experience, from faculty clinical psychologist to predoctoral interns, although they all had at least three years of training in cognitive therapy (average years of therapy experience was 7.0, SD = 4.1). Seven of the therapists met patients in both treatment conditions. All of the therapists received training in self-system therapy before and during the study, and all received weekly clinical supervision for their cognitive therapy cases and attended weekly group meetings for discussion, case presentation and videotape or audiotape review for the SST condition. The treatments were comparable concerning length and common features such as agenda setting, homework assignment, and other elements. Although there were no overall differences in the efficacy of the treatments, patients with deficits in socialization toward promotion goals that were assigned to self-system therapy showed significantly more improvement compared to those with the same deficit assigned to cognitive therapy. As expected and in line with the hypothesis, those assigned to self-system therapy reported a greater reduction in dysphoric responses to promotion goal priming compared to cognitive therapy patients. The results above and the equal overall efficacy of cognitive therapy and SST suggest that the two treatments may have different mechanisms of action. Strauman et al. (2006) suggested that these findings illustrate the potential of a translational approach to the treatment matching in depression. Persons and colleagues conducted two other interesting, but uncontrolled, studies of case-conceptualization in the treatment of depressed patients (Persons, Bostrom and Bertagnolli, 1999; Persons et al., 2006). In their 1999 study, the outcome of 45 depressed patients in the first author’s private practice was comparable to the outcome of depressed patients in randomized controlled trials (RCTs) of depression. The patients (mean age of 32 years, SD = 9.9, and 60% females) had an initial score of 14 or greater on Beck Depression Inventory (BDI; Beck et al., 1979) and 69% met the criteria for Major Depression at intake. In addition, 16% of the private
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
67
practice patients had major medical problems and 13 patients had Panic Disorder. The outcome was compared to two RCTs of depression, one by Murphy et al. (1984), and another important RCT by Elkin et al. (1989), called Treatment of Depression Collaborative Research Program (TDCRP). Compared to the patients in Persons’ private practice, all of the patients in the RCTs met the criteria for Major Depression, but those with significant medical difficulties and concurrent panic disorder were excluded. Furthermore, those in the private practice received an average of 34.8 sessions of treatment compared to a maximum of 20 sessions in RCTs. Finally, 22% of private practice patients received adjunct treatment compared to none of the research patients. One of the conditions in the RCT by Murphy and colleagues (1984) consisted of cognitive behavior therapy (CBT) plus pharmacotherapy. It allowed direct comparison of patients receiving both CBT and medication in private practice and the RCTs, as well as comparing those who only received CBT in both private practice and RCTs. The same protocol was used in treating patients for depression in private practice and RCTs, but some modifications and adjustments were made due to multiple comorbidities that were not addressed in the RCT protocol. Furthermore, in private practice the interventions were provided more flexibly and the treatment was open-ended. To make the adjustments more systematic, the therapist used individualized case formulation (Persons, 1989). The post-treatment BDI scores were almost identical in both the private practice and RCT samples, but those in the RCTs had significantly higher pretreatment BDI score, due to inclusion criteria (i.e., having Major Depression). The proportion of patients showing clinically significant change was comparable between the groups, with somewhat fewer private practice patients (17%) achieving reliable change and moving into the distribution of asymptomatic individuals than patients in the TDCRP (28%). The authors concluded that the outcome for the patients treated in private practice using case formulation was comparable to those treated in RCTs of depression, even though those in the private practice sample had multiple comorbidities and, in some cases, were more severely ill (e.g., 20% had a current substance abuse) than the sample in the RCTs, who had higher mean score on depression, but significantly fewer comorbidities. Furthermore, the private practice sample had higher education, received on average twice as many treatment sessions as those in the RCTs, and 46.7% of them had a concurrent treatment, compared to none in the RCTs. Nevertheless, the outcome suggests that the result of RCTs of CBT for depression do generalize to private practice and that case conceptualization seems to have an important function in translating and adapting the treatment to the need of patients with substantial complexity. In the second study, Persons et al. (2006) treated 58 anxious depressed patients, aged 19–75, with case-formulation driven CBT, once again showing comparable outcomes to those in RCTs of depression in terms of effect sizes and clinically significant change. The outcomes of a meta-analysis by Weston and Morrison (2001) and the review paper by Barlow and Lehman (1996) were used as benchmark. Of the 58 patients with a mean age of 36.4 years (SD = 12.7), 35 (60%) were female. The mean years of education in the sample were 14.3 years (SD = 7.2). All the patients received a primary diagnosis of nonpsychotic mood disorder, an anxiety disorder, or both, and all reported both symptoms of depression and anxiety on self-report
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
68
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
measures. Concerning comorbidities, the majority had several mood or anxiety disorders and 22 had axis-II disorders. BDI (Beck et al., 1979) and Burn’s Anxiety Inventory (BAI; Burns, 1998) were used as outcome measures. All the patients received case formulation-driven individual CBT. In addition, 66% received adjunct pharmacotherapy, 19% received family therapy or 12-step group treatment, or occasionally insight-oriented psychotherapy. Using a benchmarking strategy, the authors showed that the outcome of the formulation-driven treatment was within the range of those in the large and well-conducted RCTs, although the patients had slightly lower levels of BDI at the start of the treatment. In a subanalysis, the authors selected patients with a pretreatment score of at least 20 or higher on BDI, as was the case in RCTs. Once again, the outcome of the formulation-guided treatment was comparable to those of patients in RCTs of depression. The authors concluded that anxious-depressed patients with multiple comorbidities might benefit from empirically supported treatment, guided by a cognitive-behavioral case formulation and weekly outcome monitoring. This study was, in a way, a replication of the previous study by Persons and colleagues (1999), but the outcome was compared to meta-analysis of RCTs as well as a review study providing suitable data (i.e., with patients reporting both depression and anxiety).
Case Formulation in the Treatment of Marital Problems An early study of the effect of case formulation and individualization on treatment outcome was done by Jacobson and colleagues (1989) in the context of marital therapy. Thirty distressed married couples were randomized into two behavioral treatments, one of which was highly standardized (20 sessions), while the other, although based on the same principles, allowed more individualization and flexibility in choice and delivery of methods and strategies as well as the length of treatment (M = 22 sessions, range 8 to 53). The couples ranged in age from 27 to 64 years with a mean age of 41 years – and they had been married for an average of 11.4 years (range 0.5 to 38 years). Mean years of education for husbands were 15.8 years and the corresponding figure for wives was 14.8 years. Couples were excluded if significant physical violence had occurred in the relationship within the year prior to seeking treatment. The main outcome measures were Marital Satisfaction Inventory (Snyder, 1979, 1981), Dyadic Adjustment Scale (Spanier, 1976), and a direct observational measure of communication (Hahlweg et al., 1984). There were no differences between the conditions at post-treatment with regard to global marital satisfaction, the couples’ reports of functioning in specific areas, and direct observational measures of communication; however, couples in the standardized treatment were more likely to deteriorate and relapse at six-months follow-up compared to those in the individual treatment, who were more likely to have maintained their treatment gains. In other words, although couples were better off at six-months follow-up than when they were enrolled in treatment, they were not as happy six months after the end of the therapy as they were at post-treatment. By dividing the couples into subgroups that improved, recovered, or were unchanged or deteriorated, the investigators found that 45% of couples in the standardized treatment deteriorated from post-treatment to six-months follow-up compared to
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
69
14% in the flexible individualized condition. Despite its small sample and lack of power in some of the analyses, this study is methodologically an important one with high internal and ecological validity due to randomization, similar mean number of therapy sessions, valid and informative outcome measures, modular nature of the treatment, good manipulation check, and diversity in sample as well as the experience of the therapists. Given its design and sample size, this study should be viewed as a conservative test of the effect of individualization and, as the authors mention themselves, the difference in durability favoring the individualized condition is an impressive outcome. An interesting issue worth mentioning is the later development of Behavioral Couple Therapy (BCT) that was used in this study and its evolvement into Integrative Couple Therapy; which, to some extent, was due to work on individualization and case formulation within BCT. This led to further understanding of the variables maintaining couple distress and the importance of using acceptance strategies along with methods and strategies to promote change.
Case Formulation in the Treatment of Phobic Anxiety The effect of tailoring the treatment to the specific characteristics and needs of patients in line with functional analysis of cases was done in a study of phobic anxiety (78% with agoraphobia, and 88% reporting panic attacks) with a very clever design and a fairly large sample for such a study. Schulte and colleagues (1992) compared standardized manual-based treatment (exposure in vivo), individualized treatment based on an analysis done by the therapist, and a yoked control condition (where each patient receives the therapy individualized for another patient in the individualized condition) by randomizing 120 patients with phobias in one of these conditions. The mean age of the patients (36% males, 64% females) was 39.4 years and the mean duration of phobic anxiety symptoms was, on average, more than 17 years. The vast majority of patients had seen a psychotherapist or a medical doctor for their phobia and 66.7% had been prescribed medication, but only about a third were taking medication at the beginning of treatment, with no significant differences between the three groups regarding these variables. A strength of the study was the use of several outcome measures, most of them showing good psychometric properties. Patients responded to Anxiety Reaction Questionnaire (Kast, 1980), Situation Appraisal System (Ullrich and Ullrich de Munyck, 1976), Fear Thermometer (Walk, 1956), and Fear Survey Schedule (Wolpe and Lange, 1964). In addition, the treatment effect was evaluated in relation to achieving the specific treatment goals of an individual patient using the Goal Attainment Scaling (Kiresuk and Sherman, 1968) as agreed on during the work on the individualization of the treatment. Finally, patients and therapists made independent global ratings at the end of the treatment regarding symptoms and the ability to cope with anxiety. Data showed that the standardized manual-based treatment was superior to the other two treatments. Interestingly, the pattern of results was the same for experienced and inexperienced therapists, with no differences among patients with different phobias and panic disorders of various severities. Altogether, this study
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
70
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
found no evidence to support individualization in the treatment of phobic anxiety as operationalized according to the DSM-III (American Psychiatric Association, 1980) by means of a standardized interview. In comparison to other psychiatric disorders, the symptoms of phobia and panic disorder are rather homogeneous and limited, so the results could be a special feature of the treatment of patients with panic and phobic disorders. Accordingly, after investigating the scope of the results, the authors concluded that the superiority of the standardized treatment, which seemed to be related to superiority of exposure, might be true only for specific phobias. It is easy to agree with this statement given the nature of the sample of patients in the study, as only those with a diagnosis of any phobia according to the DSM-III and without any other diagnoses were included in the sample. Of those who requested therapy, 30% were excluded because of additional diagnoses, mainly depression. Such a selection, which is one of the major weaknesses of many otherwise well-done and older RCTs, will probably result in the exclusion of the most complex and treatment-refractory cases. Given the power of exposure therapy, there is probably not much room for improvements from individualization. In addition, the therapists in the study were 28 young psychologists with substantially variable levels of experience, but a median of nine patients treated. This means that more than half of the therapists had very little practical experience. Future studies need to investigate the role of the therapist’s experience in the effect of case conceptualization for outcome. Despite findings showing no evidence of the superiority of individualization, Schulte argued in his later writings (Schulte, 1996) that treatment individualization and standardization should be viewed as complementary strategies. This is in accord with the current view of standardization of treatment in the era of evidencebased and empirically supported treatments and their application in both research and clinical practice, where a level of individualization is ubiquitous.
Case Formulation in the Treatment of Obsessive-Compulsive Disorder (OCD) The effect of individualization was investigated in a small study focusing on obsessive-compulsive disorder (OCD) (Emmelkamp, Bouman and Blaauw, 1994). A total of 22 patients (18–65 years, M = 29.4 years, SD=6.7) were randomly assigned to either tailor-made CBT or standardized exposure in vivo therapy. The patients were referred to the study by mental health agencies and general practitioners. Those with a severe enough DSM-III-R diagnosis of OCD that warranted intensive treatment with no previous cognitive or behavioral treatment, no psychosis, and not suicidal at enrollment were included. Twenty-five patients met these criteria and two were excluded due to ethical reasons. One patient did not accept the treatment offer, leaving 22 patients (17 females and 5 males) who were equally divided across the groups. Patients reported an average duration of 7.5 years of complaints (SD = 7.5), with a range from 1 to 17 years. It is also important to keep in mind that the patients were instructed not to take any antidepressants or anxiety-reducing drugs during the trial. Outcome was measured using Maudsley
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
71
Obsessive-Compulsive Inventory (Rachman and Hodgson, 1980), the Rational Behaviour Inventory (Shorkey and Whiteman, 1977), the Self-Rating Depression Scale (Zung, 1965), and the Revised Symptom Checklist (Derogatis et al., 1973). Contrary to expectations, the authors found no differences between the two conditions. Treatment in both conditions led to significant improvement in OCD symptoms, but the individualized condition was not more effective than the standard treatment. Interestingly, two experienced behavior therapists interviewed the patients, made the functional behavior assessments, and devised a treatment plan, which can be considered a strength of the study; however, therapy was conducted by advanced clinical psychology students who had followed an extensive course in behavior therapy and received further training in CBT and cognitive therapy with OCD patients before treating the patients in this study. Despite twice-weekly group supervision sessions, the distinction between those who did the individualization and those who conducted the treatment, and lack of experience among the therapists, as well as low power of the study may partly account for the null results of the study. Another interpretation, in line with what Emmelkamp et al. (1994) suggested, is the power of self-controlled exposure in vivo. This is similar to findings in the study by Schulte et al. (1992), in reducing not only the specific anxiety symptoms, but also making generalized improvements on depressed mood, social anxiety, and psychopathology in broad terms, although, the exposure is highly focused on the specific anxiety symptoms. Once again, such a powerful treatment may not leave much space for further improvement and, therefore, has not been beaten by an individualized treatment.
Case Formulation in the Treatment of Bulimia Nervosa Ghaderi (2006) investigated the role of individualization in the treatment of bulimia nervosa (BN) in a randomized study where 48 patients of the included 50 completed the treatment in either standardized manual-based CBT for BN (Fairburn, Marcus and Wilson, 1993; Wilson, Fairburn and Agras, 1997) or an individualized condition, where the treatment was guided by logical functional analysis (Ghaderi, 2007). Participants were recruited through adverts or referrals from psychiatric clinics and the student health center. The initial telephone interview of applicants resulted in excluding 75 of the 146. In most cases, this exclusion was due to applicants using psychotropic medication, not fulfilling the binge and purge severity criteria for bulimia nervosa according to the DSM-IV (American Psychiatric Association, 2000), and current psychosocial treatments for eating disorders. In addition, current anorexia nervosa, BMI below 18 years, younger than 18 years of age, pregnancy, psychotic and bipolar disorders, severe substance abuse, as well as obstacles for committing to the study (e.g., plans for moving out of town) were also exclusion criteria. No participants were excluded due to the presence of personality disorders or other comorbid conditions such as anxiety or mood disorders, with the exception of very severe and recurring depression (e.g., BDI above 45) or acute risk of suicide. The mean age of the patients was 27.2 years (SD = 7.8, range: 19–51 years). The mean age of onset of eating disorders was 18 (SD = 4.4, range: 10–34 years), and the mean duration of eating disorders was 9.2 years (SD = 6.3,
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
72
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
range: 1–26 years). Most patients had normal weight, with a mean BMI of 25.0 (SD = 5.1, range: 18.3–40.5). Patients were diagnosed using the Eating Disorders Examination (Cooper and Fairburn, 1987). In addition to the interview, the Eating Disorders Examination Questionnaire (Fairburn and Beglin, 1994), Body Shape Questionnaire (Cooper et al., 1987), Rosenberg Self Esteem Scale (RSE) (Rosenberg, 1979), Perceived Social Support (Procidano and Heller, 1983), Beck Depression Inventory (Beck and Steer, 1993), and the Hopkins symptom checklist-90 (Lipman, Covi and Shapiro, 1979) were used to measure specific and general outcome of the treatment. This is probably one of the more demanding comparisons of a standardized and individualized condition because the individualized condition was initially based on the same protocol as the standardized condition. Methods, techniques and strategies were added to or withdrawn from the protocol for each patient in the individualized condition, depending on the conclusions drawn from the functional analysis for that patient. The number of sessions and the length of sessions were, however, the same in both conditions. At the end of the 20-session long treatment, both conditions led to significant improvements in both primary and secondary outcome variables. There were no significant differences between the conditions concerning the efficacy of treatment, with the exception of abstinence from objective bulimic episodes as measured by the number of weeks of abstinance, eating concerns, and dissatisfaction with body shape, all favoring the broader, individualized condition. The outcome was maintained at follow-up, six months after the end of the treatment. There was a statistically significant group difference regarding improvement as observed by the diagnoses at follow-up (χ 2 (1, 50) = 3.9, p =.048), but no significant differences in the percentage of improvement in objective bulimic episodes from pre- to post-treatment or follow-up. Ten patients (20%) did not respond to the treatment. Notably, a majority of nonresponders (80%) were in the manual-based condition. This study provided some preliminary support for the superiority of higher level of individualization in terms of the response to treatment and relapses; however, the magnitude of effects was moderate and independent replications with blind assessment procedures and a larger sample sizes are needed before more clear-cut conclusions can be drawn.
Case Formulation in the Treatment of Psychosis Another area where the effect of individualization has been investigated is psychosis. Using a series of single-case designs in two experiments, the effect of case conceptualization on process measures (e.g., therapeutic alliance) rated by clients and therapists and the outcome of the therapy (strength of delusional and selfevaluative beliefs) was examined (Chadwick, Williams and Mackenzie, 2003). Patients in the first experiment focusing on the therapeutic alliance were referred for CBT for psychosis because of distressing positive symptoms with at least six months duration and a poor response to drugs. The 13 patients who were enrolled in the study were diagnosed with paranoid schizophrenia, schizo-affective disorder or delusional disorder (7 men and 6 women, with a mean age of 31.5 years). The therapists were two highly skilled, accredited CBT therapists and their trainee
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
73
clinical psychologists. Outcome was measured using the Helping Alliance Questionnaire (Alexander and Luborsky, 1986) and Hospital Anxiety and Depression Scale (HADS: Zigmond and Snaith, 1983). In the second experiment focusing on the outcome, four people with distressing auditory hallucinations, and secondary paranoid delusions were enrolled in a multiple-baseline design. The main outcome measures were Psychotic Symptom Rating Scales (Haddock et al., 1999) and HADS (Zigmond and Snaith, 1983), as well as visual analog scales measuring belief conviction. Following at least five baseline data points, the three main active ingredients in cognitive therapy for psychosis (i.e., case formulation, cognitive restructuring of negative self-evaluative beliefs, and cognitive restructuring of secondary delusions) were introduced in a sequence of at least 4 sessions, each in a multiple base-line design. Case conceptualization had no significant impact on the participants’ rating of alliance or outcome variables, although descriptive data suggested some benefits for certain participants. Interestingly, therapists’ perception of the therapeutic relationship did improve following the session in which the case conceptualization had been shared with the client.
Case Formulation in the Treatment of Childhood Behavioral Disorders The potential advantages of individualization in relation to outcome have also been investigated in a study of children with behavioral disorders. In a blocked randomized controlled study, 28 boys and 7 girls (age range: 7–13 years) received either individualized skills training, nonindividualized skills training or no treatment at all during the course of the 12 weeks long treatment (Schneider and Byrne, 1987). Recruitment base for the study was a center for children with behavior disorders. All the children, except two whose parents refused consent, were included. About half of the children were in residential care and the remainders were in a day program. The primary diagnoses were Attention Deficit, Conduct or Anxiety disorder. Behavioral observations were conducted as an outcome measure and four trained psychology graduate students, who received weekly videotape supervision sessions, delivered the treatments. Role-plays showed that children in the individualized condition achieved their behavioral training objectives at an average of 87.5% compared to 76.8% in the nonindividualized condition (a nonsignificant difference) and 42.3% in the waitlist condition (significantly lower than in both treatment conditions). Interestingly, analyses of observational data during recess play showed significant findings for cooperation, but not aggression. Gain scores for the individualized condition were significantly higher than the other groups with no significant difference among them. The waitlist group received the individualized training after post-treatment data were collected. According to role-play, 94.1% of the training objectives had been mastered to criterion after completed individualized training. This study provided some limited support for the individualization of social skills training. However, given some methodological shortcomings and limitation,
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
74
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
such as small sample size and skewed observations scores, its outcome should be interpreted carefully. It provides, however, a good basis for further research on the impact of individualization of social skills training.
Case Formulation in the Management of Mental Retardation Case formulation in terms of ideographic functional analysis, or so-called chain analysis, has been used extensively with good outcome. Some of these areas include the treatment and management of children, adolescents and adults with mental retardation (Didden et al., 2006), self-injurious behavior (Pelios et al., 1999), aberrant behavior (Mace, Lalli and Lalli, 1991), and borderline personality disorder (e.g., Linehan et al., 2006). Functional analysis and individualization are extensively used, and indirectly evaluated, within the treatment of disabilities and intellectual disabilities. The analysis is an integrated part of the treatment and its treatment utility is axiomatically assumed to be high given the total outcome of treatment. This is the area where the power of case conceptualization is most evident. Similarly, when it comes to functional analysis of problem behaviors, almost every single treatment study based on behavior analysis (see any issue of Journal of Applied Behavior Analysis) can be seen as a source of evidence for the power of the analysis and its effect on outcome. Hanley, Iwata and McCord (2003) found over 500 graphed individual data sets that were published depicting the results of functional analysis through the year 2000 with a very large portion of differentiated functional analysis showing behavioral maintenance. In other words, a large number of studies over decades, using single subject design, have shown that specific variables can be measured and manipulated with precision and large effects.
CONCLUSIONS FROM EMPIRICAL RESEARCH AND IMPLICATIONS FOR PRACTICE Currently, it is difficult to make any firm assertions regarding the relationship between individual case conceptualization in its diverse forms and outcome, given the mix of negative, null, and positive findings given the limited number of studies, lack of power, methodological shortcomings, and extremely few studies with long-term follow-up data. The evidence for the treatment utility of case conceptualization with intellectual and other developmental disabilities is the strongest, but this field might be viewed as significantly different from the field of psychotherapy (talking therapies) in general. Within this area, more specific research is needed to experimentally investigate the reliability and validity of the analysis and its specific effect on outcome with focus on manipulation of the analysis itself. Current use of mediational analysis can be used as an example here. In many studies, the mediators are never manipulated – they are only measured. Through regression analysis, we see if the mediator changes and explain some of the effect. This is, however, not a solid evidence of mediation. Mediators should be
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
75
manipulated experimentally as well as maximally increased in some cases and not at all affected, or even decreased, in other cases randomly: Do we see mediation then? Well, in that case we can suggest mediation in the causal way we otherwise used to. These studies are difficult to do, but possible! This line of research means independent identification of target behaviors and schedules of reinforcement given the problem at hand. It also means assessing the interrater reliability in the assessment, as well as experimental test of alternative hypothesis and their effect on outcome. On the other hand, it might be argued that the reliability of case conceptualization, often in terms of an ideographic functional analysis, is shown by the high reliability of measuring the target behavior and reliably reinforcing or extinguishing the target behaviors in the course of treatment. This is usually done in an experimental design, usually some form of single subject design, or quasi-experimental design, generally in terms of time-series analysis. The validity of analysis is evident from outcome, which in most cases is good, with medium to large effects (for a review see Hanley et al., 2003). To summarize, the positive outcome of treatments based on applied behavior analysis (Austin and Carr, 2010) is an indirect evidence of the treatment utility of functional analysis and ideographic case formulation. Direct evaluations of the role of analysis and individual adaptations of the treatment based on the analysis should be the next step in research, by manipulating the analysis and testing alternative hypotheses. To summarize, it can be concluded that reliability, validity and treatment utility of case conceptualization, matching, or individualization certainly merit further research. The positive findings are encouraging. Moreover, given the fact that some level of adjustment and individualization occurs in all psychotherapy, including manual-driven treatments, we need to study the effect of case conceptualization and individualization more systematically. Beyond the field of applied behavior analysis and its use to manage problems within mental retardation and disabilities, as Persons and Tompkins (2007) suggested, the strongest evidence for the treatment utility of cognitive-behavioral case formulation still comes from its reliance on nomothetic formulations as a general model for idiographic formulation. This strong evidence also comes from the ideographic data collected by the therapist during the course of the therapy to test the clinical utility of the formulation for each patient. Furthermore, as summarized by Eells (2007), most proponents of individual case conceptualization refer to the efficacy studies of their school of psychotherapy in which their method of case conceptualization is used as an indirect evidence of the treatment utility of case formulation (Eells, 2007). Given different sources of cognitive bias among clinicians and researchers during assessment and treatment, there is a risk for circularity in this line of reasoning and rigorous experimental investigations of the effect of case conceptualization is needed. We do not reduce our patients or clients to diagnoses or symptoms. We try to understand them and help them as individuals in a context, which is usually complex. We try to apply our methods and techniques with sensitivity and in a manner that maximizes the effect, within a frame of good therapeutic alliance and relationship. Some level of individualization is inevitable in both assessment and treatment. Consequently, we need to learn more about its effect and how to do it best.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
76
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
FUTURE RESEARCH In addition to studies on reliability, more experimental studies are needed where the case conceptualization is manipulated. Future studies should include experienced therapists, be designed to have good power a priori, and make conservative test of the effect of case conceptualization on outcome. Testing the significance of the therapist’s experience, training, and supervision for the reliability, accuracy, quality and validity of case conceptualization is an important research agenda. Future research must show what kind of training and supervision leads to good case conceptualization. Kuyken et al. (2005) showed that both reliability and quality of case formulation were associated with levels of experience and accreditation status. In their study, the quality of only 44% of the clinicians were rated as at least good enough. More research is needed to help us learn how we can improve the quality and, thereby, the reliability and validity of case conceptualization. The effect of case conceptualization on outcome should be investigated through multivariate analysis where several important variables, such as client characteristics, severity of problem, level of comorbidity, therapist experience, etc., can be taken into account. Furthermore, as suggested by Kuyken, Padesky and Dudley (2008), we need to develop a psychometrically robust measure of the quality of case conceptualizations for further research. With the exception of functional analysis within developmental disabilities and some problem behaviors, the current state of evidence for case conceptualization is mixed. In addition to using the best practices (see, for example, Hanley, Iwata and McCord, 2003), future research should also consider incorporating recent findings on individual differences related to genetics and neuro-cognitions (for a good example of recent findings, see Furmak et al., 2008). This taps into the old notions ¨ Jeremalm and Johansson, 1981) and the of individual response patterns (e.g., Ost, effect of different methods in the treatment that is in line with ideographic case conceptualization. Another important research area is to consider the effect of case conceptualization on complex cased with multiple comorbidities against manual-based treatments as well as investigating the outcome among treatment refractory cases where manual-based treatments fail. Individualization might be the answer for increasing the efficacy of manual-based treatments, but more research is needed. However, as a note on methodology, it is recommended that there be more extensive use of single subject designs before running large scale RCTs.
REFERENCES Alexander, L.B. and Luborsky, L. (1986) The Penn Helping Alliance Scales. In L.S. Greenberg, and W. Pinsof (eds), The Psychotherapeutic Process: A Research Handbook. London: Guildford Press. American Psychiatric Association (1980) Diagnostic and Statistical Manual of Mental Disorders (3rd edn). Washington, D.C.: American Psychiatric Press, Inc. American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental Disorders (4th edn, text revision). Arlington, VA: American Psychiatric Press, Inc. Austin, J. and Carr, J.E. (2010) Handbook of Applied Behavior Analysis. Reno: Context Press.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
77
Barlow, D.H. and Lehman, C.L. (1996) Advances in psychosocial treatment of anxiety disorders. Archives of General Psychiatry, 53, 727–35. Beck, A.T., Rush, J.A., Shaw, B.F. and Emery, G. (1979) Cognitive Therapy for Depression. New York, NY: Guilford Press. Beck, A.T. and Steer, R.A. (1993) Manual for the Beck Depression Inventory. San Antonio, TX: The Psychological Corporation. Burns, D.D. (1998) Therapist Toolkit. Unpublished manuscript, Los Altos, CA. Chadwick, P., Williams, C. and Mackenzie, J. (2003) Impact of case formulation in cognitive behaviour therapy for psychosis. Behaviour Research and Therapy, 41(6), 671–80. Cooper, Z. and Fairburn, C.G. (1987) The eating disorders examination: A semi-structured interview for the assessment of the specific psychopathology of eating disorders. International Journal of Eating Disorders, 6, 1–8. Craighead, W.E. (1980) Away from a unitary model of depression. Behavior Therapy, 11, 122–8. Crits-Cristoph, P., Cooper, A. and Luborsky, L. (1988) The accuracy of therapists’ interpretations and the outcome of dynamic psychotherapy. Journal of Consulting and Clinical Psychology, 56(4), 490–5. Derogatis, L.R., Lipman, R.S. and Covi, L. (1973) SCL-90: An outpatient psychiatric rating scale. Psychopharmacological Bulletin, 9, 13–28. Didden, R., Korzilius, H., van Oorsouw, W. and Sturmey, P. (2006) Behavioral treatment of challenging behaviors in individuals with mild mental retardation: meta-analysis of single-subject research. American Journal of Mental Retardation, 111(4), 290–8. Eells, T.D. (2007) Handbook of Psychotherapy Case Formulation. New York: Guilford Press. Elkin, I., Shea, M.T., Watkins, J.T. et al. (1989) National Institute of Mental Health Treatment of Depression Collaborative Research Program: General effectiveness of treatments. Archives of General Psychiatry, 46, 971–82. Emmelkamp, P.M.G., Bouman, E.K. and Blaauw, E. (1994) Individualized versus standardized therapy: A comparative evaluation with obsessive-compulsive patients. Clinical Psychology and Psychotherapy, 1(2), 95–100. Endicott, J. and Spitzer, R.L. (1978) A diagnostic interview: The schedule for affective disorders and schizophrenia. Archives of General Psychiatry, 35, 837–844. Fairburn, C.G. and Beglin, S.J. (1994) Assessment of eating disorders: interview or self-report questionnaire? International Journal of Eating Disorders, 16, 363–70. Fairburn, C.G., Marcus, M.D. and Wilson, G.T. (1993) Cognitive behaviour therapy for binge eating and bulimia nervosa: A comprehensive treatment manual. In C.G. Fairburn and G.T. Wilson (eds), Binge Eating: Nature, Assessment and Treatment (pp. 361–404). New York: Guilford Press. Furmark, T., Appel, L., Henningsson, S., et al. (2008) A link between serotonin-related gene polymorphisms, amygdala activity, and placebo-induced relief from social anxiety. Journal of Neuroscience, 28(49),13066–74. Ghaderi, A. (2006) Does individualization matter? A randomized trial of standardized (focused) versus individualized (broad) cognitive behavior therapy for bulimia nervosa. Behaviour Research and Therapy, 44(2), 273–88. Ghaderi, A. (2007) Logical functional analysis in the assessment and treatment of eating disorders. Clinical Psychologist, 11, 1–12. Haddock, G., McCarron, J., Tarrier, N. and Faragher, E.B. (1999) Scales to measure dimensions of hallucinations and delusions. The psychotic symptom rating scales (PSYRATS). Psychological Medicine, 29(4), 879–89. Hahlweg, K., Relsner, L., Kohli, G. et al. (1984) Development and validity of a new system to analyze interpersonal communication (Kategorien- system fur partnerschaftliche interaktaon). In K. Hahlweg and N. Jacobson (eds), Marital Interaction: Analysis and Modification. New York: Guilford Press. Hanley, G.P. Iwata, B.A. and McCord, B.E. (2003) Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36, 147–85. Hathaway, S.R. (1946) The Minnesota Multiphasic Personality Inventory. Modern Hospital, 66, 65–7.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
78
May 28, 2011
9:36
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Hayes, S.C., Nelson, R.O. and Jarrett, R.B. (1987) The treatment utility of assessment. A functional approach to evaluating assessment quality. American Psychologist, 42(11), 963–74. Jacobson, N.S., Schmaling, K.B., Holtzworth-Munroe, A. et al. (1989) Research-structured vs clinically flexible versions of social learning-based marital therapy. Behaviour Research and Therapy, 27(2), 173–80. Kast, A. (1980) Entwicklung and empirische Uberpriifung eines Angstreaktionsfragebogens [Development and empirical examination of an anxiety-reaction-questionnaire]. Unpublished examination paper. Ruhr-Universit¨at, Bochum, Germany. Kiresuk, T.I. and Sherman, R.E. (1968) Goal attainment scaling: A general method for evaluating comprehensive community mental health programs. Community Mental Health Journal, 4, 443–53. Kuyken, W., Fothergill C D., Musa, M. and Chadwick, P. (2005) The reliability and quality of cognitive case formulation. Behaviour Research and Therapy, 34, 1187–1201. Kuyken, W., Padesky, C.A. and Dudley, R. (2008) Collaborative Case Conceptualization: Relieve Distress and Build Resilience. New York: Guilford. Liberman, R.P. (1981) A model for individualizing treatment. In L.P. Rehm (ed.), Behavior Therapy for Depression (pp. 231–53). New York: Academic. Linehan, M.M., Comtois, K.A., Murray, et al. (2006) Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs therapy by experts for suicidal behaviors and borderline personality disorder. Archives of General Psychiatry, 63(7), 757–66. Lipman, R.S., Covi, L. and Shapiro, A.K. (1979) The Hopkins Symptom Checklist (HSCL). Journal of Affective Disorders, 1, 9–24. Lubin, B. (1967) Depression Adjective Check lists: Manual. San Diego, CA: Educational and Industrial Testing Services. Mace, F.C., Lalli, J.S. and Lalli, E.P. (1991) Functional analysis and treatment of aberrant behavior. Research in Developmental Disabilities, 12(2), 155–80. McKnight, E.L., Nelson, R.O., Hayes, S.C. and Jarrett, R.B. (1984) Importance of treating individually assessed response classes on the amelioration of depression. Behavior Therapy, 15, 315–35. Messer, S.B. and Wolitzky, D.L. (2007) The psychoanalytic approach to case formulation. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation. New York: Guilford Press. Murphy, G.E., Simons, A.D., Wetzel, R .D. and Lustman, P.L. (1984). Cognitive therapy and pharmacotherapy, singly and together in the treatment of depression. Archives of General Psychiatry, 41, 33–41. ¨ Ost, L.G., Jeremalm, A. and Jansson, J. (1981) Individual response patterns and the effects of different behavioral methods in the treatment of social phobia. Behaviour Research and Therapy, 19, 1–16. Pelios, L., Morren, J., Tesch, D. and Axelrod, S. (1999) The impact of functional analysis methodology on treatment choice for self-injurious and aggressive behavior. Journal of Applied Behavior Analysis, 32(2), 185–95. Persons, J.B. (1989) Cognitive Therapy in Practice: A Case Formulation Approach. New York: W.W. Norton and Co. Persons, J.B., Bostrom, A. and Bertagnolli, A. (1999) Results of randomized controlled trials of cognitive therapy for depression generalize to private practice. Cognitive Therapy and Research, 23(5), 535–48. Persons, J.B., Roberts, N.A., Zalecki, C.A. and Brechwald, W.A. (2006) Naturalistic outcome of case formulation-driven cognitive-behavior therapy for anxious depressed outpatients. Behaviour Research and Therapy, 44(7), 1041–51. Persons, J.B. and Tompkins, M.A. (2007) Cognitive-behavioral case formulation. In T.D. Eells (ed.), Handbook of Psychotherapy Case Formulation (pp. 290–316). New York: Guilford Press. Procidano, M.E. and Heller, K. (1983) Measures of perceived social support from friends and from family: Three validation studies. American Journal Community Psychology, 11, 1–24. Rachman, S. and Hodgson, R.J. (1980) Obssessions and Compulsion. New York: Prentice-Hall. Rosenberg, M. (1979) Conceiving the Self. New York: Basic Books, Inc.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
DOES CASE FORMULATION MAKE A DIFFERENCE TO TREATMENT OUTCOME?
79
Schneider, B.H. and Byrne, B.M. (1987) Individualizing social skills training for behaviordisordered children. Journal of Consulting and Clinical Psychology, 55(3), 444–5. Schulte, D. (1996) Tailor-made and standardized therapy: complementary tasks in behavior therapy. A contrarian view. Journal of Behavior Therapy and Experimental Psychiatry, 27(2), 119–26. ¨ Schulte, D., Kunzel, R., Pepping, G. and Schulte-Bahrenberg, T. (1992) Tailor-made versus standardized therapy of phobic patients. Advances in Behaviour Research and Therapy, 14(2), 67–92. Shorkey, C.T. and Whiteman, V.L. (1977) Development of the Rational Behaviour Inventory: Initial validity and reliability. Educational ad Psychological Measurement, 37, 527–34. Snyder, D.K. (1979). Marital Satisfaction Inventory. Los Angeles, CA: Western Psychological Services. Snyder, D.K. and Wills, R.M. (1989) Behavioral versus insight-oriented marital therapy: Effects on individual and interspousal functioning. Journal of Consulting and Clinical Psychology, 57, 39–46. Spanier, G.B. (1976) Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38, 15–28. Strauman, T.J., Vieth, A.Z., Merrill, K.A. et al. (2006) Self-system therapy as an intervention for self-regulatory dysfunction in depression: a randomized comparison with cognitive therapy. Journal of Consulting and Clinical Psychology, 74(2), 367–76. Sturmey, P. (2008) Behavioral Case Formulation and Intervention: A Functional Analytic Approach. West Sussex: Wiley-Blackwell. Trower, P., Yardley, K., Bryant, B.M. and Shaw, P. (1987) The treatment of social failure: A comparison of anxiety-reduction and social skills acquisition procedures on two social problems. Behavior Modification, 2, 41–60. Ullrich, R. and Ullrich de Muynck, R. (1976) Das Situationsbewertungssystem SB-EMandS. Anleitung fiir den Therapeuten, Teil Ill [The situation-appraisal-system SB-EMI-S. Manual for the therapist, part III]. Mtinchen: Pfeiffer. Walk, R.D. (1956) Self-rating of fear in fear evoking situations. Journal of Abnormal Psychology, 52, 171–8. Waller, G. (2009) Evidence-based treatment and therapist drift. Behaviour Research and Therapy, 47, 119–27. Westen, D. and Morrison, K. (2001) A multidimensional meta-analysis of treatments for depression, panic, and generalized anxiety disorder: An empirical examination of the status of empirically supported therapies. Journal of Consulting and Clinical Psychology, 69, 875–99. Wilson, G.T., Fairburn, C.G. and Agras, W.S. (1997) Cognitive behavioral therapy for bulimia nervosa. In D.M. Garner and P.E. Garfinkel (eds), Handbook of Treatment for Eating Disorders (pp. 67–93). New York: Guilford Press. Wolpe, J. and Lange, P.J. (1964) A fear survey schedule for use in behavior therapy. Behavior Research and Therapy, 2, 27–30. Zigmond, A.S. and Snaith, R.P. (1983) The Hospital Anxiety and Depression Scale. Acta Psychiatrica Scandinavica, 67, 361–70. Zung, W.E.K. (1965) A self rating depression scale. Archives of General Psychiatry, 12, 63–70.
P1: OTA/XYZ P2: ABC JWST073-03 JWST073-McMurran
May 28, 2011
9:36
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
PART II
VIOLENCE
May 26, 2011
1:30
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
Chapter 4
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS: THE STRUCTURED PROFESSIONAL JUDGMENT APPROACH1 STEPHEN D. HART Simon Fraser University, Canada and University of Bergen, Norway
CAROLINE LOGAN Greater Manchester West Mental Health NHS Foundation Trust and University of Manchester, UK
Formulation of violence risk has been a neglected topic in forensic mental health. This neglect may have been due to the proliferation of actuarial risk assessment procedures in the 1990s and 2000s, procedures that (as we will discuss) do not provide a basis for thinking sensibly about formulation. Thankfully, it appears the field is shifting away from numbers-based actuarial thinking about violence risk and toward evidenced-based clinical thinking – or, as Hart (2008) put it, from formula to formulation. Our goal in writing this chapter is to help stimulate discussion and improve practice with respect to the formulation of violence risk. The first part of the chapter discusses the practice of violence risk assessment. We define evidence-based risk violence assessment and describe two major approaches, focusing on a group of discretionary procedures known collectively as structured professional judgment (SPJ) guidelines. In the second part, we discuss formulation in (forensic) mental health and then review approaches to formulation of violence risk, including the 1 The views expressed herein are those of the authors and do not necessarily reflect those of their employers or any other organization with which they are affiliated. The authors contributed equally to the preparation of this chapter. Address correspondence to Stephen D. Hart, Department of Psychology, Simon Fraser University, 8888 University Drive, Burnaby, British Columbia, Canada, V5A 1S6. E-mail:
[email protected]
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
84
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
SPJ approach. The third and final part presents a case study that illustrates formulation of violence risk using SPJ guidelines.
PART 1: THE PRACTICE OF VIOLENCE RISK ASSESSMENT Violence risk assessment and management are critical elements of the delivery of forensic mental health services by professionals from disciplines including psychology, psychiatry, nursing, social work, and occupational therapy (Hart, 2001; Doyle and Dolan, 2002). Forensic mental health professionals have various legal and ethical duties to prevent violence. One set of duties, owed to the general public, involves protecting them from violence perpetrated by people treated in community settings. A second set of duties, owed to health care staff and service users, involves protecting them from violence perpetrated by people treated or detained in institutional settings. Finally, a third set of duties, owed to people who receive services, is to help them avoid the self-harmful consequences of perpetrating violence in either community or institutional settings. There have been many advances in the field of violence risk assessment over the past two decades or so (Hart, 2006; Monahan, 2006). Epidemiological research has revealed that violence is not the rare phenomenon it was once believed to be, but rather a pervasive social problem that takes many distinct forms. Also, epidemiological and clinical research has identified important risk factors for various forms of violence. These risk factors are characteristics of people and the social and physical environments in which they live that appear to play a causal role in violence. The advances in risk assessment were not limited to research. The growing research literature provided a solid foundation for the development of evidencebased risk assessment procedures. These procedures are designed for use in forensic mental health settings, as well as in general mental health, corrections, law enforcement, and other settings. They structure the way that evaluators gather, weigh, and combine information so that their risk assessments reflect current views of best practice. No longer are evaluators forced to rely solely on personal experience, intuition, or instinct when conducting violence risk assessments. Evaluators now have access to a host of risk assessment procedures, each designed for use in specific settings, with specific populations, and for specific forms of violence.
What is Evidence-Based Violence Risk Assessment? Evidence-based means an action or decision was guided by, based on, or made after reviewing relevant information in the form of observation, research, statistics, or well-validated theory (Sackett and Rosenberg, 1995). The concept, now used widely in many disciplines, was popularized in medicine, where the classic definition is “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients” (Sackett et al., 1996, p. 71). With respect to the assessment of risk for violence, there is a large evidence base. For example, considerable attention has been devoted to the identification of (putative) risk factors for violence. Thousands of studies have been conducted around the world by researchers from various disciplines, and there have been
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
85
several excellent summaries of the research literature in recent years (e.g., Otto and Douglas, 2010). Other research has examined the efficacy of various forms of treatment or the impact of moderators, such as age, on violence. Causal theories of violence also have been developed. Yet the evidence base is still insufficient to identify exactly what are critical risk factors, how they relate to each other, or the causal roles they play with respect to violence. Evidence-based assessment of violence risk, then, may be defined as the process of gathering information about people in a way that is consistent with and guided by the best available scientific and professional knowledge to (a) understand their potential for engaging in violence against others in the future and (b) determine what should be done to prevent this violence from occurring (Hart, 2009). But even those who accept this definition in principle may have widely divergent views concerning how to apply it in practice. In abstract or general terms, the concept of evidence-based decision making is widely recognized and accepted across disciplines. But the devil is in the details. There is considerable disagreement concerning even a basic definition of evidence-based decision making (e.g., Justice, 2008; Tanenbaum, 2005; Timmermans and Mauck, 2005). Disagreement concerning its principles and practice is even more pronounced (Miles, Polychronis and Grey, 2006), especially with respect to cumulating disparate and contradictory research findings, using research based on individuals to make systems-level decisions (and vice versa), and incorporating values and ethics into the decision-making process (e.g., Atkins, Slegel and Slutsky, 2005; Borry, Schotsmans and Dierickx, 2006; Kemm, 2006; Steinberg and Luce, 2005). Opinions about evidence-based practice with respect to assessment of violence risk generally fall into two camps. On the one hand are people who hold broad views, a group that may be characterized as latitudinarian (Hart, 2003a). They consider evidence-based decision making to be a guiding philosophy, core value, or aspirational standard. They believe ‘evidence-based’ describes the general process underlying a decision, not just the specific procedures use to make the decision. They emphasize that the evidence base itself is always inadequate, flawed, or incomplete, and decision makers must always use their judgment or discretion to fill in the gaps. On the other hand are people who hold narrow views, who may be characterized as orthodox (Hart, 2003a). They consider decision making to be evidence-based only when the specific procedures used are directly derived from or supported (i.e., confirmed or validated) by empirical research. They emphasize the frailties and inadequacies of human cognition, and so attempt to find ways to minimize reliance on judgment or discretion. The debate between those with broad versus narrow views of evidence-based assessment of violence risk is active and intense in the field. Unless one appreciates the profound difference between latitudinarian and orthodox views, it is difficult to appreciate how the search for evidence-based risk assessment has led professionals in such different directions, as we will discuss below.
Approaches to Evidence-Based Violence Risk Assessment There are two basic evidence-based approaches to reach opinions about violence risk: discretionary and nondiscretionary (e.g., Meehl, 1954/1996; Monahan,
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
86
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
1981/1995, 2006). These terms refer to how information is weighted and combined to reach a final decision, regardless of the information that is considered and how it was collected (Meehl, 1954/1996). The hallmark of the discretionary approach – also referred to as the clinical or judgmental approach – is that the evaluator exercises substantial professional judgment in the decision-making process, including which information to consider and how to gather it, as well as how to weight and combine it. It is sometimes characterized as “informal, subjective, [and] impressionistic” (Grove and Meehl, 1996). In contrast, the hallmark of the nondiscretionary approach – also referred to as the actuarial or statistical approach – is that, based on the information available to them, evaluators make an ultimate decision according to fixed and explicit rules, developed a priori (Meehl, 1954/1996). It is also generally the case that the nondiscretionary approach relies on empirical research to determine which information to consider, how to gather it, and how to weight and combine it. It is very specific in focus, designed to predict certain outcomes over certain timeframes in certain populations. The nondiscretionary approach is sometimes characterized as “mechanical” and “algorithmic” (Grove and Meehl, 1996). It is worth noting here that, to some extent, all violence risk assessment is discretionary. Even when using “nondiscretionary” risk assessment procedures, professionals still must use their judgment to determine which procedure to use and how to administer, score, and interpret the procedure. Furthermore, the use of nondiscretionary procedures does not constitute a comprehensive evaluation, and evaluators typically are urged or required to consider the relevance of information in addition to that included in the procedure, such as aging or physical health or condition. Finally, nondiscretionary risk assessment procedures provide limited, if any, guidance concerning the identification of potentially effective risk management strategies.
Discretionary Procedures The discretionary approach comprises three major procedures. The first is unstructured professional judgment, also referred to as unaided clinical judgment. This is decision-making in the complete absence of structure, a process that is fairly and accurately characterized as “intuitive” or “experiential.” Historically, it is the most commonly used procedure for assessing violence risk and therefore is very familiar to mental health professionals, as well as to courts and tribunals. But there is no way to determine the extent to which unstructured professional judgment was informed, guided, or structured by the scientific and professional literature. Also, there is no body of scientific evidence supporting the usefulness of unstructured professional judgments of violence risk. Despite its strengths, then, unstructured professional judgment cannot be considered evidence-based according to either narrow or broad definitions and we will not consider it further in this chapter. The second is referred to anamnestic risk assessment (e.g., Otto, 2000). Anamnesis – from the Greek for “remembrance” or “recollection” – refers to the process of history-taking in medicine. Anamnestic risk assessment imposes a limited degree of structure on the evaluation process to the extent that the evaluator must,
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
87
at a minimum, identify the personal and situational factors associated with the offender’s past violence. The assumption here is that a series of events and circumstances, a kind of behavioral chain, led up to the offender’s act of violence. The evaluator’s task therefore is to understand the links in this chain and suggest ways in which the chain could be broken. But anamnestic risk assessment cannot be considered evidence-based for the same reasons that unstructured professional judgment cannot, and so we will not consider it further. The third procedure is structured professional judgment, also known as guided clinical judgment. Here, decision-making is assisted by guidelines that have been developed to reflect the “state of the discipline” with respect to scientific knowledge and professional practice. Such guidelines – sometimes referred to as clinical guidelines, consensus guidelines, or clinical practice parameters – are quite common in medicine, although used less frequently in psychiatric, psychological, or correctional assessment (Kapp and Mossman, 1996). Structured professional judgment procedures are evidence-based according to both broad and narrow definitions. First, the guidelines are directly informed, guided, and structured by the scientific and professional literature. Second, there is a substantial and growing body of scientific evidence supporting the view that assessments of violence risk made using structured professional judgment guidelines are both reliable and valid (e.g., Otto and Douglas, 2010). A good example of a structured professional judgment procedure for assessing risk for sexual violence is the Risk for Sexual Violence Protocol or RSVP (Hart et al., 2003; see also Hart and Boer, 2010). The RSVP guidelines are presented in the form of a reference book or manual. They are intended for use in a wide range of civil and criminal justice contexts to assist forward planning in individual cases by guiding decisions about risk assessment and management. The target population comprises people aged 18 and older who have a known or suspected history of sexual violence. Administration of the RSVP involves 6 steps. In Step 1, evaluators gather case information, guided by a number of recommendations presented in the manual. In Step 2, evaluators code the presence of 22 individual risk factors from 5 domains, as well as any additional case-specific risk factors. Presence ratings are made for two timeframes: more than one year prior to the evaluation (“past”) and within the year prior to the evaluation (“recent”). Presence ratings for each timeframe are made using a three-point ordinal scale (Absent, Possibly or partially present, or Present), and may be omitted when there is insufficient information. In Step 3, evaluators determine the relevance of the individual risk factors. “Relevant” risk factors are those the evaluator believes are functionally (i.e., causally) related to the examinee’s perpetration of sexual violence in the future, or are likely to substantially impair the effectiveness of risk management strategies designed to prevent future sexual violence. Relevance ratings are using a three-point ordinal scale (Not relevant, Possibly or partially relevant, or Relevant). In Step 4, evaluators identify and describe the most likely scenarios of future sexual violence. They conjecture about what might happen in the future in light of information about the examinee’s sexual violence history gathered in Step 1, risk factors identified as present and relevant in Steps 2 and 3, and probable living circumstances in the future. These descriptions of “possible futures” or “feared outcomes” are referred to as scenarios, short narratives designed to simplify complex
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
88
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
forecasts in a way that facilitates planning. Evaluators are encouraged to develop multiple scenarios, then “prune” those that are implausible in light of the facts of the case at hand or more general knowledge about sexual violence. The remaining plausible scenarios form the basis for the development of risk management strategies. The procedures used in this step of the RSVP were derived from more general scenario planning methodology, which has been used successfully for many years in other fields to plan under situations of great or unbounded uncertainty. In Step 5, evaluators develop strategies for managing sexual violence risk in light of the relevant risk factors and scenarios of risk. The development of strategies is based on consideration of the sexual violence that might occur under each scenario, as well as relevance of individual risk factors. To ensure the risk management strategies are comprehensive, evaluators are encouraged to consider four general categories: monitoring, treatment, supervision, and victim safety planning. Within each category, users identify specific strategies, and then are encouraged to consider in explicit and specific terms how these strategies should be implemented in the case at hand (i.e., to move from strategies to tactics). The development of good risk management plans in Step 5 depends strongly and directly on the quality of the scenarios developed in Step 4. In Step 6, evaluators document their judgments regarding overall risk in the case. This facilitates clear communication, and is also very important for liability management. Evaluators are encouraged to make judgments concerning case prioritization or overall risk, risk for serious physical harm, any indication of other risks the examinee may pose, any immediate actions taken or required, and critical dates or triggers for case review. Critically, the RSVP does not provide a formula or other algorithm for calculating risk based on the presence of various factors; instead, evaluators must use their discretion to consider, decide, and explain the relevance or meaningfulness of any factors that are present with respect to the risks posed and management of those risks.
Nondiscretionary Procedures There are two major types of nondiscretionary procedures. The first is the actuarial use of psychological tests. Classically, psychological tests are structured samples of behavior designed to measure a personal disposition, that is, an attempt to quantify an individual’s standing on some dimension of psychosocial functioning. On the basis of past research, one can identify cutoff scores on the test that maximize some aspect of predictive accuracy. For example, psychopathy, as measured by the Hare Psychopathy Checklist-Revised (Hare, 2003), may be associated with violence risk in a meaningful way. The actuarial use of psychological tests is not evidence-based according to either broad or narrow definitions. First, reliance on a single test clearly is not informed, guided, and structured by the entirety of the scientific and professional literature, but only by a tiny fraction of that literature. Second, although empirical evidence may support the reliability and validity of individual psychological tests, it also indicates that the same tests are inferior to other available procedures. To be blunt, the actuarial use of psychological tests may be demonstrably better than nothing, but it is also demonstrably worse
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
89
than some alternatives. We will not consider the actuarial use of psychological tests further. The second type of procedure is the use of actuarial risk assessment instruments, also known as actuarial tests, tools, or aids. In contrast to psychological tests, actuarial instruments are designed not to measure anything, but solely to predict the future. Typically, they are high fidelity, optimized to predict a specific outcome in a specific population over a specific period of time. The items in the scale are selected either rationally (on the basis of theory or experience) or empirically (on the basis of their association with the outcome in test construction research). The items are weighted and combined according to some algorithm to yield a decision. In violence risk assessment, the “decision” generally is the estimated likelihood of future violence (e.g., re-arrest for a sexual crime against persons) over some period of time. Items not included in the actuarial instrument may not be taken into account. Actuarial instruments may be considered evidence-based according to the narrow definition, to the extent there is direct empirical evidence supporting their reliability and validity (Quinsey et al., 2006). It is arguable, however, whether they are evidence-based according to the broad definition. The fact that they include only a relatively small set of risk factors and ignore all others means that they are not informed, guided, and structured by the entirety of the scientific and professional literature. A good example of an actuarial instrument for assessing risk for sexual violence is the STATIC-99 (Hanson and Thornton, 1999; see also Anderson and Hanson, 2010). The STATIC-99 is an actuarial instrument developed to assess risk for sexual and violent recidivism in adult males who have been charged with or convicted of a sexually motivated offense. It was created by combining items from two other actuarial instruments. The 10 items in the STATIC-99 were not selected because of their relevance or importance according to the scientific and professional literature, but rather because they significantly discriminated between known groups of recidivists and nonrecidivists in four samples of sex offenders from Canada and the United Kingdom. Administration of the STATIC-99 begins with a review of the offender’s official records; an interview may also be conducted, if the evaluator deems it necessary. Next, the evaluator gives a numerical score between 0 and 3 for each item, following detailed instructions contained in the test manual. The item scores are then summed to yield total scores, ranging from 0 to 12. Finally, evaluators can use information in the test manual to assign the offender to a relative risk category or to estimate the absolute risk (specific probability) he will commit sexual violence over the next 5, 10, or 15 years. Evaluators are advised not to use judgment or discretion to change the STATIC-99 risk estimates in light of factors not included in the test (such as age, physical health, treatment completion, etc.), but rather to administer additional actuarial instruments.
Evaluation of Evidence-Based Violence Risk Assessment Many different criteria can be used to evaluate the adequacy of violence risk assessment procedures (Hart, 2001). They can be divided into three major groups: efficacy, effectiveness, and utility. In this context, efficacy is the consistency and
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
90
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
accuracy with which risk assessment procedures can be used to forecast future violence in controlled research settings. In technical terms, this means there should be some body of research demonstrating that evidence-based risk assessment procedures have acceptable levels of interrater reliability and predictive validity. Demonstrating the efficacy of violence risk assessment procedures may seem a simple matter, but it is not. Evaluating interrater reliability requires multiple evaluators to assess the same group of patients or offenders under a range of conditions – ideally, with samples of evaluators, patients or offenders, and conditions that are sufficiently large and diverse to permit systematic analysis of factors that may moderate interrater reliability. This type of research is extremely resource intensive. Evaluating predictive validity is even more difficult. It is necessary to recruit a cohort of patients or offenders, assess them, follow them up over long periods of time, and then detect violence that occurs in institutional or community settings. The sample must be sufficiently large and the follow-up sufficiently long to yield a base rate of violence amenable to statistical analysis. Ideally, the design permits analysis of potential moderating factors such as demographic characteristics, changes over time in risk factors, and critical life events or interventions that occur during the follow up. In contrast, effectiveness here is the consistency and accuracy of violence risk assessment procedures in field settings. The objective of effectiveness research is to evaluate the extent to which the findings of efficacy research generalize to less controlled conditions. It is, arguably, even more difficult to conduct than efficacy research and typically is conducted only after some evidence of efficacy has been found. Finally, utility refers to the more general usefulness or social validity of violence risk assessment procedures. Utility research addresses issues such as the acceptability of procedures in the eyes of various consumer groups (e.g., service users, service providers, courts, tribunals and review boards, policy makers), the relative costs and benefits of implementing procedures, and so forth. It is usually conducted only after research has supported the effectiveness of a procedure. Evaluative research may be difficult to conduct, but it is clearly not impossible (for summaries, see Otto and Douglas, 2010). With respect to efficacy, the interrater reliability of discretionary (SPJ) and nondiscretionary (actuarial) violence risk assessment procedures has been supported in a large number of studies. An even larger number of studies have established their predictive validity using true prospective, pseudo-prospective (follow-back), and retrospective designs, although most were not able to control for the effect of changes over time in risk factors and life events or interventions during follow-up. In crude terms, the interrater reliability of both types of assessment procedures may be characterized as “good” to “excellent,” with actuarial procedures being slightly more reliable than SPJ procedures on average. Their predictive validity may be characterized “fair” or “moderate,” with no difference between actuarial and SPJ procedures. With respect to effectiveness, there has also been some research supporting the reliability and validity of SPJ and actuarial risk assessment procedures in field settings. The interrater reliability of actuarial procedures appears to be no better than that of SPJ procedures in field settings, with both being “good” on average. The predictive validity of both actuarial and SPJ procedures is “fair” or “moderate” in field settings. Finally, a relatively small number of studies have examined the utility of
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
91
risk assessment procedures, with generally positive results. SPJ procedures tend to be more readily accepted by some stakeholders than are actuarial procedures, in part because they more directly guide decision making with respect to risk management. This last point is crucial for the present chapter. Actuarial risk assessment procedures are prediction-oriented. They conceptualize risk solely in terms of the probability or likelihood of future violence. Their goal is to discriminate between individuals who have a high versus low likelihood of future violence, based on the extent to which they resemble statistical profiles of recidivists versus nonrecidivists from a particular study. The two criteria used to select items for inclusion are the accuracy and efficiency with which they can differentiate between known groups of recidivists and nonrecidivists. The items need not be conceptually or practically relevant – they may be neither causal risk factors (i.e., treatment targets) nor responsivity factors (i.e., characteristics that mediate or moderate potential treatments). SPJ procedures, in contrast, go beyond prediction to focus on prevention. They conceptualize risk in terms of the nature, severity, imminence, frequency, duration, and likelihood of future violence. Their goal is to speculate about the types of violence individuals might plausibly perpetrate, and then to use these speculations or forecasts to develop management plans aimed at preventing violence. The criteria used to select risk factors include scientific, practical, and legal relevance or acceptability. Some commentators have argued that, in certain clinical and legal contexts, risk assessments need only be predictive (e.g., Heilbrun, 1997; Quinsey et al., 2006). With respect, we disagree. In clinical contexts, evaluators need to know what to do with a patient or offender. A probability estimate is of no assistance here. Say the finding of an actuarial risk assessment procedure is that a patient has a “10% [25%, 50% . . .] likelihood of violence”. Is this a 10% chance that, laboring under persecutory delusions, he will take an axe and kill his family within the next week or so? Or a 10% chance sometime within the next 10 years he may do at least one thing to at least one person for some reason that falls within the general definition of violence, and which may or may not have serious consequences? The actuarial instrument cannot discriminate these two interpretations – it is not worried about such distinctions. Similarly, in legal contexts, we can think of no issue that concerns only the probability of violence. Also relevant are the nature of the violence (e.g., imminent risk of serious physical injury, long-term risk of sexual harm), what caused the violence (e.g., mental disorder), whether certain conditions can manage the risks (e.g., total confinement versus something less restrictive), and so forth. To the extent that these latter issues are not addressed by an actuarial risk assessment procedure, it is useless or even misleading, appearing to give a legally relevant answer when it does not. We conclude that of the two approaches to evidence-based violence risk assessment, only the SPJ approach assists the development of risk management plans based on an understanding of the causes of past violence. The actuarial approach is not intended and cannot be used for this purpose. Therefore, we focus in the remainder of our discussion on formulation of violence risk using SPJ assessment procedures, referring to actuarial procedures only for the purposes of comparison.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
92
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
PART 2: FORMULATION OF VIOLENCE RISK To date, four major approaches to formulating violence risk have been developed. In this section, we provide a brief description and evaluation of them.
Offense Paralleling Behavior The Offense Paralleling Behavior, or OPB, approach to formulation was developed Lawrence Jones, Michael Daffern, John Shine, and colleagues (e.g., Jones, 2002, 2004; Daffern et al., 2007). In some respects, OPB may be considered a refinement of the anamnestic approach to focus specifically on violence. Like the anamnestic approach, it is not tied to a specific theory of violence and relies heavily on systematic analysis of past violence. One refinement is that it focuses evaluators to consider the psychological functions of violence – that is, the ways in which violence is potentially rewarding for the individual. Another refinement is that evaluators look for evidence of behavior that parallels past violent offenses in topographical or functional terms (hence, “offense paralleling behavior”). OPB is clinically very useful, as it provides a target for treatment that is easier to assess and treat than is actual violence. Yet another refinement is the focus on building motivation to change as part of the development of case management plans. With respect to strengths, the refinements of OPB increase its clinical utility relative to the anamnestic approach. It provides some additional guidance concerning the identification of risk factors and the development of case management plans. But it has weaknesses similar to that of the anamnestic approach. First, it does not provide a list of theory-derived principles or empirically based risk factors on which to focus. Second, aside from encouraging attention to the functional aspects of violence, it does not facilitate thinking about the causal roles played by risk factors. Third, it does not specify a procedure for speculating about future violence in light of the risk factors identified.
Good Lives Model Tony Ward and his colleagues have pioneered the application of the Good Lives Model, or GLM, to risk assessment of sexual and violent offenders (e.g., Ward, 2002; Whitehead, Ward and Collie, 2007). This was a natural extension of their earlier work on treatment formulation for sexual offenders (e.g., Drake and Ward, 2003; Ward et al., 2000). Like the biopsychosocial model, the GLM is not a theory of violence per se, but rather a way of conceptualizing complex problems that was imported from health care. Within the framework of the GLM, Ward views violence as a problematic means of trying to obtain primary goods – that is, “activities, experiences, or situations that are sought for their own sake and that benefit individuals and increase their sense of fulfillment and happiness” (Whitehead et al., 2007, p. 581), including such things as relatedness, autonomy, knowledge, mastery, play, and physical health. Problems arise when the strategies cannot obtain primary goods. Such problems typically take four forms: neglect of important
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
93
primary goods; use of ineffective strategies to secure goods; conflict of strategies to secure goods; and inability to implement strategies for securing goods. The GLM is based on simple principles, but actually facilitates complex thinking about internal or external factors that prevent people from obtaining primary goods and how to build their strengths and resources so they can obtain primary goods in a prosocial way. With respect to strengths, GLM is clearly articulated. It also has a welcome focus on strengths and resources in the development of case management plans. Its major limitations are its failure to provide guidance concerning risk factors most likely to be related to violence, either in terms of principles derived or a list of empirically derived risk factors. Second, it provides little structure in terms of thinking about the causal roles of risk factors. Third, it does not specify a procedure for speculating about future violence or developing management plans in light of the risk factors identified.
Risk-Needs-Responsivity (RNR) The RNR approach was developed by Don Andrews, James Bonta, and colleagues (e.g., Andrews, Bonta and Hoge, 1990; Gendreau and Ross, 1979). It is not based on a theory of violence per se, but rather a theory of criminal behavior known as the Psychology of Criminal Conduct (PCC) or the General Personality and Cognitive Social Learning (GPCSL) perspective (Andrews and Bonta, 2006). The RNR approach comprises three core principles, derived from research on correctional treatment and interpreted within the broader framework of PCC/GPCSL theory. According to the risk principle, the level of services delivered to offenders should be commensurate with the risks they pose to reoffend. This means offenders at high risk for recidivism should receive more intensive assessment and management, relative to offenders at moderate or low risk. According to the need principle, offender assessment and management should focus on criminogenic needs. This means services for offenders should target causal risk factors for antisocial behavior that have been validated by empirical research. According to the responsivity principle, services should be delivered in ways that maximize their effectiveness. This means two things. First, in general terms, the focus of management programs should be on skills acquisition and enhancement through prosocial modeling, the appropriate use of reinforcement and disapproval, and problem solving, as research suggests this is the most efficient and effective way to change people’s behavior. Second, more specifically, it means that the design and management of programs delivered to offenders should match their individual learning styles, motivations, abilities, and strengths. To assist risk assessment and management using RNR, Andrews, Bonta, and colleagues have developed tools such as the Level of Service-Case Management Inventory, or LS-CMI (Andrews, Bonta and Wormith, 2004). The LS-CMI is intended for use with male and female offenders, aged 16 and older, in institutions or the community. It comprises 11 sections that require evaluators to make a series of ratings based on a semi-structured interview with the offender and a review of relevant records, then document various opinions, recommendations, and decisions. Evaluators recommend, implement, evaluate, and document case
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
94
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
management strategies based on the findings of structured assessments. This is done rationally or logically, rather than using an algorithm or formula. The RNR approach has several noteworthy strengths. First, it is based on a wellarticulated and well-established theory of criminal behavior. Second, it provides clear structure in terms of identifying risk factors and thinking about their causal roles. Third, it provides clear structure in terms of developing case management plans. Finally, and crucially, there is good research support for at least some aspects of the RNR approach (Andrews, Bonta and Wormith, 2006). With respect to potential weaknesses, PCC/GPSCL is a theory of general criminality, not of violence, and therefore the RNR approach may undervalue some risk factors associated with violence but not other forms of criminal behavior. Second, this approach is perhaps too structured. For example, it relies heavily on aggregate statistics derived from research on adult male offenders to identify and weight risk factors. This reduces its usefulness for developing individualized formulations within the population of adult male offenders, as well as its applicability to diverse populations (e.g., patients, females, youth, non-Western cultures). Third, this approach has been criticized for focusing too much on risk-enhancing factors and deficits to the exclusion of risk-reducing or protective factors (i.e., strengths, resources, and buffers). Finally, it does not facilitate speculation about future violence, and so relies on the assumption that it will resemble past violence.
Structured Professional Judgment Approach This approach has grown out of the application of SPJ guidelines. It is outlined in detail in the manuals for the RSVP (Hart et al., 2003) and the Guidelines for Stalking Assessment and Management (SAM; Kropp, Hart and Lyon, 2008). It structures formulation in two different ways. First, it analyzes past violence using a decision theory framework. Second, it speculates about future violence and develops case management plans using scenario planning. The decision theory framework may be considered a version of PCC/GPSCL tailored specifically to violence. It views violence as a choice, that is, purposive behavior intended to achieve one or more goals. The decision may be made quickly, based on bad information, and with little care and attention – that is, it may be a bad decision or a decision made badly – but it is a decision nonetheless. The bottom line is that people who engage in violence think before they act. Violence is clearly considered: Violent people choose who they commit violence against, what kinds of violence they will commit, and when they will commit it. Even the most violent people spend most of their lives not committing violence, so we must ask ourselves, why did they choose to commit specific acts of violence against specific people at specific times, but not other kinds of violence against other people at other times? Within the framework of decision theory, it is assumed that before people engage in violence, they have gone through the following four-step thought process. First, the possibility of acting violently in a given situation entered their conscious awareness, and they entertained this notion rather than dismissing it or pushing it out of their minds. This begs the question, why is it these people thought about violence or entertained the notion, whereas others did not? Second, they evaluated the possible positive consequences of violence and determined that
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
95
it might result in reward or benefit for them. Put simply, they perceived violence might pay off. People are essentially lazy; they do not engage in purposive behavior unless it pays off in some way. This point is critical: If you do not know in what ways violence was perceived as a potential reward or benefit by people, you really do not understand why they committed violence. Third, they evaluated the possible negative consequences of violence and determined the costs were acceptable. Violence always has costs, and sometimes they are enormous. It costs perpetrators time and energy; it may make them feel bad as a result of empathy with their victims, remorse for what they have done, or anxiety about possible punishment for what they have done; it may cause them distress by challenging their self-concepts as good or decent people; and it can cost them in terms of the loss of respect and love of others, the loss of relationships and employment, and even the loss of their freedom. But people who act violently have decided they are willing to accept these costs. Finally, they evaluated the options for committing violence and determined it was feasible. Even if people want to engage in violence, they may have to overcome all kinds of practical barriers. They have to figure out how to locate the victim; how to contact or approach the victim; how to threaten or harm someone; and how to try to get away and avoid detection or capture. According to the decision theory framework, the task of risk assessment is to understand how and why people made decisions to engage in violence, and to understand the various factors that impinged on or influenced their decision making. Risk factors are things that influence decision making. They can play several causal roles. They can motivate, disinhibit, or destabilize decisions. Motivators increase the perceived rewards or benefits of violence. Disinhibitors decrease the perceived costs or negative consequences of violence. Destabilizers generally disturb people’s ability to monitor and control their decision making. Also according to decision theory, the task of risk management is to determine effective strategies to encourage decisions to act prosocially and discourage decisions to act nonviolently. SPJ guidelines such as the RSVP and SAM direct the attention of evaluators to risk factors that are considered generally important according to systematic reviews of the scientific and professional literature, and then to help evaluators understand the relevance of these risk factors in the case at hand using decision theory. Scenario planning is a management strategy used for more than 50 years in such fields as business, health care, and the military (Ringland, 1998; van der Heijden, 1997). According to Chermack and Lynham (2002, p. 366), [s]cenario planning is a process of positing several informed, plausible and imagined alternative future environments in which decisions about the future may be played out, for the purpose of changing current thinking, improving decision making, enhancing human and organization learning and improving performance.
It is most appropriate for situations in which decisions must be based on incomplete knowledge – that is, in situations where various sources of uncertainty make it impossible to predict an outcome using approaches based on frequentist views of probability (e.g., van der Heijden, 1994). In our view, this accurately characterizes the state of affairs with respect to violence risk assessment (Hart, 2003b).
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
96
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
With respect to violence risk assessment, each scenario is a story about violence the person might commit. It is not a prediction about what will happen; rather, it is a general forecast or speculation about what could happen, in light of the evaluator’s general knowledge and experience and the specifics of the case at hand. Although the number of possible scenarios that could be constructed is almost limitless, in any given case only a few distinct scenarios seem reasonable, credible, or internally consistent to evaluators in light of what is known about fact and theory (e.g., Chermack and van der Merwe, 2003). Other scenarios developed may be theoretically possible, but they will be perceived as implausible and subsequently dismissed or “pruned” (e.g., Pomerol, 2001). There are some useful strategies for generating scenarios (e.g., van Notten et al., 2003). To start the process, consider one in which the person commits violence similar to the current or most recent act – what might be called a repeat, “flat trajectory,” linear projection, or point projection scenario. Imagine how this would unfold: What would it take for the person to decide to commit violence of that sort again? Then consider a “better case” or optimistic scenario, one in which the trajectory of violence decreases and the person commits a less serious act. Next consider a “worst case” scenario, also known as a pessimistic or “doom” scenario, one in which the trajectory increases and the person commits a more serious, and perhaps even life-threatening, act of violence. Finally, consider some “twist” or “sideways trajectory” scenarios in which the nature of violence changes or evolves, such as with respect to the manner of victim selection or the type of coercion used. In the SPJ approach, the evaluator constructs as many scenarios as seem plausible based on theory, research, experience, and case facts. In our experience, three to five general scenarios usually are sufficient to capture the range of plausible outcomes. Then, for each scenario, the evaluator develops a detailed description in terms of the nature, severity, imminence, frequency or duration, and likelihood of violence. The scenarios of future violence are then used to develop case management plans that identify general strategies and specific tactics with respect to monitoring, supervision, treatment, and victim safety planning (e.g., Hart, Douglas and Webster, 2001; Hart et al., 2003; Kropp et al., 2008). Like the RNR approach, the SPJ approach to formulation has the strengths of being based on a clearly articulated and well-established theory. It provides structure or guidance for identifying risk factors, conceptualizing their causal roles, speculating about future violence, and developing case management plans. It also has research support (for reviews, see Otto and Douglas, 2010). Unlike RNR, the risk factors it focuses evaluators on are related to violence, rather than criminal behavior more generally. Also, although the SPJ approach is structured, it is more flexible than the RNR approach and thus more readily adapted to new or unusual contexts. Finally, the SPJ approach is the only one of those reviewed in this chapter that focuses as much on the future as it does on the past.
PART 3: AN ILLUSTRATIVE CASE FORMULATION To illustrate our approach to formulation of violence risk using SPJ guidelines, we present below the findings of a sexual violence risk assessment conducted for sentencing purposes (Hart, 2009). Identifying details have been changed to
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
97
protect the privacy of those involved. The findings are followed by a formulation of the case developed using the RSVP. In addition, to highlight the differences between SPJ and actuarial approaches to violence risk assessment, we then present a formulation of the same case using the STATIC-99.
Summary of Findings Mr David Hackett, a 37-year-old man, was the subject of a sexual violence risk assessment following convictions for sexual assault, aggravated assault, and assault causing bodily harm. The risk assessment was requested by the prosecutor and undertaken with the consent and participation of the defendant and his counsel. Mr Hackett was in custody at the time the assessment was completed.
Index Offenses Mr Hackett was convicted of a sexual assault against Ms Easton, a 38-year-old sex trade worker. According to the victim, the offense stemmed from an incident that occurred on 21 July 2006 at about 0245 hrs. Mr Hackett approached her while she was working as a prostitute. Ms Easton and Mr Hackett agreed upon a price for certain sexual acts, Ms Easton entered the SUV that Mr Hackett was driving, and Mr Hackett drove to another location and parked the vehicle. Ms Easton and Mr Hackett engaged in sexual activity, but he was unable to reach orgasm. When Ms Easton complained about the length of time it was taking Mr Hackett to reach orgasm, he became angry, grabbed her by the throat, and threatened her with his fist. Mr Hackett ordered Ms Easton to exit the vehicle, and she complied. Mr Hackett then ordered Ms Easton to bend over the back seat of the car while standing, restrained her by placing a hand on her back, and raped her vaginally and anally. During the rape, Mr Hackett threatened to kill Ms Easton if she resisted. The use of physical coercion did not adversely affect Mr Hackett’s sexual arousal; indeed, if anything, it increased his arousal. The sexual assault continued until another vehicle approached, at which time Ms Easton was able to gather her belongings and leave. She ran to a pay phone and reported the sexual assault to police, providing a description of Mr Hackett and the license plate number of his vehicle. Mr Hackett was also convicted of an aggravated assault and an assault causing bodily harm – both sexually motivated – against Ms Jackson, a 32-year-old sex trade worker. According to the victim, the offenses stemmed from an incident that occurred on 25 August 2006 at about 2245 hrs. Mr Hackett approached her while she was working as a prostitute. Ms Jackson and Mr Hackett agreed upon a price for certain sexual acts, Ms Jackson entered the SUV that Mr Hackett was driving, and Mr Hackett drove to another location and parked the vehicle. Ms Jackson and Mr Hackett engaged in sexual activity, but initially he was unable to achieve an erection and subsequently was unable to reach orgasm. Mr Hackett asked Ms Jackson to exit the vehicle and bend over the car back seat of the car while standing up so that he could have intercourse with her while standing behind her. When Ms Jackson complained about the length of time it was taking Mr Hackett to reach
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
98
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
orgasm, he grabbed her by the throat, threatened to kill her, and continued the sexual activity. The use of physical coercion appeared to increase Mr Hackett’s sexual arousal. Ms Jackson lost consciousness briefly during the assault. When she regained consciousness, she struggled, escaped Mr Hackett’s choke hold, and fell to the ground. Mr Hackett hit her while she screamed and kicked at him, and eventually he got into his vehicle and fled the scene. The incident was reported by witnesses who heard Ms Jackson screaming. Ms Jackson provided a description of Mr Hackett and his vehicle to police when they attended the scene, and a witness provided a description of the vehicle. Ms Jackson suffered significant physical injury as a result of the assault, and medical evidence suggested that the strangulation could have resulted in her death. Mr Hackett was arrested by police based on the information provided by the victims. He was released on bail pending trial. While on bail, Mr Hackett contacted one of the victims, Ms Jackson. He approached her while she was working as a prostitute and attempted to negotiate a price for certain sexual acts. Ms Jackson recognized Mr Hackett immediately, but pretended she did not. She entered Mr Hackett’s vehicle, and he drove to a gas station. When he entered the station, she took his personalized cheques from the glove compartment of the vehicle, left the scene, and contacted the police. Mr Hackett said that he had looked for Ms Jackson on several occasions. He said he wanted to speak with her in an attempt to convince her to drop the charges against him by apologizing or offering her money. No charges were recommended or laid as a result of this incident. The version of events that Mr Hackett gave to police differed dramatically from those of the victims. Regarding the first offense, he denied completely any involvement in the incident on 21 July 2006 involving Ms Easton. Regarding the second offense, he admitted to hiring Ms Jackson to have sex with him, and further admitted that he was unable to achieve an erection. According to Mr Hackett, however, when he exited the vehicle to urinate, he believed he saw Ms Jackson going through the pockets of his jeans. When he confronted her, she became angry and aggressive. He grabbed her briefly by the neck to restrain her, slapped her with an open fist, and she fell to the ground. He then left the scene in his vehicle. Mr Hackett was unable to provide a plausible explanation for the discrepancies between his version of events and that of the victims. He insisted that he was not in possession of his own vehicle at the time of the attack on Ms Easton, having traded vehicles with a friend for about a week. He speculated that perhaps his friend committed the offense. Mr Hackett also speculated that Ms Easton and Ms Jackson had fabricated much of their statements, perhaps in collaboration with each other, in an attempt to profit from criminal injuries compensation claims. During the interview conducted as part of the risk assessment, he reiterated this account even though he readily acknowledged it was completely implausible.
Prior Offenses Mr Hackett had no previous charges or convictions for criminal offenses as a juvenile or an adult. He denied any significant conduct problems in childhood or adolescence, aside from driving on rural roads prior to receiving his driver’s license. He admitted to drinking and driving on occasion as an adult and received
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
99
roadside suspensions from the police for suspicion of driving while intoxicated, but was never arrested for or charged with an offense. Mr Hackett was the subject of a complaint of sexual assault in 1999. His wife at the time, Carrie Hackett, was pregnant with the couple’s first child. Mr Hackett met a woman while drinking in a bar, went for a car ride with her, and engaged in sexual relations with her. According to the complainant, she did not provide consent to the full extent of the sexual activity in which they engaged; she described Mr Hackett’s actions as coercive. Mr Hackett admitted to police that he had sexual relations with the complainant, but denied any coercion. The complainant was ambivalent about participating in the police investigation, and no charges were laid in the matter.
Social History Mr Hackett was raised by his natural parents in a large family. The family members enjoyed close relationships. For example, in childhood and adolescence Mr Hackett and his brothers were coached in boxing by their father; and Mr Hackett worked with his father and brother for a time after graduating from high school. Mr Hackett’s academic adjustment was average. His grades at school were average. His attendance and behavior were also average; he had some minor problems and was suspended on one occasion for fighting, but was never expelled. He was sociable and had many friends. He was active in athletics. Following graduation from high school, Mr Hackett attended college for three years and was qualified as a steel fabricator. He worked steadily in the field since completing college, obtaining additional specialty qualifications. He was employed steadily, receiving unemployment insurance benefits briefly on a single occasion. He was never fired from a job. In addition to typical fabrication jobs, Mr Hackett was involved in building race cars. Mr Hackett started dating in early adolescence. He started dating his future wife, Carrie, when he was 18 years old and she was 17. They moved in together a year later (when he was about 19 years old), and married 8 years later (when he was about 28 years old). The couple had a stable and caring relationship, but grew apart over the years. They separated 31/2 years ago (when he was about 33 years old). There was no physical violence in the relationship, before or after separation. While together, the Hacketts had two children: a son now aged 8 and a daughter now aged 6. Ms Hackett had primary responsibility for raising the children, but Mr Hackett played an active role as father before and after the separation, Mr and Ms Hackett got along well together, and reported no difficulties raising their children together despite the dissolution of their marital relationship. Some three years ago (when he was about 34 years old), Mr Hackett started dating Ms Dawn Hunter. The relationship was stable, with no history of physical violence. Although Mr Hackett reported the relationship is ongoing, Ms Hunter indicated that it ended about six months previously. Mr Hackett had his first sexual relationship at the age of 15. His sexual relations with intimate partners have been typical in terms of the focus and intensity of his appetite. He has had sexual relationships with about 8 women.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
100
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Mr Hackett admitted to trying marijuana on a few occasions in adolescence, but denied any other use of illegal drugs. He admitted to drinking alcohol since the age of 14, and said he drank socially as an adult, primarily on weekends. In the past, he has denied any problems stemming from alcohol abuse; but in the interview conducted as part of the risk assessment, he acknowledged that his drinking had contributed to marital and legal problems. Mr Hackett had no history of significant physical or mental health problems, aside from substance use.
Analysis Using the RSVP Risk Factors Analysis of the case using the RSVP indicated the presence of several basic risk factors, all of which were possibly relevant either to the risks posed by Mr Hackett or the management of those risks. With respect to his history of sexual violence, Mr Hackett had used physical coercion. With respect to psychological functioning, his account of his index offenses and relations with women indicated possible or definite problems with respect to minimization and denial, attitudes that support or condone sexual violence, and problems with self-awareness. With respect to mental disorder, he had definite problems with substance use and, despite a lack of direct evidence, the nature of his offenses raised the possibility of sexual deviation (specifically, a paraphilia such as biastophilia, sometimes referred to as paraphilic rape) and violent ideation. With respect to social adjustment, he had problems with intimate relationships. Finally, with respect to manageability, he had problems with supervision. No additional risk factors were identified.
Formulation Based on his findings, the evaluator developed two competing formulations of Mr Hackett’s sexual violence. According to the first, Mr Hackett’s offenses were motivated primarily by the desire to regain a sense of mastery or agency. Mr Hackett experienced life stresses that made him feel distressed, angry, or insecure, especially with respect to his sexual relationships with women. Disinhibited by somewhat negative attitudes toward women (a hyper-masculine but fragile selfconcept, a sense of male prerogative or sexual entitlement), a tendency to minimize and deny his problems to self and others, and alcohol intoxication, he used physical coercion to “steal sex” from women. According to the second formulation, Mr Hackett’s sexual violence was motivated at least in part by gratification of violent sexual fantasies, possibly the result of an undiagnosed paraphilia, instead of or in addition to the desire to enhance a sense of agency.
Scenarios Based on Mr Hackett’s history of past sexual violence and the formulations described above, the evaluator developed two scenarios of future sexual violence.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
101
The first scenario was a repeat of the index offenses. This scenario, based on the first formulation, was as follows: Mr Hackett experiences life stresses in his relationships with women that make him feel distressed, angry, and insecure; motivated by the desire to regain a sense of mastery and disinhibited by negative attitudes toward women, a tendency to minimize and deny his problems to self and others, or alcohol intoxication, he uses physical coercion to “steal sex” from women. The most likely victims in this scenario are adult females – including intimate partners, casual acquaintances, or prostitutes – who he perceives are denying him the sexual gratification to which he feels entitled. The violence would likely cause serious psychological harm and moderate to severe physical harm. The risk of sexual violence of this sort appears to be chronic (i.e., long-term) rather than acute, although one potential warning sign of imminent risk might be the onset of problems in an intimate relationship. The second scenario was an escalation scenario, based on the second formulation. According to this scenario, Mr Hackett experiences a recurrence or possibly even an escalation of sexually violent fantasies and, motivated primarily by the desire for sexual gratification (and perhaps secondarily by desire for a sense of mastery), acts on them; he may be disinhibited by negative attitudes toward women, a tendency to minimize and deny his problems to self and others, or alcohol intoxication. The likely victims of sexual violence are strangers, such as prostitutes, targeted because it is easy to dehumanize them and also easier to avoid apprehension. The harm to victims could be moderate to severe; indeed, depending on the nature of the paraphilia or the lengths to which he might go to avoid apprehension in the future, it might be life threatening. The risk for this sort of sexual violence also appears to be chronic, rather than acute, although a recurrence or escalation of sexually violent fantasies could be triggered by threats to masculinity or sexuality, and could be accompanied by warning signs such as increased use of alcohol or pornography (especially violent pornography) or preoccupation with prostitutes (e.g., cruising areas frequented by prostitutes).
Management Plans Taking into account the risk factors, formulation, and scenarios, the evaluator recommended several potential case management strategies. The first strategy was participation in assessment and treatment programs. In light of a cluster of risk factors related to Mr Hackett’s minimization and denial, problems with selfawareness, and problems with supervision, it was recommended that treatment should focus initially on enhancing his motivation for positive change by sensitizing him to the seriousness of his problems – past, present, and future – and developing a positive working alliance with a treatment provider. Subsequently, treatment should address his problems with substance use, as well as a cluster of risk factors related to his negative attitudes toward women and intimate relationship problems. A focus of the later stages of treatment should be his negative attitudes related to masculinity. It was recommended that Mr Hackett undergo further assessment of possible paraphilia concurrent with his treatment; if further evidence of paraphilia was uncovered, then it should be added to the list of treatment targets. Give his intellectual functions and basic social skills were intact, and that
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
102
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
he did not present with symptoms of psychopathic personality disorder, the evaluator viewed Mr Hackett as a good candidate for group and individual treatment programs, including those with cognitive-behavioral or emotional-interpersonal theoretical orientations. The second strategy was close monitoring by corrections or mental health professionals with expertise in sexual violence to detect any changes in risk factors that may be triggers (distal or proximal) of sexual violence, such as intimate relationship problems, problems with alcohol use, and possible sexual deviation. The third general strategy was close supervision either in the form of a custodial sentence or an intensive supervision program that included electronic monitoring, conditions to refrain from associating with prostitutes and abstain from drinking alcohol. The evaluator again emphasized the problems developing effective management strategies until the causes of Mr Hackett’s sexual violence were better understood.
Conclusory Opinions The evaluator opined that, despite a limited history of sexual offenses, Mr Hackett posed substantial risks for future sexual violence. The level of risk (also known as case prioritization) was characterized as moderate, meaning that an elevated or above-average level of effort and intervention would be required to prevent future sexual violence. The evaluator was very concerned that the sexual violence was severe in nature, and may have involved sexual arousal after or in reaction to physical coercion. Also, he was concerned that the offenses appeared to occur suddenly and at a relatively late age, without clear motives, warning signs, or triggers.
Analysis Using the STATIC-99 Based on his age, Mr Hackett received a score of 0 on Item 1. Based on the length of his marriage, he received a score of 0 on Item 2. As one of his index offenses was for nonsexual violence, he received a score of 1 on Item 3. As he had no prior convictions, he received a score of 0 on Items 4, 5, and 6. Based on the nature of his index offenses, he received a score of 0 on Item 7, 1 on Item 8, 1 on Item 9, and 0 on Item 10. Summing the item scores, Mr Hackett’s total scores on the STATIC-99 was 3 out of a possible 12. Following is a narrative interpretation of these findings, based on data and recommendations presented in the test manual: The recidivism estimates provided by the STATIC-99 are group estimates based upon reconvictions and were derived from groups of individuals with these characteristics. As such, these estimates do not directly correspond to the recidivism risk of an individual offender. The offender’s risk may be higher or lower than the probabilities estimated in the STATIC-99 depending on other risk factors not measured by this instrument.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
103
Mr Hackett scored a 3 on this risk assessment instrument. Individuals with these characteristics, on average, sexually reoffend at 12% over five years, 14% over ten years, and 19% over 15 years. Based upon the STATIC-99 score, this places Mr Hackett in the Moderate-Low risk category (between the 24th and the 61st percentile) relative to other adult male sex offenders.
Comment This case study clearly illustrates the SPJ approach to formulation. Evidence-based SPJ guidelines focus evaluators on risk factors that are present and relevant, which in turn are used to develop a case formulation. A decision theory framework is used to structure thinking about the past and present, whereas scenario planning is used to structure thinking about the future. The SPJ risk assessment is an attempt to both synthesize and analyze what is known and, just as important, what is not known about the offender. The evaluation is comprehensive in nature, reflecting the breadth of factors deemed critical in the relevant literature. Rather than offer a prediction of what will happen, the evaluator speculates systematically about what the offender might or could do in the future and how to prevent it. The case study also illustrates actuarial risk assessment. The actuarial approach tries to provide an estimate of the probability of future sexual violence based on statistical profiles of known groups of recidivists and nonrecidivists. The evaluation strives for simple and objective findings based on consideration of a small number of specific factors. It is not intended to and, in fact, cannot assist the formulation of violence risk or the development of case management plans except in the most superficial of ways. It is focused on prediction, not prevention.
CONCLUSION In this chapter, we have addressed the critical task of formulation in violence risk assessment and management. We began by discussing the practice of violence risk assessment, defining evidence-based practice and describing the major approaches. Next, we discussed formulation in mental health and reviewed the different ways in which formulation may be undertaken in respect of violence risk assessment, including the SPJ approach. Finally, we presented a case study in which a formulation was proposed linking risk assessment using a set of SPJ guidelines (in this instance, the RSVP) to risk management. Our goal in writing this chapter was to help to stimulate discussion about the critical role of formulation in risk assessment and to examine options for best practice. We hope we have accomplished this, and more. Violence risk assessment is a critical task in mental health and correctional settings, yet its fundamental purpose – the prevention of harm – has for too long been overshadowed by the pursuit of precision in the assessment of probability estimates, arguably impossible in the individual case (Hart et al., 2007), using tools whose application generates little in the way of understanding about harm potential and even less about what to do to prevent it. This chapter has prioritized the pursuit of understanding in its focus on formulation and provided a theoretically sound and empirically based
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
104
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
framework for its achievement. It is only through understanding the risk potential of our clients through the process of formulation that rational and proportionate risk management can be achieved.
REFERENCES Anderson, D. and Hanson, R.K. (2010) Static-99: An actuarial tool to assess risk of sexual and violent recidivism among sexual offenders. In R.K. Otto and K.S. Douglas (eds), Handbook of Violence Risk Assessment Tools (pp. 251–68). Milton Park, UK: Routledge. Andrews, D.A. and Bonta, J. (2006) The Psychology of Criminal Conduct (4th edn). Cincinnati, OH: Anderson. Andrews, D.A., Bonta, J. and Hoge, R.D. (1990) Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, 19–52. Andrews, D.A., Bonta, J. and Wormith, S.J. (2004) Level of Service/Case Management Inventory (LS/CMI). Toronto: Multi-Health Systems Inc. Andrews, D.A., Bonta, J. and Wormith, S.J. (2006) The recent past and near future of risk and/or need assessment. Crime and Delinquency, 52, 7–27. Atkins, D., Slegel, J. and Slutsky, J. (2005) Making policy when the evidence is in dispute. Health Affairs, 24, 102–13. Borry, P., Schotsmans, P. and Dierickx, K. (2006) Evidence-based medicine and its role in ethical decision-making. Journal of Evaluation in Clinical Practice, 12, 306–11. Chermack, T.J. and Lynham, S.A. (2002) Definitions and outcome variables of resource planning. Human Resource Development Review, 1, 366–83. Chermack, T.J. and van der Merwe, L. (2003) The role of constructivist learning in scenario planning. Futures, 35, 445–60. Daffern, M., Jones, L., Howells, K. et al. (2007) Refining the definition of Offence Paralleling Behaviour. Criminal Behaviour and Mental Health, 17, 265–73. Doyle, M. and Dolan, M. (2002) Violence risk assessment: Combining actuarial and clinical information to structure clinical judgements for the formulation and management of risk. Journal of Psychiatric and Mental Health Nursing, 9, 649–57. Drake, C.R. and Ward, T. (2003) Treatment models for sex offenders: A move toward a formulation-based approach. In T. Ward, D.R. Laws and S.M. Hudson (eds), Sexual Deviance: Issues and Controversies (pp. 226–43). Thousand Oaks, CA: Sage. Gendreau, P. and Ross, B. (1979) Effective correctional treatment: Bibliotherapy for cynics. Crime and Delinquency, 25, 463–89. Grove, W.M. and Meehl, P.E. (1996) Comparative efficiency of informal (subjective, impressionistic) and formal (mechanical, algorithmic) prediction procedures: The clinicalstatistical controversy. Psychology, Public Policy, and Law, 2, 293–323. Hanson, R.K. and Thornton, D.M. (1999) Static-99: Improving Actuarial Risk Assessment for Sexual Offenders. Ottawa: Solicitor General of Canada (Corrections Research User Report 1999-02). Hare, R D. (2003) Manual for the Hare Psychopathy Checklist – Revised (2nd edn). Toronto, Canada: Multi Health Systems. Hart, S.D. (2001) Assessing and managing violence risk. In K.S. Douglas, C.D. Webster, S.D. Hart, D. Eaves and J.R.P. Ogloff (eds), HCR-20 Violence Risk Management Companion Guide (pp. 13–25). Burnaby, British Columbia: Mental Health, Law, and Policy Institute, Simon Fraser University, and Department of Mental Health Law and Policy, Florida Mental Health Institute, University of South Florida. Hart, S.D. (2003a) Actuarial risk assessment: Commentary on Berlin et al. Sexual Abuse: A Journal of Research and Treatment, 15, 383–8. Hart, S.D. (2003b) Violence risk assessment: An anchored narrative approach. In M. Vanderhallen, G. Vervaeke, P.J. Van Koppen and J. Goethals (eds), Much Ado about Crime: Chapters on Psychology and Law (pp. 209–30). Brussels: Uitgeverij Politeia NV.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
May 26, 2011
1:30
Printer Name: Yet to Come
FORMULATION OF VIOLENCE RISK USING EVIDENCE-BASED ASSESSMENTS
105
Hart, S.D. (2006, December) Assessing Risk for Violence: Promise and Peril. Public lecture presented at The Assessment and Management of Personality Disordered Offenders, co-sponsored by the Northern Ireland Branch of the British Psychological Society and the Psychological Society of Ireland, Belfast, Northern Ireland. Hart, S.D. (2009) Evidence-based assessment of risk for sexual violence. Chapman Journal of Criminal Justice, 1, 143–65. Hart, S.D. and Boer, D.P. (2010) Structured professional judgment guidelines for sexual violence risk assessment: The Sexual Violence Risk–20 (SVR-20) and Risk for Sexual Violence Protocol (RSVP). In R.K. Otto and K.S. Douglas (eds), Handbook of Violence Risk Assessment Tools (pp. 269–94). Milton Park, UK: Routledge. Hart, S.D., Douglas, K.S. and Webster, C.D. (2001) Risk management using the HCR-20: A general overview focusing on historical factors. In K.S. Douglas, C.D. Webster, S.D. Hart, D. Eaves, and J.R.P. Ogloff (eds), HCR-20 Violence Risk Management Companion Guide (pp. 27-40). Burnaby, British Columbia: Mental Health, Law, and Policy Institute, Simon Fraser University. Hart, S.D., Kropp, P.R., Laws, D.R. et al.. (2003) The Risk for Sexual Violence Protocol (RSVP): Structured Professional Guidelines for Assessing Risk of Sexual Violence. Burnaby, British Columbia: Mental Health, Law, and Policy Institute, Simon Fraser University. Hart, S.D., Michie, C. and Cooke, D.J. (2007) Precision of actuarial risk assessment instruments: Evaluating the “margins of error” of group v. individual predictions of violence. British Journal of Psychiatry, 190, Suppl. 49, S60–S65. Heilbrun, K.S. (1997) Prediction versus management models relevant to risk assessment: The importance of legal decision-making context. Law and Human Behavior, 21, 347–59. Jones, L. (2002) An individual case formulation approach to the assessment of motivation. In M. McMurran (ed.), Motivating Offenders to Change (pp. 31–54). Chichester: John Wiley & Sons, Ltd. Jones, L. (2004) Offence paralleling behavior (OPB) as a framework for assessment and interventions with offenders. In A. Needs and G.J. Towl (eds), Applying Psychology to Forensic Practice (pp. 34–63). Oxford: British Psychological Society Blackwell. Justice, L. (2008) Evidence-based terminology. American Journal of Speech-Language Pathology, 17, 324–5. Kapp, M.B. and Mossman, D. (1996) Measuring decisional competency: Cautions on the construction of a “capacimeter.” Psychology, Public Policy, and Law, 2, 73–95. Kemm, J. (2006) The limitations of ‘evidence-based’ public health. Journal of Evaluation in Clinical Practice, 12, 319–24. Kropp, P.R., Hart, S.D. and Lyon, D. (2008) Guidelines for Stalking Assessment and Management (SAM): User Manual. Vancouver, Canada: ProActive ReSolutions Inc. Meehl, P.E. (1996) Clinical versus Statistical Prediction: A Theoretical Analysis and a Review of the Literature. Northvale, NJ: Jason Aronson. (Original work published in 1954.) Miles, A., Polychronis, A. and Grey, J.E. (2006) The evidence-based health care debate – 2006. Where are we now? Journal of Evaluation in Clinical Practice, 12, 239–47. Monahan, J. (1995) The Clinical Prediction of Violent Behavior. Northvale, NJ: Jason Aronson. (Original work published in 1981.) Monahan, J. (2006) A jurisprudence of risk assessment: Forecasting harm among prisoners, predators, and patients. Virginia Law Review, 92, 391–435. Otto, R.K. (2000) Assessing and managing violence risk in outpatient settings. Journal of Clinical Psychology, 56, 1239–62. Otto, R.K. and Douglas, K.S. (eds) (2010) Handbook of Violence Risk Assessment Tools. Milton Park, UK: Routledge. Pomerol, J.-C. (2001) Scenario development and practical decision making under uncertainty. Decision Support Systems, 31, 197–204. Quinsey, V.L., Harris, G.T., Rice, M.E. and Cormier, C. (2006) Violent Offenders: Appraising and Managing Risk (2nd edn). Washington, DC: American Psychological Association. Ringland, G. (1998) Scenario Planning: Managing for the Future. Chichester: John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-04 JWST073-McMurran
106
May 26, 2011
1:30
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Sackett, D.L. and Rosenberg, W.M. (1995) On the need for evidence-based medicine. Journal of Public Health Medicine, 17, 330–4. Sackett, D.L., Rosenberg, W.M., Muir Gray, J.A. et al. (1996) Evidence based medicine: What it is and what it isn’t – it’s about integrating individual clinical expertise and the best external evidence. British Medical Journal, 312, 71–2. Steinberg, E.P. and Luce, B.R. (2005) Evidence based? Caveat emptor! Health Affairs, 24, 80–92. Tanenbaum, S.J. (2005) Evidence-based practice as mental health policy: Three controversies and a caveat. Health Affairs, 24, 163–73. Timmermans, S. and Mauck, A. (2005) The promises and pitfalls of evidence-based medicine. Health Affairs, 24, 18–28. van der Heijden, K. (1994) Probabilistic planning and scenario planning. In G. Wright and P. Ayton (eds), Subjective Probability (pp. 549–72). Chichester: John Wiley & Sons, Ltd. van der Heijden, K. (1997) Scenarios: The Art of Strategic Conversation. New York: John Wiley & Sons, Inc. van Notten, P.W.F., Rotmans, J., van Asselt, M.B.A. and Rothman, D.S. (2003) An updated scenario typology. Futures, 35, 423–43. Ward, T. (2002) The management of risk and the design of good lives. Australian Psychologist, 37, 172–9. Ward, T., Nathan, P., Drake, C.R. et al. (2000) The role of formulation-based treatment for sexual offenders. Behaviour Change, 17, 251–64. Whitehead, P.R., Ward, T. and Collie, R.M. (2007) Time for a change: Applying the Good Lives Model of rehabilitation to a high-risk violent offender. International Journal of Offender Therapy and Comparative Criminology, 51, 578–98.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
Chapter 5
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION AND VIOLENCE KEVIN HOWELLS Institute of Mental Health and Division of Psychiatry, University of Nottingham and Peaks Unit Rampton Hospital, UK
This chapter is concerned with the formulation of problems of aggression and violence. There are many potential levels of formulation – the problem, the person, or the symptoms of the disorder may be the focus for formulation (Persons, 2008). Thus, the individual may have multiple formulations, including all three levels, and several presenting problems which need to be addressed, for example, aggression, self-harm and substance abuse. For this chapter I shall assume that the major presenting problem is aggression itself, though it is acknowledged that aggression often occurs in the context of other problems, including, for example, Axis I disorders such as schizophrenia or Axis II disorders, particularly personality disorders such as Antisocial and Borderline (see Chapter 12 by Lawrence Jones) or other behavioral disorders. I shall suggest that it is sometimes the case that the clinician is distracted from addressing the core problem in the formulation, particularly in mental health services. Thus, the person’s aggression may be the major reason they have been admitted to a service but the focus becomes treating the personality disorder or schizophrenia. This makes sense only when the formulation and assessments make it clear that such disorders are functionally or causally linked to the problem of aggression. The social and economic costs of aggression and violent crime are considerable (Rand Corporation, 2009). The Rand audit report estimated that in the UK, while violent offenses comprise approximately 10% of offenses, they account for approximately 20% of the costs of crime and that the average costs incurred by one homicide, for example, amount to £1.5 million. The need to formulate a problem is driven, in part, by the significance and impact of the problem under consideration. Apart from economic aspects, by most standards aggression and violence are Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
108
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
major contributors to human suffering and thus require serious efforts directed at explanation (formulation) and, ultimately, prevention; however the formulation of aggression and violence has complexities not always encountered, or perhaps not so readily identified, in the formulation of conventional clinical problems, such as anxiety, depression or psychosis. One complexity is that the analysis of causation of aggression needs to be multilevel and to include contextual as well as individual factors. The Rand Report (2009) concluded that:
As evidence has emerged about how rates of violence and crime vary crossnationally, over time, seasonally and between neighbourhoods, it has become increasingly clear that while some people may have greater propensity to commit crime, other factors also significantly affect the incidence. (p. 20)
Contextual information of this sort has some implications for clinical formulation. Clinical formulation is conventionally construed as an individual-level task, but social conditions and situations need to be acknowledged in the specification of distal factors (historical causes) and of proximal factors to which the individual is exposed at the time of acting violently. Nomothetic theories of aggression invariably point to aggression being a response to social events, typically of an aversive sort (Anderson and Bushman, 2002; Sestir and Bartholomew, 2007). While aversive social events may sometimes be difficult to change in a therapeutic sense, effecting such changes is by no means impossible. This may involve social change or political action, or even assisting the individual to be alert to the importance of such external factors to avoid or prevent aversive situations arising, to focus on managing his or her reactions to them, or to work toward developing skills which may enhance the satisfactions and rewards the environment holds for them. The distal factors associated with aggression, violence and violent crime are well documented and would include being male; young; from a family background involving poor parenting, modeling of aggressive behavior, and poor coping strategies; experiencing harsh or inconsistent discipline; as well as having certain traits and dispositions, such as impulsivity, anger-proneness and low empathy. Many of these factors are static (unchangeable) and, although they require reference in a full formulation, the main focus of formulation clearly needs to be on dynamic variables which are capable of change (Douglas and Skeem, 2005). One clinical task, therefore, when a static factor is identified as an important causal antecedent, for example, exposure to violent abuse in childhood, is to translate this factor into a dynamic proximal one capable of influencing behavior in the here-and-now. How does being abused affect currently, or at the time of the aggressive incident, this person’s propensity for aggression? It may be, for example that the abuse has made the person sensitive to controlling or humiliating behaviors by authority figures, have high trait anger, or hold views as to the normality and acceptability of aggression as a strategy for dealing with problems. Such translation is a familiar task in functional analysis assessments (Sturmey, 1996).
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
109
DEFINITIONAL PROBLEMS Defining aggression and violence is not a straightforward task (Serin, 2004) and the definitions provided in the literature are highly variable. In general, it is useful to view violence as a subset within the broader category of aggression and this is the approach that will be followed for the present chapter. Aggression itself has many definitions ranging from Buss’s (1961) “response that delivers noxious stimuli to another organism”, to Baron and Richardson’s (2004) “any form of behavior directed toward the goal of harming or injuring another living being who is motivated to avoid such treatment”. Violence is a form of aggression where physical injury ensues (Blackburn, 1993). In clinical or forensic settings such as prisons and high security hospitals, it is common for terms such as aggression and violence to be used interchangeably with anger and hostility. The cognitive-behavioral formulator of the clinical problem needs to be very clear about the distinctions between such terms. Anger is an internal emotional response with characteristic psychophysiological and facial accompaniments (Novaco, 2007). Hostility refers to negative evaluations of people or events. Both anger and hostility may influence the occurrence of behavioral aggression but this is not inevitably so. Aggression, violence, anger and hostility may be temporally and situationally specific, referring to particular acts or internal responses, but they may also be dispositional, referring to a general propensity on the part of the person to act in such ways. The formulator may be called on to address and explain either the particular, or the dispositional, or both. Typically, it is the behavioral manifestation of violence that brings the individual into a treatment service, rather than anger or hostility or even aggression per se, though anger, hostility and aggression may sometimes be antecedents for violence. Aggression may be the problem when it takes the form of extreme verbal aggression or threats. For the purposes of this chapter the term aggression will be used to cover both verbal and behavioral problems, including violence.
A PROBLEM FOR WHOM? In defining aggression as the problem requiring formulation, the question arises of “a problem for whom?” While this may sometimes be an issue for other problems, such as depression or anxiety, it is more salient for aggression (Howells and Day, 2003). Behaving aggressively can be quite congruent with the individual’s short and long-term goals and thus they have no reason to seek treatment or to comply with the assessments and treatment prescribed by the therapist. Anger regulation problems, for example, may be associated with active, angry rejection of the notion that they have a problem and attribution of life difficulties to the unfair treatment or malevolence of others, particularly others who may be perceived as having pressured or coerced the person into treatment (Howells and Day, 2003). The fact that aggression may not fit the ideal medical model of someone with a distressing problem seeking therapeutic help for it has important implications for the
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
110
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
therapeutic relationship with the professional charged with conducting an assessment, constructing a formulation, and delivering therapy. In recent years, problems of low motivation for treatment, latterly defined as problems of low treatment readiness (Day et al., 2010), have begun to be analysed theoretically and investigated empirically (Howells et al., 2005; Day et al., 2008; Casey et al., 2007). The implication of such work is that low treatment readiness routinely needs to be assessed, formulated and, where necessary, addressed therapeutically as a precursor to attending to the aggressive behavior itself. Methodologies are beginning to emerge which are suited to this task (Day et al., 2010). In a forensic context, such as the criminal justice system or forensic mental health services, perceived coercion into treatment (Day et al., 2004), rather than self-initiated help-seeking may be the norm. Thus a clear statement of for whom the behavior is a problem is required from the formulator.
WHY FORMULATE THE INDIVIDUAL CASE? There are alternatives to cognitive behavioral individual case formulation in understanding and devising interventions for aggression. Indeed, it is likely to be the case that in forensic services the move toward therapeutic programs, an important part of the What Works movement in offender rehabilitation (Andrews and Bonta, 2006; Hollin and Palmer, 2006), has led to the neglect of the individual’s unique causal profile (Thomas-Peter, 2006). There are two prevailing alternatives to individual formulation: the StructuralDiagnostic and the Offense-Topographic (Daffern and Howells, 2002; Howells, 2010). The former most commonly thrives in forensic mental health services where offenders have been identified as having significant mental disorders, for example, psychotic conditions or personality disorders. Thus, a person’s violent act, for example a homicide, might be largely, or even exclusively, explained in terms of their mental disorder, with the implicit corollary that addressing the mental disorder is the most important, or even only, treatment target. Amongst the many difficulties with the Structural-Diagnostic approach are the deficiencies in the empirical literature on mental disorders as causal factors for violence. The role of mental disorders as antecedents for violence is discussed in more detail below. In recent years there has been increasing interest in the potential role of personality disorders as causal antecedents for violence (McMurran and Howard, 2009), but as yet it remains to be demonstrated that the correlation between personality disorder and violence reflects mechanisms that meet established criteria for causality (Haynes, 1992; Duggan and Howard, 2009; Howells, 2010). The inference of causality is stronger in relation to schizophrenia and violence but also remains problematic (Hodgins, 2008; Elbogen and Johnson, 2009). Thus, while mental disorder variables may be included in a cognitive-behavioral formulation, caution is required. The clinician could attempt to demonstrate that, for this particular individual, variations in, for example, psychotic states parallel variations in aggressive behavior. Such within-individual analyses are difficult to conduct clinically where the personality disorder is likely to show little temporal variation (Howells, 2010).
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
111
The second explanatory approach, the Offense-Topographic, is often implicit rather than explicit and assumes that topographic categories of offenders, in this case violent offenders, share common causal antecedents. Once this assumption is made, it appears sensible to organize treatment programs for such topographic categories of offenders – the “violent offender program” or the “sex offender program”, for example (Howells, 2010). Both the Structural-Diagnostic and the Offense-Topographic approach share the problem of heterogeneity of the categories used in causal explanation, requiring finer and finer subgroups to be identified. Thus, it may be argued, only particular types of schizophrenia (Hodgins, 2008) or particular types of personality disorder (Blackburn, 2009) are likely to be causally linked to violence, and then only when other factors are also present. Equally, ‘violent offenders’ is not a homogenous category (Serin, 2004; McGuire, 2008). It is proposed here that, ultimately, the heterogeneity of mental disorder and offense categories leads inevitably to a need to take an individual formulation approach in which a wide range of causal factors are investigated, in addition to the mental disorders themselves. I shall return later in this review to the heterogeneity of aggressive acts, particularly in relation to anger as an antecedent.
WHAT TYPES OF ANTECEDENT MIGHT FEATURE IN A COGNITIVE-BEHAVIORAL FORMULATION OF AGGRESSION? The range of antecedent factors is potentially very broad and includes: situational and environmental factors and external triggers; cognitive appraisals, beliefs, values and schemas supporting aggression; affective and emotional states (particularly anger); Axis I mental disorders; personality disorders; personality traits; interpersonal problems; self-regulatory deficits; deficits in empathy and other positive emotions; poor coping and problem-solving skills and consequent difficulties in establishing “good lives”; disinhibitors, particularly substance use and misuse; identity and valued goals; and opportunity factors, including weapon and victim availability. This list includes the likely categories requiring investigation but is not exhaustive. It is important to bear in mind that such a list is relevant to the formulation of the problem (acts of aggression), and may be less relevant to the formulation of the case (the person as a whole), or the formulation of symptoms – see discussion of these types of formulation by Persons (2008). It is not possible to review the large literatures relating to each category here. A longer review is provided in Howells et al. (2008). It is clear that these categories overlap considerably and that they focus on intrapersonal factors, though it is acknowledged that many such factors have as distal antecedents powerful cultural, sociological and historical forces (McGuire, 2008). The importance of cultural and historical context in relation to the explanation of aggression and violence has been illustrated in an Australasian setting in Day et al.’s (2008) analysis of high rates of aggression and violence in indigenous peoples. These authors suggested that high violent offending rates could be linked to intergenerational effects, to collective trauma, and to extreme experiences of loss
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
112
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
and family disruption in indigenous peoples. A purely intrapersonal formulation of aggression in such a context is likely to be of limited utility and to distract attention from cultural-level interventions and required remedies. One limitation of lists of relevant factors, such as those in preceding paragraphs, is that the relative neglect of strengths and protective factors as variables that need to feature in any explanation of why people act aggressively or engage in offending behavior. Ward and Maruna (2007) suggested that meeting a number of human needs or ‘primary goods’ is important to leading an offense-free life and thus need to feature in a comprehensive formulation. The model suggests, therefore, that attention should be paid in assessment and in programs to the necessary conditions (e.g., skills, values, opportunities, social supports) that will ensure that the individual’s needs are met in adaptive ways which, in turn, minimizes the likelihood of harm to others. The implied approach to formulation and to intervention is thus one which is much broader and based on a viewpoint that sees people as a “complex interrelated system in which changes in dynamic factors influence those of another” (Thomas-Peter, 2006, p. 35). Systematic approaches to the measurement of strengths, and their subsequent inclusion in protocols for formulation are likely to be important future developments in this field.
ANGRY AFFECT IN THE FORMULATION OF AGGRESSION The role of anger in formulation of aggression merits special consideration, for two reasons. First, whilst few would argue that anger is a necessary or sufficient cause for aggression, there is considerable evidence that anger is an important contributing factor, in combination with other antecedent factors (Cavell and Malcolm, 2007; Day et al., 2008; Howells, 2009). Second, anger treatments are one of the most widely delivered therapeutic interventions for aggressive individuals, particularly in criminal justice settings, though evidence for effectiveness for violent offenders is, as yet, ambiguous (Howells et al., 2005; Heseltine et al., 2009). The importance of anger as an antecedent is underlined by the fact that this particular emotion has been and continues to be the subject of considerable theoretical and conceptual analysis (Novaco and Welsh, 1989; Davey et al., 2005; Novaco, 2007; Howells, 2009). This confers a major benefit – allowing the identification by the problem formulator, for example, of the important component processes involved in anger, including eliciting events, cognitive appraisals and schema, patterns of psychophysiological activation, self-regulatory strategies and action tendencies. This benefit is further increased by the fact that an array of psychometric measures exists to assess some of these components.
HETEROGENEITY In the past decade, evidence has increasingly demonstrated that aggressive offenders are indeed heterogeneous in terms of their needs and causal influences
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
113
on their offending behavior. Although anger is important for some, for others it is irrelevant. One of the most long-enduring distinctions in the literature has been between hostile and instrumental forms of aggression (McEllistrem, 2004). Hostile aggression comprises a triggering event, typically in the form of a frustration and an internal state of negative emotional arousal and an impulse to hurt or harm the provoking agent. Instrumental aggression is intended to secure an environmental reward and negative emotional arousal is less likely to be present as an antecedent. Some robberies illustrate instrumental violence, though not all, while homicides appear to be predominantly anger mediated forms of violence (Howells, 2009). A wide variety of terms have been used in the literature to describe what is essentially the same distinction, for instance angry, affective, hostile, reactive, hot-blooded vs. non-angry, predatory, proactive, planned, cold-blooded (see McEllistrem, 2004). Offenders who engage in hostile aggression would typically be seen as requiring interventions to develop more effective self-regulation of anger and other negative affective states (Cavell and Malcolm, 2007), while instrumental aggressors have different and poorly understood treatment needs. The developmental literature also reinforces the utility of the angry versus instrumental distinction discussed above. Hubbard and colleagues (2007), in reviewing this area, concluded: in our opinion, the distinction between reactive and proactive aggression is critical . . . . . some episodes of children’s aggressive behavior are strongly driven by anger, whereas other instances of aggression are markedly lacking in anger, being driven instead by a desire to achieve an instrumental or social goal. (p. 270)
Hubbard et al. point out that, although angry and instrumental violence are correlated in children, the two forms of aggression have different correlates. Hubbard et al. interpret the developmental literature as revealing that children’s level of angry aggression is related to particular social, cognitive, and emotional characteristics (hostile attributional biases, depression, rejection by peers) while instrumental aggression has different correlates (positive outcome expectancies for aggression, instrumental goal orientation and deviant peer friendships (see Hubbard et al. 2007, pp. 270–1). The importance of emotional functioning is being increasingly acknowledged by those studying developmental antecedents for aggression and violence. Delays in important socio-emotional skills are linked with aggressive behavior. Bierman (2007) points to critical skills: labeling of own feelings, recognizing the feelings of others, empathy and interpersonal understanding, regulation of emotional and behavioral reactivity, coping with frustration and distress, and effective communication of emotion in problem-solving. Emotional arousal, particularly anger activation, interacts with cognitions and interpersonal contingencies in influencing subsequent aggression. This work on the early development of aggression is relevant to understanding adult violent offenders, given the increasing evidence that the aggression problems of severely aggressive adults can be traced back to this early developmental
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
114
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
stage. Tremblay (2008) has shown that the developmental trajectory for violence was such that “the postulated learning of physical aggression by males, as well as the onset of CPA (chronic physical aggression) was either occurring during the kindergarten year or before kindergarten, not during the elementary school years, and certainly not during adolescence” (p. 2615). Amongst the important conclusions drawn by Tremblay (2008, p. 2616) are: (1) that children spontaneously use physical aggression; (2) rather than learn to physically aggress, they learn not to physically aggress; and (3) children learn not to physically aggress mainly in the preschool years. This contradicts widely held public views that violence is a product of adolescence, though, of course, the physical growth accompanying adolescence may make violence more damaging when it occurs. The data also appear, as Tremblay suggests, to contradict the view that violence is largely a product of exposure to media violence. It is likely that these early problems persist through adulthood in such individuals and are still in need of remediation. Helping violent offenders cope with severe developmental failures that persist into their adult life is a challenge indeed. Current neurobiological and developmental studies of violence appear to concur and to support the importance of the angry-instrumental distinction. In the authoritative review of the neurobiology of violence published by the Royal Society (Hodgins et al., 2009), for example, Michael Rutter concluded that violence is heterogeneous and that angry, instrumental and sadistic forms of violence should be distinguished. Blair’s work on brain systems suggests different neuro-anatomical pathways for angry and instrumental violence (Blair, 2004, 2008) and observed different developmental pathways to violence are broadly consistent with this way of dividing up the violent offending category. Dadds and Rhodes (2008) also point to heterogeneity of response to behavioral interventions with antisocial and violent children. In this case, callous and unemotional traits in boys were associated with poor responsiveness to behavioral programs. There are a number of problematic features of the angry versus instrumental distinction (Bushman and Anderson, 2001; Howard, 2009). The first is that the distinction is most appropriately applied to acts than to actors. An aggressive individual may engage in both forms of aggression, though a particular type of offense may still predominate in his history. Second, the distinction between the two forms is often difficult to make in practice (Barratt and Slaughter, 1998; Bushman and Anderson, 2001). Third, hostile aggression appears to be wrongly confounded with impulsive aggression and instrumental with planned violence (Howard, 2009). Many crimes of aggression indicate that angry reactions to a provocation can be carefully rehearsed and nurtured over time until a planned retaliation, delayed revenge is enacted. While some clinicians might readily equate instrumental with “cold” psychopathic characteristics, Patrick’s analysis (2006) suggests a complex picture emerging on the role of angry emotion in psychopathy (Howells, 2009). The main value of the hostile-instrumental distinction, perhaps, is not to introduce a further typology but to alert the formulator to the variation between aggressive individuals in terms of triggering events, cognitive and affective mediators and the goals of the act. The treatment targets for someone with an anger
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
115
regulation problem will differ substantially, for example, from the person who impulsively uses aggression to secure a range of reinforcers in the environment.
MENTAL DISORDER AND AGGRESSION The link between mental disorder and violence has been a contentious issue for more than three decades. A very substantial recent epidemiological study in the USA (Elbogen and Johnson, 2009) confirms the complexity of the relationship and casts some doubt on the notion that significant associations are necessarily causal. In this longitudinal study, severe mental illness alone was not a predictor of subsequent violent acts, though it became a significant predictor in the presence of comorbid substance abuse. Historical, dispositional and contextual factors were stronger predictors of violence than mental disorder per se. Elbogen and Johnson suggested that “clinicians (need) to look beyond diagnosis and consider a patient’s historical and current life situation more closely when assessing risk of violence” (p. 156). Elbogen and Johnson suggested that substance abuse and mental disorder interact in the sense that comorbid groups have higher rates of violence than either independently has. The authors conclude that “if a person has severe mental illness without substance abuse and a history of violence, he or she has the same chances of being violent in the next 3 years as any other person in the general population” (p. 157). This report draws attention to the importance of environmental stressors (unemployment, relationship breakdown) to violence. This research and other studies like it confirm that the formulation task with a violent or aggressive mentally disordered person should not stop short and focus exclusively on disorder factors but needs to be wide-ranging, including the potential antecedents demonstrated to be relevant in studies of aggression in the nondisordered.
PERSONALITY DISORDER AS AN ANTECEDENT In the past decade there has been considerable interest in the extent to which personality disorder is an antecedent for aggressive and violent behavior (McMurran and Howard, 2009; Tennant and Howells, 2010). Indeed, the increasing belief that a functional link existed between personality disorder and dangerous, predominantly violent, behaviors provided a rationale in England for the development of a raft of new therapeutic programs intended to reduce risk by targeting and treating personality disorder itself (Howells et al., 2007); however, the extent to which personality disorder explains and predicts violence remains unclear (Duggan and Howard, 2009). Duggan and Howard (2009), following Haynes (1992), point to four necessary conditions for a causal relationship: covariation between variables, temporal precedence of the causal variable, exclusion of alternative explanations, and establishing a logical connection between variables. They propose that Haynes’s conditions have not yet been met in relation to personality disorder as a cause of violence:
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
116
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
“The evidence, such as it is, suggests that any relationship that exists between PD and violence is weak, and that PD, including ASPD, probably accounts for only a very small proportion of the variance in violent behaviour” (Duggan and Howard, 2009, p. 29). This sceptical view, if correct, has considerable implications for formulation of cases and ultimately for treatment interventions. It implies that the co-occurrence of personality disorder and violent behavior should not necessarily be interpreted as indicating a causal relationship is present. Thus, in some cases, the presence of personality disorder may be discarded as irrelevant, if the aim of a formulation is to identify causal influences and to translate such influences into treatment targets. Personality disorder, in the language of offender rehabilitation, has not yet been unequivocally demonstrated to be what has been termed a criminogenic need (Andrews and Bonta, 2006). It is possible of course that personality disorder is still relevant as a responsivity factor, needing to be addressed therapeutically because it is likely to interfere with engagement in or compliance with treatment (Howells and Tennant, 2010; Ward et al., 2004). Reducing aggressive and high risk behavior is typically not the only objective of treatment services and it may be that the reduction of the interpersonal dysfunctions associated with personality disorder is deemed worthwhile in itself. In this instance, personality disorder is a relevant antecedent, but for the problem of social dysfunction rather than aggression per se. The other implication of the sceptical view is that there is a clear need to identify in the formulation other variables (apart from personality disorder) which are likely to be functionally linked to violence. This suggestion parallels the conclusions of Elbogen and Johnson (2009) above in relation to Axis I disorders. Fortunately, the scientific and empirical literature on causes of violence is a substantial one, far more robust than that relating to personality disorder (Howells et al., 2008; Cavell and Malcolm, 2007; Sestir and Bartholomew, 2007; Gannon, 2009). The clinical formulator is typically required to assess the particular individual rather than to determine whether there is a nomothetic relationship. That is, to determine whether for this person the personality disorder is causally linked to aggression. This is frequently a difficult, sometimes impossible, task. There are several methodological problems in trying to answer this question. The fact that an individual belongs to two populations that overlap (the personality disordered and violent offenders) does not prove that personality disorder is a functional (causal) antecedent for that person’s offending. The most common ways clinically to determine whether a causal relationship exists are to (1) observe natural covariation between the two variables for example, by asking the question “Is s/he more aggressive when they have the disorder than when they do not?” and (2) To treat the disorder and then observe the effect on aggression. Such methods would work well in relation to, for example, depression but not for personality disorder which shows little variation over time and which is, as yet, largely untreatable. Where the central aim of a clinical service is to change aggressive behavior, the emphasis on structural, diagnostic antecedents may be unhelpful in that appropriate attention is not given to variables likely to have a genuinely causal status. The disorder becomes the problem rather than the aggressive behavior itself. A formulation needs to focus on the central problem at hand and to be sufficiently comprehensive that a broad array of causal factors is considered. Offense-paralleling
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
117
behavior (Daffern et al., 2010) then becomes a core rather than a peripheral phenomenon requiring attention in the clinical formulation.
PERSONALITY AS AN ANTECEDENT It is plausible that long-term dispositions or traits, often summarized as personality, but falling short of personality disorder, are causally significant and thus need consideration in a clinical formulation of aggression. In a large theoretical and metaanalytic investigation, Bettencourt et al. (2006) investigated personality correlates of different types of aggression using a wide range of methodologies. A consistent pattern emerged across studies: that some personality variables predict aggression under neutral situational conditions and others under provocative conditions. In particular, aggression under provocation was predicted by trait anger, type-A personality, rumination, narcissism and impulsivity. Such studies are important in identifying variables that might be investigated in a formulation and also in drawing attention to the importance of situational factors – in this case to the need for a provocation to be present for aggressive behavior to be elicited. What provokes an individual will vary and may be idiosyncratic, with signature eliciting events needing to be identified for the individual (Mischel, 2004). Such situational and state factors have been neglected in formulations which are often focused on trait variables. A balanced approach is required in clinical settings with state, situational and dispositional factors, and their interactions, all requiring full consideration. Such a comprehensive approach to formulation is entirely consistent with general aggression theory (Bushman and Anderson, 2001; Sestir and Bartholomew, 2008) with its emphasis on a multiplicity of interacting causative factors.
INDIVIDUAL AND POPULATION FUNCTIONAL ASSESSMENTS The formulation of aggression may be either idiographic (that is at the individual level) or nomothetic (at the population level). A nomothetic needs analysis might suggest, for example, that alcohol misuse and emotional dysregulation are known antecedents of aggression and that, therefore, treatment programs to address these problems are required in an aggressor population. An idiographic analysis of an individual (rather than the group) might suggest, on the other hand, that a particular person’s aggression is influenced, for example, by exposure to humiliating provocations and subsequent paranoid ideation. Population analyses are essential in service planning while individual analysis is required to plan treatment for the individual aggressor. Idiographic assessment of causal influences allows for the likelihood that some influences will be idiosyncratic and specific to the individual under consideration. Such analysis is likely to be particularly important in complex populations. An example of the latter would be those with high risk of violence associated with the presence of severe personality disorder such as those deemed dangerous and severe personality disorders (DSPD) in England (Howells, Krishnan and Daffern, 2007; Tennant and Howells, 2010).
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
118
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Function-analytic assessments (Daffern and Howells, 2002, 2009; Sturmey, 1996, 2008) seek to clarify the factors responsible for the development, expression and maintenance of problem behaviors. Such methods may be of assistance in understanding the factors contributing to the development, expression and maintenance of aggression. Function-analytic assessments are typically achieved through assessment of the behavior of interest, assessment of the individual’s predisposing characteristics, and consideration of the antecedent events, which are important for the initiation of the behavior, and the consequences of the behavior, which maintain and direct its developmental course (Haynes, 1992).
FURTHER ASSESSMENT ISSUES Assessment in relation to aggressive offenders is best considered within two frameworks, firstly the Risk-Needs-Responsivity (RNR) framework (Andrews and Bonta, 2006), also known as ‘what works’, which has been a dominant influence on offender rehabilitation over more than 15 years. Second, it needs to be considered within the framework provided by aggression theory. RNR highlights three important assessment tasks, following from the accepted rehabilitation principles of risk, criminogenic needs, and responsivity. Aggressive offenders overlap with offenders in general in terms of their patterns and type of offending and in the criminogenic needs they have. Thus, assessments and formulations relevant to offenders in general will be relevant to those who offend aggressively. In addition, however, aggressive offenders will have criminogenic needs (and consequent treatment needs) specific to their aggressive acts. A wide range of assessment tools to assess aggression and violence are currently available (Suris et al., 2004). Many of these have been developed to suit the needs of the particular setting in which they are used, and tools developed for use in correctional settings usually seek to identify individual differences that discriminate between those who are likely to reoffend and those who are not. It follows that variables such as situational triggers, empathy deficits, cognitive and information-processing biases, affective-regulation deficits (particularly in relation to anger), overcontrol (Davey et al., 2005), impulsivity, personality dispositions such as antagonism and neuroticism (Bettencourt et al., 2006; Blackburn, 2007), DSM-IV personality disorders, psychopathy, Axis I mental disorders and substance abuse are all areas requiring assessment prior to effective formulation and treatment of aggressive offending. Treatment itself, of course, will need to also focus on the psychological mechanisms underlying these factors. In some settings such extensive assessment may be seen to be overly burdensome and expensive to implement, with the result that detailed and extensive assessment and formulation is implemented only with those who pose a very high risk of future violence, as in the dangerous and severe personality disorder program in England (Howells et al., 2007). Consideration of state and situational factors and their effects on disturbed behavior have been bread-and-butter tasks in clinical assessments in general mental
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
119
health for many years, a product, perhaps, of the emphasis on comprehensive formulation in clinical psychology (Sturmey, 2008), but this has been curiously neglected, until recently, in offender assessment. The state and situational emphasis is very congruent with the functional analytic approach to violence assessment described below. Douglas and Skeem (2005) have described methodologies for assessing dynamic factors influencing violence. Ogloff and Daffern (2007) have devised a method for 24 hour prediction of violence (the Dynamic Appraisal of Situational Aggression), based on daily ratings of behavior by staff, which is based on the variation in violence propensity from day to day.
FUNCTION ASSESSMENTS OF AGGRESSION AND VIOLENCE An assessment framework for analysing the functions of aggressive actions (assessed through review of proximal antecedents and consequences within the context of the individual’s predisposing personal attributes and limitations) has been proposed by Daffern and colleagues (Daffern et al., 2007; Daffern and Howells, 2007; Daffern and Howells, 2009). This classification system, the ‘Assessment and Classification of Function’ (ACF; Daffern et al., 2007), acknowledges that multiple functions may be present for any particular act and that perpetrators may have different goals for different acts. These functions are: 1. 2. 3. 4. 5. 6. 7. 8. 9.
demand avoidance; to force compliance; to express anger; to reduce tension (catharsis); to obtain tangibles; social distance reduction (attention-seeking); to enhance status or social approval; compliance with instruction; and to observe suffering.
For the ACF, each function is recognized through its characteristic antecedents and consequences and scored as present or absent for a particular aggressive behavior. The ACF acknowledges that violence may have multiple functions and goals for the individual as well as for the group. Indeed any one violent act may have multiple functions, suggesting it is unhelpful to think of functions as necessarily characteristic of the individual offender. Thus, methods such as the ACF classify acts rather than actors. So far, the major applications of the ACF have been to mentally disordered violent offenders (Daffern et al., 2007) and to personality disordered offenders (Daffern and Howells, 2009). It would seem to hold promise as a method for assessing violent offenders in the criminal justice system. In a study of 502 aggressive incidents in a high security forensic psychiatric hospital in Victoria, Australia, Daffern
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
120
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
et al. (2007) found that anger expression was the most frequent function of aggression but that functions differed for aggressive behaviors toward staff and those toward fellow patients. Demand avoidance was a common function for aggression toward staff but rare for aggression to patients. To obtain tangibles (an instrumental function) was rare for all incidents. Daffern and Howells (2009) have recently extended this work to high risk offenders with personality disorders in Rampton Hospital in the United Kingdom and added two further categories of function to the original nine functions, namely “sensation seeking” and “sexual gratification” to capture apparent sexual/sadistic functions occurring in this very high risk and complex population. In the latter study the function of the violent offenses leading to admission (index offenses) proved to differ substantially from the functions of violent behaviors within the institution. The importance of functional assessments of this sort is that they clearly suggest different therapeutic strategies for different violent offenders, depending on the exact functions of their violence. An offender who is predominantly “anger expressive”, for example, would have different needs from someone who is mainly seeking sensation or sexual arousal. The general direction of therapy would be to develop alternative, adaptive means of achieving personal functional goals and to change circumstances and other factors within the person in such a way that the need to pursue the problematic goal is reduced.
IMPLEMENTATION OF THE CLINICAL FORMULATION OF AGGRESSION IN PRACTICE: SOME AWKWARD QUESTIONS Within forensic mental health and criminal justice systems considerable resources are devoted to treatment interventions which have as a primary objective the reduction of aggressive behavior (McGuire, 2008). Formulation of the individual case is frequently endorsed, both in principle and in practice. There exist comprehensive treatment programs explicitly based on an individual formulation model (Hogue et al., 2007; Jones, 2010). Nevertheless, it is possible that formulations in practice fall short and do not live up to the standards required for a genuinely systematic and scientific approach, as encapsulated in the scientist-practitioner model of practice. An important task for the future will be to survey the quality of clinical formulations as currently delivered by practitioners. To my knowledge, no such surveys have been reported in the literature. There is clearly a need for formulation guidelines to be developed so that a consistent approach can be adopted across clinical settings. Questions relating to quality and practicality would include the following: 1. Do you have a systematic, standardized protocol for conducting formulations of aggression in your service? 2. Are staff trained to conduct formulations of aggression? Whilst many professionals, particularly clinical psychologists, have received generic training in formulation as part of their pre-qualification training, few may have applied formulation skills to aggression.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
121
3. Are formulations sufficiently comprehensive, covering the categories of antecedents and functions outlined above? 4. Are the formulations scientific in terms of precision, capacity for revision and falsifiability (Kuyken, 2006; Sturmey, 2009; Haynes et al., 2009)? 5. Have efforts been made to determine the reliability of formulations of aggression? that is, having more than one practitioner provide a formulation for the same case material? This is essential when a formulation may have important implications for risk management, detention and discharge. 6. Are the treatment targets and therapies delivered consistent with the original formulation made? Complex formulations are redundant if they do not explicitly shape treatment targets, risk assessments and the treatments delivered. 7. What is the process for conducting formulations? Which profession is responsible and how is multidisciplinary team functioning accommodated in the formulation? 8. Are formulations produced collaboratively with the patient or offender? 9. Are formulations disseminated effectively to those that need to know them? 10. How is long-term staff awareness of the formulation maintained so that the initial formulation, often conducted close to the time of the violent act, for example shortly after admission to a facility, is not forgotten?
Anecdotally, it appears that an initial formulation may be forgotten so that attention is increasingly, and sometimes unhelpfully focused on the person’s recent behavior. There are some indications (Daffern and Howells, 2009) that the antecedents and functions for institutional aggressive behavior may differ from those pertaining to the original index offense that may have led to admission. Effectiveness is likely to depend on the clarity and brevity of the formulation (Sturmey, 1996). Aggression problems are linked to issues of risk of harm to other patients, staff and members of the community. A technically expert formulation is vitiated if it is inaccessible to another professional further down the track in the treatment pathway, or even to other staff in the same service who are dealing with a patient, for example, in an occupational therapy or similar setting. It is likely that more than one version of a formulation is required to allow for the differential level of detailed needed or appropriate for professionals with different roles in relation to the patient or offender. A potential therapist, for example, may need a detailed formulation of the problem, while an ancillary worker may require a more schematic summary.
TWO CONTRASTING CASE FORMULATIONS The following vignettes have been modified in details to protect anonymity. They formed part of a study interviewing imprisoned violent men in detail about their index violent offense and its immediate antecedents. Interview accounts were subjects to the ACF analysis for functions of violence developed by Daffern et al. (2007) and described in this chapter.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
122
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Jim Jim was in his early 20s at the time of the offense. In the afternoon of the day of his offense he received news that his mother had been severely injured in a car accident. He immediately began consuming alcohol. After visiting her in hospital he became distressed and tearful. Later, a friend offered to take him drinking in a club. He drank heavily and his mood improved initially though he later became morose and despondent. On leaving the club, in a very inebriated state, he saw a male who he recognized as someone who had previously been verbally abusive to his mother. He approached him and a row started. It ended but Jim decided to follow the man home, with strong feelings of anger and the need for revenge. He remembers thinking that the man deserved to be punished for the previous bad behavior to Jim’s mother. “I just wanted to make him afraid. I threw him against a wall, kicked him and kept on hitting him. I was so angry, I just lost control. The violence just helped me to reduce my stress and tension from what had happened earlier in the day. I needed it. I was also glad I got my revenge.” In terms of Daffern’s ACF functions the violent act was classified as predominantly “anger expression” and “tension reduction”. There was little indication of other functions such as “obtaining tangibles” or enhancing “status and social approval” – the offense was committed alone rather than in a peer-group. His previous occasional episodes of violence showed a similar pattern. All had occurred at times of distress and anger at someone who he perceived as mistreating himself or his family and occurred when he had been drinking heavily.
Chris “It was just a normal day in . . . I got up early and started drinking at 8.30 am. I was in an OK mood and my plans were just to chill. I was just hanging with my cousin. We then decided to just walk around, I suppose we were a bit bored . . . I was a bit short of cash.” Later in the morning, Chris, with friends, saw a stranger walking toward them on a deserted street. He and his friends “put their hoods up” like a balaclava which they collectively took as a signal that they were about to do something antisocial. “I didn’t want to stand out and be different.” The group threatened the stranger and demanded he hand over “something”. The man resisted and the group began hitting and kicking him. He fought back and their violence to him became more severe. Chris reported no feelings of anger or resentment. He initially felt bored but became excited as the fight progressed and enjoyed the “adrenaline rush” of “laying into him”. He reported indifference to the man’s significant injuries and subsequent hospitalization and felt satisfied that they had been successful in securing a wallet and a mobile phone which could be traded for money and further alcohol. Chris’s violence was categorized as in pursuit of several goals “to force compliance” from the victim, “to obtain tangibles”, to “enhance status” in his peer group and “sensation seeking” at a time of boredom.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
123
Comment The functions of Chris’s aggression are clearly more instrumental in nature than those of Jim above and, when compared, both cases illustrate the diversity of functions of these acts of violence. Were both acts of violence typical of other similar acts of violence for each person, it is clear that different patterns of proximal antecedent events exist and that, in a therapy context, both would have very different treatment targets.
SUMMARY In summary, clinical formulation is a core task in the assessment and treatment of aggressive behavior. The available evidence and contemporary theories suggest that a very wide range of potentially causal variables need to be considered. A broad approach allows for the demonstrated heterogeneity of aggressive acts and aggressive people and for the unique constellation of factors that may characterize the causation of aggression for an individual. Formulation needs to include intrapersonal but also broader situational, state and cultural variables. Mental disorder alone is rarely a sufficient explanation of aggressive acts. Future studies and service improvement initiatives need to consider how adequately clinical formulation is implemented in practice.
REFERENCES Anderson, C.A. and Bushman, B.J. (2002) Human aggression. Annual Review of Psychology, 53, 27–51. Andrews, D. and Bonta, J. (2006) The Psychology of Criminal Conduct (3rd edn). Cincinnatti OH: Anderson. Baron, R.A. and Richardson, D.R. (1994) Human Aggression. New York: Plenum. Barratt, E.S. and Slaughter, L. (1998) Defining, measuring and predicting impulsive aggression: A heuristic model. Behavioral Sciences and the Law, 16, 285–302. Bettencourt, B.A., Talley, A., Benjamin, A.J. and Valentine, J. (2006) Personality and aggressive behaviour under provoking and neutral conditions: A meta-analytic review. Psychological Bulletin, 132, 751–77. Bierman, K.L. (2007) Anger and aggression: A developmental perspective. In T.A. Cavell and K.T. Malcolm (eds), Anger, Aggression and Interventions for Interpersonal Violence. Mahwah NJ: Lawrence Erlbaum, pp. 215–38. Blackburn, R. (1993) The Psychology of Criminal Conduct. Chichester: John Wiley & Sons, Ltd. Blackburn, R. (2007) Personality disorder and antisocial deviance: Comments on the debate on the structure of the Psychopathy Checklist-Revised. Journal of Personality Disorders, 21, 142–59. Blackburn, R. (2009) Subtypes of psychopath. In M. McMurran and R. Howard (eds), Personality, Personality Disorder and Risk of Violence. Chichester: John Wiley & Sons, Ltd, pp. 113–32. Blair, R.J.R. (2004) The roles of orbital frontal cortex in the modulation of antisocial behaviour. Brain and Cognition, 55, 198–208. Blair, R.J.R. (2008) The amygdala and ventomedial prefrontal cortex: Functional contributions and dysfunction in psychopathy. In S. Hodgins, E. Viding and A. Plodowski (eds),
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
124
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
The neurobiology of violence: Implications for prevention and treatment. Philosophical Transactions of the Royal Society B, 363, 2557–65. Bushman, B.J. and Anderson, C.A. (2001) Is it time to pull the plug on the hostile versus hostile versus instrumental aggression dichotomy? Psychological Review, 108, 273–9. Buss, A. (1961) The Psychology of Aggression. New York: John Wiley & Sons, Inc. Campbell, A. (2006) Sex differences in direct aggression: What are the psychological mediators? Aggression and Violent Behavior, 11, 237–64. Casey, S., Day, A., Howells, K. and Ward, T. (2007) Assessing suitability for offender rehabilitation: Development and validation of the Treatment Readiness Questionnaire. Criminal Justice and Behavior, 34, 1427–40 Cavell, T.A. and Malcolm, K.T. (eds) (2007) Anger, Aggression and Interventions for Interpersonal Violence. Mahwah NJ: Lawrence Erlbaum. Dadds, M.R. and Rhodes, T. (2008) Aggression in young children with concurrent callousunemotional traits: Can the neurosciences inform progress and innovation in treatment approaches? In S. Hodgins, E. Viding and A. Plodowski (eds), The neurobiology of violence: implications for prevention and treatment. Philosophical Transactions of the Royal Society B, 363, 2567–76. Daffern, M. and Howells, K. (2002) Psychiatric inpatient aggression: A review of structural and functional assessment approaches. Aggression and Violent Behaviour, 3, 1–21. Daffern, M. and Howells, K. (2007) Antecedents for aggression and the function analytic approach to the assessment of aggression and violence in personality disordered patients within secure settings. Personality and Mental Health, 1, 126–37. Daffern, M. and Howells, K. (2009) The function of aggression in personality disordered patients. Journal of Interpersonal Violence, 24, 586–600. Daffern, M., Howells, K. and Ogloff, J.R.P. (2007) What’s the point? Towards a methodology for assessing the function of psychiatric inpatient aggression. Behaviour Research and Therapy, 45, 101–11. Daffern, M., Jones, L. and Shine J. (2010) Offence Paralleling Behaviour: An Individualised Approach to Offender Assessment and Treatment. Chichester: Wiley-Blackwell. Davey, L., Day, A. and Howells, K. (2005) Anger, overcontrol and violent offending. Aggression and Violent Behavior, 10, 624–35. Day, A., Casey, S., Ward, T. et al. (2010) Transitions to Better Lives: Offender Readiness and Rehabilitation. Cullompton UK: Willan. Day, A., Howells, K., Casey, S. et al. (2009) Assessing treatment readiness in violent offenders. Journal of Interpersonal Violence, 24, 618–35. Day, A., Nakata, M. and Howells, K. (eds) (2008) Anger and Indigenous Men. Sydney: The Federation Press. Day, A., Tucker, K. and Howells, K. (2004) Coerced offender rehabilitation: A defensible practice? Psychology, Crime & Law, 10, 259–69. Douglas, K.S. and Skeem, J.L. (2005) Violence risk assessment: Getting specific about being dynamic. Psychology, Public Policy and Law, 11, 347–83. Duggan, C. and Howard, R.C. (2009) The “functional link” between personality disorder and violence: A critical appraisal. In M. McMurran and R. Howard (eds) Personality, Personality Disorder and Risk of Violence (pp. 19–37). Chichester: John Wiley & Sons, Ltd. Elbogen, E.B. and Johnson, S.C. (2009) The intricate link between violence and mental disorder: Results from the national epidemiological survey on alcohol and related conditions. Archives of General Psychiatry, 66, 152–61. Gannon, T.A. (2009) Social cognition and aggressive offending: An overview. Psychology, Crime and Law, 15, 97–118. Hare, R.D. (2003) The Hare Psychopathy Checklist-Revised (2nd edn). North Tonawanda NY: Multi-Health Systems. Hare, R.D. (2006) Psychopathy: a clinical and forensic overview. Psychiatric Clinics of North America, 29, 709–24. Haynes, S.N. (1992) Models of Causality in Psychopathology. New York: Macmillan.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
125
Haynes, S.N., Yoshioka, D.T., Kloezeman, K. and Bello, I. (2009) Clinical applications of behavioural assessment: Identifying and explaining behaviour problems in clinical assessment. In: J. Butcher (ed.) Oxford Handbook of Clinical Assessment. Oxford, UK: Oxford University Press. Heseltine, K., Howells, K. and Day, A. (2009) Brief anger interventions with offenders may be ineffective: A replication and extension. Behaviour Research and Therapy, 48, 246–50. Hodgins, S. (2008) Criminality among persons with severe mental illness. In K. Soothill, M. Dolan and P. Rogers (eds), The Handbook of Forensic Mental Health. Cullompton, Devon: Willan, pp. 400–23. Hodgins, S., Viding, E. and Plodowski, A. (eds) (2009) The Neurobiological Basis of Violence: Science and Rehabilitation. Oxford: Oxford University Press. Hogue T.E, Jones, L., Talkes, K. and Tennant, A. (2007) The Peaks: A clinical service for those with dangerous and severe personality disorder. Psychology, Crime and Law, 13, 57–68. Hollin, C.R. and Palmer, E.J. (2006) Offending Behaviour Programmes: Development, Application and Controversies. Chichester: John Wiley & Sons, Ltd. Howard, R. (2009) The neurobiology of affective dyscontrol: Implications for understanding “Dangerous and Severe Personality Disorder”. In M. McMurran and R. Howard (eds), Personality, Personality Disorder and Risk of Violence (pp. 157–74). Chichester: John Wiley & Sons, Ltd. Howells, K. (2009) Angry affect, aggression and personality disorder. In M. McMurran and R. Howard (eds), Personality, Personality Disorder and Risk of Violence (pp. 191–212). Chichester: John Wiley & Sons, Ltd. Howells, K. (2010) Distinctions within distinctions: The challenges of heterogeneity and causality in the formulation and treatment of violence. In Daffern, M., Jones, L. and Shine, J. (eds), Offence Paralleling Behaviour: An Individualised Approach to Offender Assessment and Treatment. Chichester: Wiley-Blackwell. Howells, K., Daffern, M. and Day, A. (2008) Aggression and violence. In K. Soothill, M. Dolan and P. Rogers (eds), The Handbook of Forensic Mental Health (pp. 351–74). Cullompton, Devon: Willan. Howells, K. and Day, A. (2003) Readiness for anger management: Clinical and theoretical issues. Clinical Psychology Review, 23, 319–37. Howells, K., Day, A., Williamson, P. et al. (2005) Brief anger management programs with offenders: Outcomes and predictors of change. Journal of Forensic Psychiatry and Psychology, 16, 296–311. Howells, K., Krishnan, G. and Daffern, M. (2007) Challenges in the treatment of dangerous and severe personality disorder. Advances in Psychiatric Treatment, 13, 325–32. Howells, K. and Tennant, A. (2010) Ready or not, they are coming: Dangerous and severe personality disorder and treatment engagement. In A. Tennant and K. Howells (eds), Using Time, Not Doing Time: Practitioner Perspectives on Personality Disorder and Risk. Chichester: John Wiley & Sons, Ltd. Hubbard, J., McAuliffe, M.D., Rubin, R.R. and Morrow, M.T. (2007) The anger-aggression relation in violent children and adolescents. In T.A. Cavell and K.T. Malcolm (eds), Anger, Aggression and Interventions for Interpersonal Violence. Mahwah NJ: Lawrence Erlbaum. Jones, L. (2010) Case formulation with personality disordered offenders. In A. Tennant and K. Howells (eds), Using Time not Doing Time: Practitioner Perspectives on Personality Disorder and Risk (pp. 45–62). Chichester: Wiley-Blackwell. Kuyken, W. (2006) Research and evidence base in case formulation. In N. Tarrier (ed.), Case Formulation in Cognitive Behaviour Therapy: The Treatment of Challenging and Complex Clinical Cases (pp. 12–35). London: Brunner Routledge. McEllistrem, J.E. (2004) Affective and predatory violence: A bimodal classification system of human aggression and violence. Aggression and Violent Behavior, 10, 1–30. McGuire, J. (2008) A review of effective interventions for reducing aggression and violence. Philosophical Transactions of the Royal Society B, 363, 2577–97.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
126
May 28, 2011
8:20
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
McMurran, M. and Howard, R. (eds) (2009) Personality, Personality Disorder and Risk of Violence. Chichester: John Wiley & Sons, Ltd. Mischel, W. (2004) Towards an integrative science of the person. Annual Review of Psychology, 55, 1–22. Novaco, R.W. (2007) Anger dysregulation. In T.A. Cavell and K.T. Malcolm (eds), Anger, Aggression and Interventions for Interpersonal Violence. Mahah NJ: Lawrence Erlbaum, pp. 3–54. Novaco, R.W. and Welsh, W.N. (1989) Anger disturbances: Cognitive mediation and clinical prescriptions. In K. Howells and C.R. Hollin (eds), Clinical Approaches to Violence (pp. 39–60). Chichester: John Wiley & Sons, Ltd. Ogloff, J.R.P. and Daffern, M. (2007) The dynamic appraisal of situational aggression: An instrument to assess risk for imminent aggression in psychiatric inpatients. Behavioral Sciences and the Law, 24, 799–813. Parrott, D.J. and Giancola, P.R. (2007) Addressing “the criterion problem” in the assessment of aggressive behaviour: Development of a new taxonomic system. Aggression and Violent Behavior, 12, 280–99. Patrick, C.J. (2006) Back to the future: Cleckley as a guide to the next generation of psychopathy research. In C.J. Patrick (ed.), Handbook of Psychopathy (pp. 605–17). New York: Guilford Press. Persons, J.B. (2008) The Case Formulation Approach to Cognitive-Behavior Therapy. New York: Guilford Press. Rand Corporation (2009) Tackling Violent Crime. Technical Report. National Audit Office. Cambridge, UK: Rand Corporation. Rutter, M. (2008) Introduction to the papers. In S. Hodgins, E. Viding and A. Plodowski (eds), The neurobiology of violence: implications for prevention and treatment. Philosophical Transactions of the Royal Society B, 363, 2485–90. Serin, R.C. (2004) Understanding violent offenders. In D.H. Fishbein (ed.), The Science, Treatment and Prevention of Antisocial Behaviors. Volume 2: Evidence-Based Practice. Kingston NJ: Civic Research Institute. Sestir, M.A. and Bartholow, B. (2007) Theoretical explanations of aggression and violence. In T.A. Gannon, T Ward, A.R. Beech and D. Fisher. Aggressive Offenders’ Cognition. Chichester: John Wiley & Sons, Ltd. Sturmey, P. (1996) Functional Analysis in Clinical Psychology. Chichester: John Wiley & Sons, Ltd. Sturmey, P. (2007) Functional Analysis in Clinical Treatment. Burlington MA: Elsevier. Sturmey, P. (2008) Behavioral Case Formulation and Intervention. A Functional Analytic Approach. Chichester: Wiley-Blackwell. Sturmey, P. (ed.) (2009) Clinical Case Formulation: Varieties of Approaches. Chichester: WileyBlackwell. Suris, A., Lind, L., Emmett, G. et al. (2004) Measures of aggressive behavior: Overview of clinical and research instruments. Aggression and Violent Behavior, 9, 165–227. Tennant, A and Howells, K. (eds) (2010) Using Time Not Doing Time: Practitioner Perspectives on Personality Disorder and Risk. Chichester: Wiley-Blackwell. Thomas-Peter, B. (2006) The modern context of psychology in corrections: Influences, limitation and values of ‘What Works’. In G.J. Towl (ed.), Psychological Research in Prisons (pp. 24–39). Oxford: Blackwell. Tremblay, R.E. (2008) Understanding development and prevention of chronic physical aggression: Towards experimental epigenetic studies. In S. Hodgins, E. Viding and A. Plodowski (eds), The neurobiology of violence: implications for prevention and treatment. Philosophical Transactions of the Royal Society B, 363, 2613–25. Ward, T., Day, A., Howells, K. and Birgden, A. (2004) The multifactor offender readiness model. Aggression and Violent Behavior, 9, 645–73. Ward, T. and Gannon, T. (2006) Rehabilitation, etiology and self-regulation: The Good Lives Model of sex offender treatment. Aggression and Violent Behavior, 11, 77–94. Ward, T. and Maruna, S. (2007) Rehabilitation. London: Routledge.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
COGNITIVE BEHAVIORAL APPROACHES TO FORMULATING AGGRESSION
127
Webster, C., Harris, G., Rice, M., Cormier, C. and Quinsey, V. (1994) The Violence Prediction Scheme: Assessing Dangerousness in High Risk Men. Toronto: University of Toronto Centre of Criminology. Webster, C.D., Douglas, K., Eaves, D., and Hart, S. (1997) HCR-20: Assessing Risk for Violence – Version 2. Vancouver BC: Simon Fraser University. Wong, S. and Gordon, A. (2006) The validity and reliability of the Violence Risk Scale: a treatment friendly violence risk assessment tool. Psychology, Public Policy and Law, 12, 279–309.
P1: OTA/XYZ P2: ABC JWST073-05 JWST073-McMurran
May 28, 2011
8:20
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
Chapter 6
FORMULATION OF SERIOUS VIOLENT OFFENDING USING MULTIPLE SEQUENTIAL FUNCTIONAL ANALYSIS AIDAN J.P. HART University of Lincoln, UK
DAVID M. GRESSWELL University of Lincoln, UK
LOUISE G. BRAHAM University of Nottingham & Nottinghamshire Healthcare NHS Trust
INTRODUCTION Core to the work of the applied psychologist working in forensic settings is the assessment and formulation of complex and problematic behaviors. The purpose of such assessments is often twofold: First, they facilitate a thorough risk assessment in order to determine the likelihood of future occurrences of the behaviors in question, and second they aid the design and implementation of appropriate interventions to reduce the reoccurrence of such behaviors. In nonforensic clinical settings, developing a case formulation based on sound psychological theory whilst simultaneously providing a parsimonious, accurate and testable account of a client’s difficulties is typically a time-consuming task. Such assessments if undertaken are often inconsistent with the currently prevailing philosophy evident in many forensic settings, that clinical risk judgements are unreliable in comparison to actuarial methods. Criticism of individual clinical assessment has become embedded in the professional culture of risk assessment and perpetuated by the culture of manualized group treatments; these standardized procedures militate against the development of individual case formulation. Nevertheless, within the current professional climate, the issue of
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
130
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
individual formulation remains an important one. Not all individuals respond to or engage with group treatments at the same level and these individual differences need to be explained and incorporated into treatment planning and assessment. Recent developments in manualized treatments for sexual or violent behavior have seen exercises specifically tailored to the particular offenses of the individual in question, even when delivered within a group format (see Braham, Jones and Hollin, 2008). It is therefore imperative that those delivering such programs have a detailed understanding of the individual’s criminal acts. While research on the predictive validity of actuarial risk assessments for violent offenders appears on the face of it to be supportive (Quinsey et al., 1996; Harris and Rice, 2007) on closer inspection the usefulness of such assessments for individual case management is limited. Even if problems with the confidence intervals in such measures (see Hart, Michie and Cooke, 2007) are set aside, such assessments only tell us that one offender’s score is similar to that of a group of offenders who reoffended at a rate of X; however, this tells us little about whether the offender is one of those who is likely to reoffend or one of those who will not, let alone anything about what circumstances increase the likelihood of this risk. Extrapolating an individual prediction and risk management plan from what is essentially a group mean can only be done in the context of a detailed individual case formulation or functional analysis of dynamic risk variables, despite prohibitions from some of the developers of such assessments against using parallel clinical judgement to interpret the results of such assessment (e.g., Quinsey et al., 1996). Indeed, other measures that incorporate dynamic variables (such as HCR-20, Webster et al., 1997; VRS, Gordon and Wong, 2000) specifically require clinical judgement in the determination of risk and the interpretation of the variables of which they are comprised. Using dynamic variables to provide an accurate description of risk requires the clinician to understand the function of the problem behaviors and to formulate the offending behavior both within the community and institutional settings; however, this process has its difficulties. With a convicted offender, the behavior under scrutiny will have typically occurred months, if not years, before the forensic practitioner begins his or her assessment (Jones, 2004). Indeed, in some cases there may be a pattern of escalating behavior which has not been directly observed or recognized. Furthermore the assessing clinician may find other co-occurring behaviors that may be clinically significant but not necessarily be functionally linked to the specific offending behavior of interest. An example of this can be seen in the hypothesized functional links between violent offending and a diagnosis of a psychotic disorder or alcohol use. The majority of people who suffer psychosis or abuse alcohol do not commit violent offenses. Therefore experiencing a psychosis or abusing alcohol do not of themselves predict who with a psychosis or alcohol misuse problem is more likely to commit a violent act or the circumstances under which this might happen. Carrying out a detailed functional analysis allows the assessing clinician to determine with much greater confidence whether and how any features of a psychotic or substance misuse disorder have influenced the offense in question. Such added understanding will therefore aid the development of a much more defensible and detailed treatment plan. In summary, whilst actuarial risk assessment tools may prove to be statistically accurate they lack precision when used with individuals and therefore tell us little
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
131
about the context in which such offending will occur and what the individual treatment implications of such risk is. Therefore, there remains a need for detailed, well structured and defensible idiographic case formulations of offending behavior. Despite the clear advantage of carrying out a detailed case formulation, there are a number of difficulties that the clinician needs to be aware of. First, the assessing clinician is typically not in the position to have witnessed the offense in question or the developmental learning history that preceded it. This first problem is compounded by the second immediate issue that the clinician has to face; that clients often minimize, distort or simply deny their offenses for obvious reasons, whether intentionally or not. The criminal justice and subsequent rehabilitation process, which typically involves repeated interviews and assessments, may also sensitize or desensitize an individual to the emotional impact of their offense. Subsequently, even cooperative well-motivated individuals may struggle to give an accurate account of their offending behavior or the emotional and cognitive factors they experienced during an offense. Despite the potential utility of formulation, some studies indicate that formulations carried out in clinical practice are often of poor quality (Kuyken et al., 2005). The purpose of this chapter is to consider the role of Multiple Sequential Functional Analysis (MSFA) in providing a framework for formulation and to highlight how MSFA can be used to address Kuyken et al.’s critique and increase the quality of formulations of serious offending behaviors. Further to this, an aim of this chapter is to demonstrate how MFSA can be used to delineate the functional properties of an offense from its topographical features and hence demonstrate that offenses that appear similar might also have key functional differences. Finally, how such a functional analysis can allow for the identification of Offense Paralleling Behaviors (Jones, 2004) and for the provision of different treatment and management pathways, will be discussed.
Multiple Sequential Functional Analysis MSFA (Gresswell and Hollin, 1992) is grounded in the philosophy of science known as radical behaviorism and the applied science of behavior analysis. It is the goal of any behavior analytic approach to understand, predict and influence behavior with precision, scope and depth. Behavioral or functional analysis as a psychological approach is grounded in the somewhat uncontroversial idea that behavior is rooted in and influenced by the context in which it occurs. Within a typical functional analysis, a specific or selected sample of behavior is analysed in terms of the environmental contexts (stimuli) that preceded it and the consequent changes in the environment (reinforcers, punishers etc.) which are hypothesized to make the behavior under consideration more or less likely to reoccur. To fully understand the process of functional analysis two questions that arise from the preceding statement must be answered; namely what is behavior and what is context? From a functional analytic perspective, behavior is the dependent variable in our analysis and it is best understood as an ongoing and ever-present stream of overt (observable) behavior and covert behavior (such as violent fantasies, or feelings of anger) that occurs in dynamic interaction with the environment (Wilson
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
132
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Table 6.1 Classes of antecedent stimuli. Antecedent stimulus
Definition
Discriminative stimulus (SD)
A stimulus that sets the occasion for operant behavior. The SD discriminates those occasions where a particular behavior is more likely to lead to a particular consequence An S is the opposite of an SD. An S is a stimulus that signals that reinforcement is not available An establishing operation is a direct or indirect procedure that temporarily increases or decreases the value of a reinforcer. A common example is that deprivation of food can increase the value of food as a reinforcer A US is a stimulus that produces an unconditioned response (UR) without direct explicit learning. Salivation at the sight of food is an example A neutral stimulus does not elicit a specific response (e.g. sound of bell before it is paired with food) A stimulus that after repeated pairing with the US comes to elicit the same response as the US. The response to the CS is known as the conditioned response (CR)
Stimulus delta (S) Establishing operation
Unconditioned stimulus (US) Neutral stimulus (NS) Conditioned stimulus (CS)
and DuFrene, 2009). To put it more simply, behaviour is whatever the organism is doing (Skinner, 1938). Context refers to anything outside of the person that is happening to them currently, has happened to them in the past, or may happen to them in the future. While context can be anything outside of the person, it does not refer to everything that is outside of the person; it is those aspects of the environment that have a functional relationship to the behavior under scrutiny. In this regard context is the independent variable in our analysis. Functional analyses are normally expressed in an A:B:C format in which the ‘A’ refers to Antecedents or stimuli thought to have triggered or elicited the behavior in question (see Table 6.1 for a breakdown of Antecedent Stimulus Classes); the ‘B’ is the Behavior itself, which may be described in terms of both overt and covert behaviors (such as thoughts, physiology changes, emotions etc.); whilst the ‘C’ refers to the Consequences that follow the behavior. In terms of considering the consequences of the behaviour, we are not just interested in any event that follows the behaviour. What we are interested in are the environmental changes contingent on the behaviour which may make it more likely to occur (positive and negative reinforcers) or less likely to occur (punishers, extinction) in the future. In addition, within a behaviorally orientated MSFA framework, it is not just the consequences themselves but the manner in which they are delivered and the relationship between behavior and consequence in terms of response ratio and time interval (Ferster and Skinner, 1957). See Table 6.2 for a summary of classes of consequential stimuli and schedules of reinforcement and punishment. MSFA is a development of this approach in which a series of functional analyses are linked together to account for complex sequences of learning, resulting in current behavior and current learning (Gresswell and Hollin, 1992).
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
133
Table 6.2 Classes of consequential stimuli, schedules of reinforcement and punishment. Consequential stimulus
Definition
Positive reinforcement
An increase in the future frequency of behavior due to the addition of a stimulus following a response upon which the consequence was contingent An increase in the future frequency of a behavior due to the removal of a stimulus following a response upon which the consequence was contingent A decrease in the future frequency of a behavior due to the addition of a stimulus following a response upon which the consequence was contingent A decrease in the future frequency of a behavior due to the removal of a stimulus following a response upon which the consequence was contingent The time between response and the contingent consequence The number of responses required to produce the contingent consequence The same time interval or number or responses is required to produce reinforcement A variable time span or number of responses is needed to produce a reinforcement. Such a schedule of reinforcement has significant power to maintain behavior over time even if reinforcement is withdrawn or with concurrent aversive stimulation (eg. a slot machine gambler)
Negative reinforcement Positive punishment Negative punishment Interval Ratio Fixed Variable
MSFA stresses the fluid, developmental nature of an individual’s learning history and examines how an individual changes through the contexts they experience and the consequences of their behavior within those contexts. Thus one discrete A:B:C sequence may become part of the antecedents for the next A:B:C sequence: for example, the stimulus that functions as a consequence in the first ABC also functions as a discriminative stimulus in the second ABC (see Figure 6.1).
A:
B:
C:
A:
B:
C:
A:
B:
C:
Figure 6.1 Relationship between antecedents, behaviour and consequences in an MSFA sequence.
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
134
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Such analyses use the form A:B:C rather than A> B > C to indicate that it is not necessary to establish causal relationships within an A:B:C sequence, but that the events in question tend to occur in a particular temporal order and are therefore likely to have some form of functional relationship. When functional relationships between As, Bs and Cs are considered it is not necessarily just cause and effect relationships that are of interest but an understanding of the relationship between the independent and dependent variables and how variation in the former relates to variations in the later (Skinner, 1953). As stated above, the goal of functional analysis is prediction and influence with precision, scope and depth. In a forensic context, this would mean that any analysis of a repertoire of offending behavior should allow us to predict the circumstances and contexts under which reoffending is likely to occur. Because an analysis of the offending behavior should allow us to identify the function of behavior, a good MSFA should also facilitate the identification of institutional offending and offense paralleling behaviors. A particular problem that arises when carrying out an MSFA in forensic settings is that it can be difficult to identify the correct controlling stimuli and the functional relationships between them. To complete an analysis, the assessing clinician often has to rely on collateral file information or narrative accounts that might be incomplete or inaccurate for a multitude of reasons. In order to aid the identification of the correct variables, it is important that some degree of analysis and assessment is ongoing. For example, if particular variables are identified as being controlling stimuli, then it is important to look for the presence or absence of those stimuli within the offender’s current and other previous environments. If the stimuli are present but no behavior is observed, or if the behavior is present and the stimuli are absent, then this suggests that there may be an inaccuracy in the analysis or that there is an additional contextual factor that has been missed. In this regard, carrying out an analysis is an ongoing dynamic interaction with the case material and client behavior.
CASE STUDIES: BOB AND LEON Two illustrative case formulations are presented: “Bob” and “Leon”, who both killed other men through inflicting multiple stab wounds. The case formulations show that during early childhood, both men had similar developmental experiences but that their developmental pathways diverged during adolescence. This nevertheless resulted in offenses which, while topographically similar, also have key functional differences that impact on treatment pathways and institutional risk management. Both cases are heavily disguised and incorporate common clinical case material based on many years of working with and assessing mentally disordered violent individuals.
Childhood and Early Secondary School Developmental Experiences Bob and Leon were both in their mid-twenties when they committed their index offenses and both men gave accounts of unsettled childhoods in which they described poor relationships with their mothers and abusive relationships with
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
135
their stepfathers. Bob gave an account of having been sexually abused during puberty by one of his mother’s boyfriends. Leon did not disclose any sexual abuse but gave an account of severe physical abuse from his stepfather, who also subjected him to harsh criticism and instilled in Leon the idea that he should be tough and that real men always fight and win. When Leon was bullied at school, his stepfather beat him and made him go out and “beat up” the boys who delivered the bullying. Table 6.3 summarizes this.
Table 6.3 Functional analysis sequence 1: Childhood and early secondary school developmental experiences. Bob
Leon
A
Distal Antecedents Experience of poor parenting Sexual abuse at hands of stepfather Proximal antecedents/establishing operations Criticized by teachers for having poor concentration in school SD Bullied by other children at school
Distal Antecedents Experience of poor parenting Proximal antecedents/establishing operations “Respected” but feared by other children at school for his violence and bullying Few friendships: associations with other children focused around his bullying SD Physical abuse at hands of stepfather
B
Overt Socially isolates self and withdraws Covert: Mixed/ambivalent feelings about abuser Violent fantasy of taking revenge on abuser Ambivalent feelings about sexual orientation Increased social anxiety
Overt Uses aggression against other children Covert: Anxieties/inhibitions about violence habituated Use of violent fantasy of taking revenge on abuser
C
Attempts at social interaction are punished through lack of peer recognition/low social status (negative punishment) and bullying (positive punishment) Social isolation keeps him safe from bullies (negative reinforcement)
Increased status/peer recognition (positive reinforcement) Financial gain (positive reinforcement)
Key learning
Cannot gain peer recognition Avoidance of others keeps you safe Other males are a threat to him He is inadequate and not deserving of friendship
Social and material gains from violent behavior Being a bully protects him from being a victim
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
136
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Both men indicated that, as boys, they had struggled at school with problems of concentration and, particularly in Leon’s case, problems with conduct. Consequently, both men indicated that they had experienced educational failure in junior school and both consequently struggled to form effective peer relationships. Bob described himself as being withdrawn, somewhat isolated and vulnerable to bullying. Withdrawal and isolation may have been negatively reinforced for Bob by avoidance of the aversive social stimulus. From Leon’s account, it would appear that he also had very few if any close friends. Despite this, Leon gained some status within his peer group because of his aggression, bullying and willingness to fight other pupils, no matter what their size or reputation. Leon recalled that, even at this early age, he gained some material benefit from his violence including acquiring possessions and influencing other children to do things for him. Indirectly, there was a further gain for Leon as a result of his aggression in that he gained his stepfather’s approval, which in turn reduced the frequency of his own beatings. Thus, for Leon aggression and bullying others may have been positively reinforced by the acquisition of tangible items and increased social status and negatively reinforced by reduction in the beatings from his stepfather. Both men reported ambivalent relationships with their abusers. Leon indicated that he was frightened of his stepfather and could recall the fear associated with his stepfather returning from drinking binges and the anticipation of violence directed at both him and his mother. Yet, despite this, he also reported respect for his stepfather who had some social standing within the community because of his reputation as a “local hard man”. Leon admitted that achieving and bettering his stepfather’s social standing was something he actively aspired to do. Despite reporting he respected his stepfather, Leon also described anger toward his stepfather and disclosed a history of violent fantasy involving taking revenge on his abuser. Bob, on the other hand, disclosed how his sexually abusive stepfather was in some ways good to him and his mother and how he could be funny and engaging. Bob recalled how his abuser would comment on Bob’s penis becoming erect during the abuse and would tell him that he was special and the abuse was their special secret. Although Bob disclosed that he found the abuse physically painful, and feared his stepfather’s attentions, his physiological responses to the abuse made him feel ashamed and confused about his own sexual orientation. Bob indicated, however, that after his abuser left the family and he became aware of the true nature of what had been done to him, he also (in common with Leon) experienced revenge fantasies. Bob reported that he became fearful and suspicious of other men, particularly those he believed to be homosexual. Bob now admits he had mistakenly connected homosexuality to sexual abuse.
Adolescent Development It would appear from the initial analysis set out above that both Bob and Leon entered adolescence and secondary school with an underlying fear of being a victim, albeit with a different repertoire of topographically different behaviors to
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
137
manage this fear. Bob reported that this fear led to him being cautious in forming friendships with others. This seems to have marked Bob out as different to the other children who began to bully and ostracize him and compounded his anxiety and led to a further reduction in his social behavior. In effect, his already limited social behavior was subjected to a combination of positive punishment in the form of bullying of a physical and emotional nature and negative punishment in the form withdrawal of social attention from others. For Bob this led to an increase in his avoidance behaviors, which functioned to help him escape from the aversive stimulation of being bullied and rejected by others (negative reinforcement). For Bob because of the level of aversive stimuli at school, the school environment may have eventually become a conditioned discriminative stimulus (SD) for eliciting a fear response in the form of physiological arousal with associated fearful cognitions and subsequent avoidance behaviors. In line with these anxious responses, it can be hypothesized that this response may also have generalized from specific bullies to other children of whom Bob became increasingly fearful and avoidant of, both in and out of school. These problems were compounded by his poor educational attainment, which also elicited aversive responses from his teachers and teasing from the other children. He developed a pattern of increasingly avoidant behavior, isolating himself and truanting from school. These coping strategies, while preventing him from being bullied, also stopped Bob forming meaningful social relationships. This created a negative feedback loop where Bob’s isolation, his inability to trust others, and his negative self-concept began to self-perpetuate. Bob indicated that an important formative experience for him occurred while he was truanting when he was picked up by some youths who were out of school and intent on stealing a car. Bob went along with them and was involved in the car theft, but looking back felt that he had been set up and that the other boys were probably motivated to involve him in the theft by the wish to ‘take the piss’. After stealing the car, which was wrecked, Bob was the only one who was caught. Although Bob was dealt with through a community sentence, this experience further lowered his social standing at school and he felt he was then made fun of for his incompetence as a car thief and for being gullible and allowing himself to be set up by the other boys. Leon, in common with Bob, also experienced educational failure and struggled with basic skills, such as literacy and numeracy. Leon also truanted from school, but, unlike Bob, he truanted with a peer group, some willing and some bullied into accompanying him. While truanting, he began to associate with older boys who were involved in minor criminal activities including burglaries, car thefts and drug misuse. Leon reflected that he had felt more accepted by this deviant peer group and had gained status for his violence and risk-taking criminal activities. Perhaps inevitably, Leon became increasingly well known to the police through his criminal activities and received a variety of community disposals for car theft and burglary before eventually being given a custodial sentence in a Young Offenders Institution (YOI). Leon acknowledged that he was certainly no longer of high status in his first YOI, where he mingled with young men who were tougher and more experienced than he was; however, the experience seems to have
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
138
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
further desensitized him to violence and further embedded him within a deviant criminal peer group. From a behavior analytic perspective two things may have happened here. Firstly, through repeated exposure to violence Leon habituated to the associated physiological arousal. Secondly, Leon’s violent behavior was also being reinforced by others. Whilst tangible reinforcement may not always be forthcoming, it was hypothesized that intermittent reinforcement in the form of a variable ratio schedule would be enough to maintain this behavior. Due to his own change in status and the lack of respect and status initially received within the YOI, Leon increased his criminal offending within the YOI by getting involved in racketeering and collecting and enforcing debts for higher status prisoners. From a behavioral analytic perspective the acquisition of new criminal behavior is evident. This new behavior was hypothesized to be maintained by tangible reinforcers, such as money or tobacco1 and other social reinforcers. In this environment, there would be little reinforcement for conventional, nondeviant behavior, which was in any case likely to have been at a relatively low response rate for Leon at this stage. The status given to Leon from the other prisoners further reinforced Leon’s antisocial behavior. On his release, Leon’s offender status had increased, with conventional social relationships and legitimate employment being of little value to him. In behavioral analytic terms, there are two classes of behavior being affected. First, his conventional social behavior is at a weak level and instances of positive social behavior unlikely to be reinforced and might have been punished. Second there were likely to have been concurrent schedules of reinforcement operating on his deviant and criminal behavior operating.
Summary It can be seen in the analyses set out in Table 6.4 that Bob and Leon’s paths have diverged at this point. A major hypothesis here is that Bob was unable to gain recognition and status with his peer group. His one significant attempt to join in with a peer group when stealing a car resulted in the extremely punitive and frightening experience of getting caught and the increased the level of ridicule he experienced as a result. Bob developed a set of negatively reinforced avoidance behaviors as a means of avoiding the aversive stimulation of his peer group. By contrast, Leon was becoming more physically dominant and increasingly enmeshed with a deviant peer group; however, for Leon although there was a unity against external forces within the peer group, he constantly felt that he had to justify and maintain his position within the hierarchy, typically with increasing levels of violence and aggression. This inevitably made Leon a high status target for other criminals and led to him carrying a concealed weapon. For example, he stated that “I always had to watch my back in those days.”
1 In
many correctional institutions tobacco is used as a currency in the absence of actual money
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
139
Table 6.4 Functional analysis sequence 2: Adolescent development. Bob
Leon
A
Previous learning set out in Functional analysis 1 (Table 6.3) Establishing operation Absence of peer approval S-d Encouraged to steal car with peers
Previous learning set out in Functional analysis 1 (Table 6.3) Establishing operation Previous violent offending reinforced by peers S-d Sees opportunity to commit burglary
B
Skips school and steals a car
Commits burglary
C
Is abandoned by peers and is caught alone, and charged – acquires criminal record (positive punishment) Ridiculed by peers and is further isolated (positive and negative punishment)
Is caught and charged, acquires criminal record (initial weak positive punishment) becomes associated with praise by his peers for criminal record (positive reinforcement)
Key learning
Other people can not be trusted and will seek to harm him He is no good at anything and is “worthless” in the eyes of others
Crime gets you status so the more crime committed the greater the status achieved You don’t get caught often so it is worth the risk
Further Adolescent Experiences Bob From adolescence onwards, Bob became increasingly socially anxious around others (see Table 6.5) and continued to use avoidance strategies as a means of coping with this anxiety and thus increased his social isolation. During this time of his life, Bob continued to develop fantasies of taking revenge against his stepfather for the sexual abuse inflicted upon him and began to use alcohol as a coping mechanism in a number of different contexts. Mostly, he reported that he began to use alcohol to increase his self-confidence in social situations. Alcohol use in this case was being concurrently reinforced on negative and positive reinforcement schedules; negatively reinforced in that there was a reduction in anxiety in social situations and positively reinforced in that it facilitated an increase in his social contact with other people. During this time, however, Bob also began to use alcohol for the secondary function of emotional regulation, namely to reduce or “block out” memories and the associated anxiety and feelings of vulnerability and defectiveness resulting from his sexual abuse. Nevertheless, Bob reported that his alcohol abuse did not have the desired effect, and he reflected upon two side-effects of this strategy. First, he noted that alcohol began to strengthen the aversive affect he experienced and made him more likely to experience revenge fantasies; the strength of these fantasies was related to
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
140
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Table 6.5 Functional analysis sequence 3: Further adolescent development and origins of offence pathway. Bob
Leon
A
Previous learning set out in functional analysis 2 (Table 6.4). Establishing operations Bullied and ridiculed following car theft S-d Other people and social situations function as aversive stimuli
Previous learning set out in functional analysis 2 (Table 6.4). Establishing operations Father dies in pub fight S-d Increased attention on Leon as a potential target
B
Overt: Uses alcohol as a coping strategy to deal with abuse flashbacks (covert) Attempts to be more social Goes out drinking Covert: Attempts to cope with abuse flashbacks through violent fantasy
Overt: Increases his offending and the level of violence he uses Covert: Has flashbacks to his father beating him Is upset at being reminded of the abuse
C
Alcohol acts as CS for eliciting negative affective states Is hospitalized
Gains more respect from others (positive reinforcement) Stature as a target increases creating the need for more violence Emotional upset decreases with violence (negative reinforcement?)
Key learning
There is something wrong with him People think he is gay and that means he is in danger Needs to continue to avoid people and places to keep safe
In order to keep safe, increasing levels of violence are needed
how drunk he seemed to be – until he passed out. Second, Bob’s revenge fantasies seemed to function to not just reduce his negative mood but to help him feel like a strong and powerful man. Bob also reported that the fantasies disturbed and scared him: emotional responses that he attempted to reduce with more alcohol. He reflected that back then he would attempt to drink himself out of his low mood and fantasies only to find himself in a downward spiral. Bob reflected that by the time he realized what was happening, he was probably an alcoholic. Bob also reported that, the more he used alcohol in this manner, the more likely he was to think about his abuse and engage in fantasies when drinking. From a behavioral perspective, it seems that alcohol and alcohol-related contexts began to function both as conditioned stimuli for eliciting abuse memories and associated revenge fantasies from Bob and as discriminative stimuli for alcohol-related avoidance behaviors.
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
141
A further key learning experience occurred for Bob during this period of his life. One evening whilst drinking with some acquaintances, Bob was assaulted in the street and was hospitalized as a result. In essence the experience is likely to have maintained the conditioned stimulus functions of “other people being dangerous”. As such, other people continued to elicit a fear arousal response from Bob. A key aspect of this assault for Bob was that his attacker called him a “poof”.2 Bob seems to have linked this to his own doubts about his sexuality that resulted from his sexual abuse. He began to think that other people could tell there was something wrong with him sexually and that he was vulnerable to harm as a result. During this time Bob became increasingly fearful that others could see there was “something not right” with him. He eventually began to suspect strangers in the street could also see this and suspected that the police were following him to see if he was gay. It can be hypothesized that this experience acted as an establishing operation for strengthening Bob’s avoidance behavior. In essence, the experience is likely to have increased the aversive stimulus properties of other people, leading to an increase in the negatively reinforcing value of his avoidance behavior. Another consideration at this stage is a possibly concurrent deterioration in Bob’s mental state. Given that Bob became increasingly isolated, misinterpreted environmental stimuli as aversive and became what could be described diagnostically as paranoid, the possibility that at this time Bob was in the early or prodromal stages of psychosis must be considered.
Leon Leon also had a number of functionally significant learning experiences during adolescence. Leon’s stepfather died following a fight in a pub. Following his stepfather’s death, Leon assumed the role of the man of the family. Leon reflected that it just seemed like he was expected to do this. He also reflected that he stepped into his stepfather’s shoes as a local hard man and felt that he had something to prove, given his stepfather’s reputation. As a result, Leon made additional efforts to demonstrate that he was bigger, better and stronger than his stepfather. Leon also stated that, despite the difficulties with his stepfather, he had somehow felt that he was protected from retaliation for his criminal actions as a result of his stepfather’s reputation. He stated that after his stepfather was killed, he had to prove he could make it on his own. He therefore increased his focus on offending and in particular on the levels of violence he would use. In practice this attempted escalation in his offending behavior seems to have led to further difficulties for Leon because, in order to preserve his status as a hard man, he became harder and tougher; however, whilst this kept rivals at bay to a certain degree, it also made him more vulnerable in other ways. As Leon explained: “the bigger the bastard, the bigger the payout for taking him down”. In this regard, Leon stated that at that period he always felt he was living on borrowed time and said you had to “watch you friends even closer than your enemies”. Leon also reported feeling under scrutiny for his lack of girlfriends and he believed that other people thought he was homosexual. He explained that, in 2A
pejorative term in the UK and Ireland for a homosexual man
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
142
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
his world, being labeled “gay” would have been a potential death sentence. He reported that he had once smashed a beer glass into the face of an associate who made a joke about his sexuality; Leon reported that this was the last time until the index offense that someone made such remarks to his face. In the context of Leon’s learning history, this behavior can be understood in the following A:B:C analysis. The antecedent in this context is likely to have been the joke about being gay. Leon’s behavior included perceiving that being seen as gay would be a signal for other people to attack him. In this regard, being labeled was a signal for impending danger. The consequences of the violent assault (the behavior) on this associate were that the threat was removed and his violence was negatively reinforced by the termination of the aversive social behavior/threat from others. During this period Leon rapidly gained an even stronger reputation for being a local hard man and as an enforcer for local gangsters. His reputation was further solidified by a number of short stints in prison/young offender institutes.
Summary While both men used coping strategies that appear topographically different (Bob primarily uses alcohol, whilst Leon uses violence and offending behavior), there are a number of functional similarities between them that are notable. Both men used negatively reinforced avoidance strategies as a means of surviving in their respective environments: social isolation for Bob and a criminal underworld for Leon. Paradoxically the longer-term consequences of both Leon and Bob’s strategies seems to have been to inadvertently increase contact with the aversive stimulation both were attempting to avoid, thus eliciting further problematic avoidance behaviors of a similar nature; for Bob maintaining his use of alcohol and for Leon maintaining his use of aggression and offending behavior.
Adulthood and Development of Offense Pathway Bob As Bob’s alcohol use and social isolation continued, he experienced deterioration in his mental health to the extent that he was sectioned under the Mental Health Act and admitted to a psychiatric ward. After an initial assessment, Bob was detained under the Mental Health Act for further assessment, was diagnosed with a psychotic illness and placed on antipsychotic medication and antidepressants. During his stay in hospital, Bob disclosed his history of being sexually abused; however, Bob reported that he did not receive the response he had hoped for. Bob perceived that the staff did not believe him and considered his report to be delusional, choosing to increase his medication rather than offering other help or support. (See Table 6.6 for a summary of adult development.) After his discharge from hospital, Bob was placed back in his own accommodation community mental health services where he felt increasingly vulnerable and isolated. He had gained weight as a side-effect of his medication and his neighbours were aware he had been in a psychiatric hospital and, according to Bob,
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
143
Table 6.6 Functional analysis sequence 4: Development of offence pathway 2. Bob
Leon
A
Previous learning set out in functional analysis 3 (Table 6.5) Distal antecedent Hospitalized SD Encouraged to “open up” by staff
Previous learning set out in functional analysis 3 (Table 6.5) Distal antecedent Incarcerated S for pro-social behavior Surrounded by senior criminal peers SD for offending Surrounded by senior criminal peers
B
Discloses history of sexual abuse
Carries on offending in prison
C
Disclosure not believed (absence of reinforcement) Medication increased (positive punishment) Remains on the periphery of his social group Told to make more efforts to be social able (Sd for further behavior – likely under aversive control)
Increased status in prison (positive reinforcement) Reduced criminal status on return to the community (withdrawal of reinforcement, Thinning out of schedule)
Key learning
Other people can not be trusted Needs to use alcohol if he is to be more sociable
Criminal behavior will need to be increased to regain status
avoided him more than usual and sniggered when he walked past. His front door was vandalized with graffiti which referred to his mental health. Bob found it increasing difficult to cope with his environment and took a medication overdose which he reported was a suicide attempt. In the context of the history presented above it can be hypothesized that this suicide attempt functioned as an effort to ultimately terminate the aversive stimulation in Bob’s environment. Whilst Bob had not attempted suicide before, and by definition completed suicide can have no reinforcement history, this behavior can still be understood from a behavioral analytic perspective. Skinner (1953) argued that suicide is a form of self-regulatory behavior, which, while having no complete reinforcement history, does have a generalized reinforcement history for its component parts of self-regulation and negative reinforcement. Bob’s self-harm resulted in a further brief stay in hospital, before he was discharged back to his own accommodation and to the care of a community mental health team. Bob’s community nurse worked actively with Bob to increase his social skills and social integration, encouraging Bob to be more socially active and going to the extent of putting Bob in contact with others in a similar situation to himself. Bob reported that he tried hard to make contacts, but could only do so with the aid of alcohol, something he hid from his community nurse. While this
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
144
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
intervention was well intentioned, in practice Bob was assessed without a functional analysis of his problems having been undertaken and consequently led to suboptimal outcome. Given the previously established functions of alcohol for Bob, this intervention led to a situation where Bob was experiencing negative thoughts and emotions about his abuse and associated revenge fantasies in the company of a number of people he did not know very well and of whom he was potentially scared. Bob reported that he remained on the periphery of this social group. He explained there were some people he knew from hospital who were OK, but there were always others he did not know coming in and out of the group and social gatherings, which Bob found very unsettling.
Leon During this time, Leon’s reputation amongst his underworld peers and superiors continued to grow. He became known as someone reliable and who could get things done. He began to be offered money for contracted threats and beatings. Leon said that because of his reputation, when he threatened people with violence, they listened to him and treated him with respect because they knew he was capable of carrying out any threats made. In effect the schedules of reinforcement established in prison that strengthened and diversified his repertoire of criminal behavior now also occurred in the community. Leon also became increasingly involved in recreational violence and alcohol use, reflecting that “if there was no alcohol or fighting involved, it was not a good night out”. His use of violence escalated to him using weapons such as screwdrivers and baseball bats. During his early adulthood, Leon spent a further longer period of time in prison, where his reputation and considerable use of violence led to him achieving top dog status, stating that “even the screws3 were afraid of me”. Although he was able to achieve this status in prison, Leon reported that, on his return to his local community, he found himself in an unexpected position: On this occasion he did not return to find his status, position and reputation unequivocally enhanced. Instead he found himself in a different position. His status was if anything diminished: his criminal record, familiarity to the police and frequent periods in prison had led to his criminal peers to question his reliability and discretion and in his absence a number of other criminals had fulfilled his role to the satisfaction of various gangland bosses. Nevertheless, shortly after his return to the community, Leon stated he was offered a job that he saw a chance to re-establish himself in the gangland hierarchy. Leon reported that a senior-ranking criminal, Dave, suspected to be conducting an affair with another’s girlfriend and Leon was instructed to “warn Dave off, physically if necessary”. Leon reported that, given the rank of the target, this was a high profile job that many others were afraid to take. Therefore, he saw it as a perfect chance to re-establish his status amongst his gangland peers. The sequences of offenses 1 and 2 for Bob and Leon are shown in Tables 6.7 and 6.8.
3A
pejorative term in the UK and Ireland for a prison officer
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
145
Table 6.7 Functional analysis sequence 5: Development of offence pathway 2. Bob
Leon
A
Previous learning set out in functional analysis 4 (Table 6.6) In the pub Meets a group of acquaintances Presence of a gay man
Previous learning set out in functional analysis 4 (Table 6.6) Discharged from prison Lowered social standing Accepts high status job Consumption of alcohol Being made fun of by his victim
B
Overt Drinks a lot of alcohol Invites people back to his flat Hides a knife for his protection Covert I must be more sociable Increased sense of threat from Tom
Overt Continues to drink alcohol Covert Increasing anger at his victim Increasing discomfort at being put down in front of others Worry about his social standing
C
People come back to the flat People drink all of this alcohol He is left alone with Tom
Dave continues to disrespect him Others are witness to Dave’s disrespect
Key learning
He is in danger. It has been a set up to get him alone with Tom Tom means to cause him harm
If he does not follow through with his plans he is finished
Index Offense Bob On the night of his index offense, Bob went out drinking and met a group of acquaintances at a local bar. There were also some people there who Bob did not know very well, including one person Bob was fearful of, Tom. Bob knew that Tom was gay and he explained during interview that he was wary of him for the reasons outlined above. At closing time, Bob invited everyone back to his flat for more drinks. Bob reported that he was unsure why he had done this, but reflected that he was trying to make an effort and do things to prove his fears wrong – as suggested by his community psychiatric nurse. Once back at Bob’s flat, the group drank Bob’s alcohol and made fun of him and the spartan conditions in which he lived. After consuming all the available alcohol the group decided to go to a night club and all left with the exception of Tom who stayed in Bob’s flat, saying he wanted to finish his drink. Once alone with Tom, Bob became anxious about Tom being in his flat, having previously mislinked homosexuality and sexual abuse in his own learning history. This fear was compounded by his alcohol use that night and by him already thinking about his abuse and experiencing some revenge fantasies. Becoming increasingly fearful for his safety Bob went to the kitchen to get a knife
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
146
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Table 6.8 Offence sequence 2. Bob
Leon
A
Previous learning set out in functional analysis 5 (Table 6.7) Offence sequence 1 Tom touches him on the shoulder Smell of alcohol
Previous learning set out in functional analysis 5 (Table 6.7) Offence sequence 1 Dave goads and makes fun of Leon
B
Covert: Memory of his own experience of abuse Immediate sense of physical threat Overt Stabs Tom
Covert: Memory of his stepfather’s abuse Threat to integrity Threat to criminal progression Increased anger Overt Stabs Dave
C
Tom is killed and the threat is immediately removed (Sr− ) Bob is arrested and sectioned under the MHA
Dave is killed, the threat is removed (Sr− ) but also the chance for reward (Sr+ ) is increased
Key learning
He is safe and the threat is removed (negative reinforcement). Aversive stimulus functions of social situations and gay men are maintained and likely to be strengthened
He is safe and that violence still works as a way of keeping himself safe. Increasing levels of violence might be needed
while Tom was using to toilet which he hid under a cushion on his seat. Bob reported that he wanted Tom to leave but was underassertive and to anxious to ask him to go. This situation frustrated Bob and reminded him of the passive role he believed he took in his own abuse and elicited further thoughts about how weak he was and how others constantly took advantage of him. Bob reported that he became increasingly angry with himself and Tom as a result but nevertheless claimed that toward the end of the night he fell asleep on the sofa only to be woken by Tom who was saying goodbye before he left. The smell of alcohol on Tom’s breath, the closeness of his face and the touch on Bob’s shoulder triggered a flash back to Bob’s own abuse. The next thing Bob says he remembered was Tom lying on the floor having been fatally stabbed.
Leon On the night of Leon’s index offense he went out drinking with a group of acquaintances, including his victim Dave (the person Leon was to be paid for “warning off”). Leon reported that, on the night of the offense, Dave was constantly calling him “sonny” and “boy”. Leon reported that he considered this to be disrespectful and made him angry, reminding him of how his stepfather would mistreat him as
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
147
a child. On leaving the pub, Dave invited everyone back to his house to continue drinking. As the night wore on everyone left and Leon recognized that he has an opportunity to deliver the warning to Dave. Dave responded to Leon’s threat by laughing at him. Leon reported that this was something he had not experienced for a long time and he became very angry, pulling out a knife and holding it to Dave’s throat. As Leon attempted to increase the level of threat against Dave, Dave responded by laughing and ridiculing Leon, and according to Leon Dave told him that he was going to take the knife off him and was going “to stick it up Leon’s arse”, implying that Leon was gay would consequently surely enjoy the experience. Leon reported that at this point he “saw red and lost it” stabbing Dave in the neck. Leon reported that Dave stumbled back holding his neck but despite his injury continued to express contempt for Leon who then stabbed Dave repeatedly – only stopping once the blade broke.
Offense Paralleling Behaviors and Clinically Relevant Institutional Behaviors Bob From this analysis of Bob’s developmental trajectory and index offense, it can be hypothesized that the primary function of Bob’s index offense was one of self-protection and the removal of a particularly threatening and aversive stimuli from his environment. The majority of the behaviors that have been identified as functionally relevant in Bob’s case can been seen to be negatively reinforced avoidance behaviors. Bob’s history is such that other people with certain characteristics may have become conditioned discriminative stimuli for threat and danger, which Bob avoided by social isolation and self-medication with alcohol. Bob’s trajectory was relatively free of offending prior to his index offense; however, based on this analysis, it may still be possible to make predictions about Bob’s institutional behavior and risk factors for further violence. Given the documented abuse history, it can be hypothesized that prison or a male-dominated secure hospital would be laden with discriminative stimuli signalling danger for Bob. In this regard, a number of institutional behaviors were observed in Bob that were hypothesized to be relevant to his offending behavior. Initially Bob refused to wash and shower within his secure environment. Given Bob’s history of mental health difficulties, this was viewed by some as a recurrence of a depressive episode or negative symptoms signifying a psychotic relapse; however, based on the analysis above it can also be hypothesized that this presentation may be connected to Bob’s previously conditioned threat sensitivity and ideas about homosexuality. Washrooms and showers in the secure setting were semi-communal. In the context of Bob’s learning history, the showers and the potential to be naked around a number of other male strangers was likely to have acted as a conditioned aversive stimulus for Bob, which, as expected, he attempted to avoid in line with his learning history. Attempts to encourage Bob to behave differently independent of his previous learning could actually be counter-therapeutic and place him and others at risk.
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
148
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Leon While the index offense for Leon had both topographical and functional similarities to Bob’s, it can be hypothesized that there were significant functional differences. Whilst Bob’s offense has been hypothesized to be a function of a prolonged and ingrained negative reinforcement schedule, Leon’s index offense can be thought of as being a function of a mixed negative and positive reinforcement schedule, in which increasing criminality and violence served to increase status and financial reward (positive reinforcement) and remove threats from his environment (negative reinforcement). While Leon’s violence has served to remove threats from his environment, such threats were usually a consequence of his previous violence. Therefore, it has been hypothesized that the most important reinforcement for Leon’s violence has been the financial reward or status. Therefore, within the prison system it can be hypothesized that Leon was likely to continue this pattern of behavior and engage in behaviors such as intimidation of others and bullying and racketeering, which he did attempt to do in the early stages of his incarceration.
DISCUSSION As this chapter highlights, forensic case formulation presents a number of unique challenges. Given the criticisms of formulation outlined previously – particularly that it is subjective, lacking in both consistency and validity and is thus unreliable (Kuyken et al., 2005), clinicians need to find a way of enhancing explanations of a forensic patient’s behavior. MFSA provides a vehicle in which to do this. As discussed in the introduction to this chapter, MSFA was originally described in the early nineties (Gresswell and Hollin, 1992). The more recent literature supports the use of functional assessment and analysis by clinicians working with aggression and violence (Daffern et al., 2007; Jones, 2004). Both Daffern et al. (2007) and Jones (2004) highlighted the importance of functional assessment and analysis in improving clinician understanding and consequently assessing risk and providing treatment. Furthermore, in arriving at a psychologically relevant formulation critical to the understanding of the patient’s actions, the clinician is required to determine the role of thought and action, and ultimately the function of the behavior leading to the offense in question. In this chapter two men have been described whose index offense behaviors are topographically similar in many ways and yet whose diagnoses, presentations, and forensic histories are very different. Furthermore, and perhaps more importantly, the treatment efforts and risk issues presented are even more at odds with each other. In both of the cases described, a difference in the function of behavior can be identified, the motivation for which is uncovered by applying MSFA to unpick the issues. MSFA can also sit within a wider formulation. For example, considering the case of Leon, the function of each of his behaviors can be considered at a number of different levels or stages. So, the understanding of the function and purpose of behavior, can sit within a wider systemic formulation to include the family, peer
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
149
group, education system and support system each of which is considered in the prior learning.
Treatment Implications In relation to intervention, management whilst in an institution and reducing the risk of future offending, it is ever more important for clinicians to defend their opinions and provide the appropriate treatments. Psychologists have recognized for a long time that to provide appropriate psychological treatments it is unwise to simply look at the end behavior without knowing what drives it. In the cases of Bob and Leon, on the surface they look similar save for the differences in forensic history. Both men experienced some form of early abuse by a stepfather, are poorly educated, used alcohol prior to the offense, had poor family relationships and generally had impaired social relationships. Both men stabbed their older male victims with in frenzied attacks in a private home within the context of having consumed alcohol and with a background of fantasy of revenge on their childhood abuser stepfather. The function of the behavior in both cases was to remove a perceived threat, with the functions established in both men’s early history of abuse. Surely, then, given these similarities, treatment considerations should be much the same? This is not the case. When considering the context/environment within the functional analysis at the different stages, changes in patterns of behavior in both cases are uncovered. In both cases, the behavior of the victim in that particular situation was the trigger to ignite the emotional response driving the overt behavior. By understanding the emotional responses (fear and anger), the function behind the resulting behavior can be better understood. In Bob’s case it was initially fear then anger and rage related to flashbacks of his stepfather and desire for revenge. In Leon’s case it was initially an instrumental plan that turned into rage following his perception that the victim was treating him disrespectfully and with contempt, triggering an emotional reaction of anger and rage and desire for revenge. In both cases, the behavioural shift happened following this trigger. For Leon, this was an unusual response to violent confrontations which were usually carried out with confidence and self-assurance. In Bob’s case this violence was unusual in that he had never committed a violent offense prior to or following this one. What may be of importance is that Bob had successfully avoided these contexts prior to the night of his index offense. Leon, however, had been in similar situations before without such a loss of control. Leon’s recent loss of criminal status and the need to recapture this was an important establishing operation in his case. In considering treatment for these two individuals, some clear similarities as well as important differences can be found. Both might be considered to need support around interpersonal relationships and the development of social skills, although both men are likely to need different sets of social skills. While it may appear that both men may also benefit from improving their self-esteem, the above MSFA would indicate that this would be a simplistic view. Both men have very different needs in this respect. Bob probably has little self-esteem, whereas Leon gets his from criminogenic sources. In this case, Bob’s deficit may be easier to work
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
150
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
with by establishing and shaping a behavioral repertoire to aid this. Leon, on the other hand, will be more difficult to work with given the significant history of reinforcement for his criminal behavior in the past. Both men might also benefit from developing skills to deal with their anger experiences resulting from their abuse histories. Both Bob and Leon need to resolve their own issues and feelings related to their stepfathers who abused them – indeed this is imperative considering the identified trigger in both cases involved a degree of stimulus generalization from the abuse situation to the one of the index offense – leading both of them to kill. Notably, though, this work would be very different in each case. Here, however, the similarities end. Bob’s treatment needs include work around vulnerability and safety as well as psycho-education in relation to mental health experiences (psychosis). He also needs to learn about the impact of alcohol on his emotional and mental health functioning. Finally, Bob needs to understand the situational or contextual factors that make him more likely to become a risk to others (real or perceived threat) as well as his lack of assertiveness and vulnerability. Bob’s risk factors include his level of fear of others. Bob might, therefore, benefit from graded exposure to aversive stimuli along with social and assertiveness skills training to allow him to stay in and tolerate aversive situations for longer so that he may learn that the threat is not great, thus extinguishing his fear response. Teaching Bob social skills would also increase the chances of successful social interactions (i.e., positive reinforcement of alternate behavior). Leon’s treatment needs to focus more on the criminogenic factors, such as previous instrumental violence and his perception of vulnerability to others if he does not maintain his status. Leon’s previous behaviors indicate an underlying acceptance of violence as an appropriate means of gaining respect, approval and cash, something borne out by his experience of the world. A treatment program for Leon will need to include a means to shape up and reinforce alternative methods of gaining these rewards in a prosocial manner; for example, Leon could be encouraged to work in the prison in roles that require some form of increasing responsibility, such as work that requires cooperating with others. Given the very low base rate of such behaviors, Leon is likely to need ecologically valid social reinforcement which may be difficult to engineer in a prison environment: normal praise and attention from staff may be aversive for him (and therefore act as punishers). Some form of tangible reinforcer that has a positive function during his incarceration (e.g., increases in privileges, for example) may be effective but would need to be well managed and clearly contingent upon a reduction of offending behaviors within the institution if his bullying is not to be accidentally maintained. A further key factor in Leon’s offending has been the lack of empathy for others, with the distress of others failing to act as inhibiting stimuli for violent behavior. This in itself has treatment implications for Leon. Many offender treatment programs contain victim empathy sections. Those working with Leon will need to be aware that attempting to increase Leon’s empathy for others may be difficult, given the very low rate of base behavior and the different stimulus functions the distress of others is likely to serve for Leon; it can be hypothesized above that the distress of others is likely to function as a generalized positive reinforcer for Leon.
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
May 28, 2011
8:22
Printer Name: Yet to Come
FORMULATION OF SERIOUS VIOLENT OFFENDING
151
Thus, the importance of the context becomes clear in the commission of the violence in both cases. Given what is known of Bob and of the functional assessment carried out, it may be considered unlikely that he would carry out such an offense again in the absence of the specific contextual factors that arose at the time. In Leon’s case, murder can perhaps be considered a progression from the many violent assaults which he carried out in the past. He has gained much that is important to him from these assaults which reinforce the necessity of doing it again; however, this was Leon’s first murder and consideration needs to be given to whether the progression would have happened without the contextual factor reminiscent of his stepfather’s abuse. What is clear is that, in terms of further violence, the functional assessment indicated a higher risk for Leon in terms of his reinforcement history because of the high likelihood of key discriminative stimuli being present within his environment in the future.
CONCLUSION Scrutinizing behaviors in the way MSFA requires raises questions that would otherwise have been missed and allows hypothesis-driven testing of such ideas. It helps provide a scientific basis for the formulation method, one that is systematic and defensible and less likely to leave scope for wide variability. Understanding the function of a given behavior is arguably imperative in any case formulation, especially one that leads us to decision making with regards treatment and risk. The versatility of MSFA means that it can be applied to all situations whether community forensic or inpatient cases. MSFA can help conceptualize assaultative behavior within institutions and allow patterns to emerge. Carrying out MSFA with each situation over time will build up a comprehensive picture to allow intervention points to emerge. Furthermore, it can be applied to individual offenses carried out by one individual, allowing an understanding to develop in relation to consistency and changes in patterns of offending behavior. Offense paralleling behaviors are beginning to gain credence in terms of what they can tell us about an offender’s motivation to offend. By identifying offense paralleling behaviors the clinician can often see the different functions of similar appearing behavior across different contexts. Discrete behavioral episodes may not carry the weight of importance often ascribed to them once entrenched within a MSFA. Ultimately, it is hoped that use of such an analysis will help improve the quality and accuracy of individual formulations improving long term the possible outcomes for these individuals.
REFERENCES Braham, L., Jones, D. and Hollin, C.R. (2008) The Violent Offender Treatment Program (VOTP): development of a treatment program for violent patients in a high security psychiatric hospital. International Journal of Forensic Mental Health, 7, 157–72. Daffern, M., Ferguson, M., Ogloff, J.R.P., Thomson, L. and Howells, K. (2007) Appropriate treatment targets or products of a demanding environment. The relationship between
P1: OTA/XYZ P2: ABC JWST073-06 JWST073-McMurran
152
May 28, 2011
8:22
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
aggression in a forensic psychiatric hospital with aggressive behaviour preceding admission and violent recidivism. Psychology, Crime and Law, 13, 431–41. Ferster, C.B. and Skinner, B.F. (1957) Schedules of Reinforcement. New York: AppletonCentury-Crofts. Gordon, A. and Wong, S.C.P. (2000) Violence Reduction Program: Facilitator’s Manual. Department of Psychology, University of Saskatchewan. Gresswell, D.M. and Hollin, C.R. (1992) Towards a new methodology for making sense of case material: an illustrative case involving attempted multiple murder. Criminal Behaviour and Mental Health, 2, 329–41. Harris, G.T. and Rice, M.E. (2007) Characterising the value of actuarial violence risk assessments. Criminal Justice and Behaviour, 34, 1638–58. Hart, S.D., Michie, C. and Cooke, D.J. (2007) Precision of actuarial risk assessments. British Journal of Psychiatry, 190, 60–5. Jones, L. (2004) Offence paralleling behaviour (OPB) as a framework for assessment and intervention for interventions with offenders. In A. Needs and G. Towl (eds), Applying Psychology to Forensic Practice (pp. 34–63). Oxford: BPS, Blackwell. Kuyken, W., Fothergill, C.D., Musa, M. and Chadwick, P. (2005) The reliability and quality of cognitive case formulation. Behaviour, Research and Therapy, 43(9), 1187–1201. Quinsey, G., Harris G., Rice, M. and Cormier, C. (1996) Violent Offenders: Appraising and Managing Risk. Washington, D.C.: American Psychological Association. Skinner, B.F. (1938) The Behaviour of Organisms. New York: Appleton-Century. Skinner, B.F. (1953) Science and Human Behaviour. New York: Macmillan. Webster, C.D., Douglas K.S., Eaves, D. and Hart, S.D. (1997) HCR-20: Assessing Risk for Violence (Version 2). Burnaby, BC: Mental Health Law and Policy Institute, Simon Fraser University. Wilson, K.G. and DuFrene, T. (2009) Mindfulness for Two: An Acceptance and Commitment Therapy Approach to Mindfulness in Therapy. Oakland, CA, New Harbinger.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
Chapter 7
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER ELLEN VEDEL Jellinek Substance Abuse Treatment Center, Amsterdam, Holland, The Netherlands
PAUL M.G. EMMELKAMP University of Amsterdam, Holland, The Netherlands
INTRODUCTION A relatively large number of substance abusing patients will come into contact with the criminal justice system. The relationship between substance use disorder and criminal behavior is complex and in the treatment of substance abusing and offending patients a thorough case formulation is needed to insure that the different problem areas, substance use, criminal behavior but also associated concurrent disorders are assessed properly and that the right screening and diagnostic tools are used. In the context of such a case formulation clear hypothesis are formulated about how these problem areas are related, which behaviors should be targeted first and which specific interventions are probably most effective. This chapter starts with a short overview of the literature on the relationship between substance abuse, criminal behavior and associated problems. Different assessment tools are discussed, after which a case is presented of a male substance dependent patient with a history of criminal and violent behavior.
Substance Use Disorders and Criminal Behavior Although there is consensus in the literature about a relationship between substance abuse and offending, different interpretations of the nature of the relationship abound, varying according to the substances used, the specific sample studied, such as normal populations (McKinney et al., 2010), forensic populations Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
154
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
(Kraanen, Scholing and Emmelkamp, 2010), substance abusing populations (Chermack et al., 2009), and the social context (Erickson, Macdonald and Hathaway, 2009). Instances of reckless criminal behavior may be directly due to intoxication and in severely dependent substance abusers crime may be a means to obtain money for financing their substance use. Also, underlying vulnerabilities and personality pathology have been held responsible for both criminal behavior and substance use disorders. For example, there is some evidence from longitudinal studies that experiencing childhood abuse in youth is a significant risk factor for not only alcohol and drug use (Burnette et al., 2008), but for crime and violence as well (Thornberry et al., 2010).
Substance Use Disorders and Violence In the US, 40–50% of all murders, over 50% of rapes and over 50% of deaths resulting from car accidents are alcohol-related. Together with bipolar disorder and psychosis, substance use disorders are associated with relative high prevalence rates of interpersonal (Corrigan and Watson, 2005) and partner violence (Kraanen, Scholing and Emmelkamp, 2010; Moore et al., 2008). In the case of alcohol, violence can often be explained by the ingestion of alcohol before the violent offense (H˚aggard-Grann et al., 2006). Cannabis use may be connected to violence not by the direct intoxicating effects of the drug but rather via withdrawal from the substance. There is some evidence that during withdrawal from cannabis subjects report greater irritability. However in the case of young marijuana dependent patients, violent offenses can best be explained by a juvenile history of conduct disorder (Arseneault et al., 2000). Also in sexual violence there is a clear relationship with substance abuse. In a recent review, it was concluded that about half of sex offenders had a history of substance abuse, a quarter to half had a history of alcohol misuse and that about one-fifth to a quarter had a history of drug misuse (Kraanen and Emmelkamp, 2011). Furthermore, about a quarter to half of the sex offenders appeared to be intoxicated at the time of the offense. Another aspect of violence is violence toward the self. Suicidal behavior and substance abuse are closely linked. In a review of studies on suicide in alcohol dependence, Inskip, Harris and Barraclough (1998) found that substance abusers were seven times more likely than nonsubstance abusers to commit suicide. In opiate users and mixed intravenous drug users, the risk of committing suicide was found to be even greater (Wilcox, Conner and Caine, 2003). This high prevalence of suicidal behavior in patients with substance use disorders can partly be explained by the high prevalence of other psychiatric disorders which are associated with suicide, like depression. However, it is not only the presence of the comorbid disorder on its own, which explains the elevated risk for suicide. Depressed patients who abuse alcohol are more likely to commit suicide compared to non-abusing depressed patients (Dumais et al., 2005). Clinicians should be particularly alert for the increased risk of suicide not only with depressed patients, but also with impulsive substance abusing patients (often with borderline or antisocial traits.) It is wrong to assume that impulsive aggression in substance abusers is limited to violence
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
155
toward others (e.g., intimate partner violence); many suicides are committed as an impulsive act when intoxicated.
Substance Use Disorders and Personality Disorders A substantial number of substance abusers share personality characteristics that predispose them to criminal behavior, such as antisocial personality disorder. In recent years a great deal of research attention has been given to the relationship between personality disorder and substance abuse, most notably with borderline personality disorder and antisocial personality disorder. In the National Comorbidity Survey (NCS) community study, the odds ratio for comorbidity between antisocial personality disorder and substance and drug use disorders were 11.3 and 11.5 respectively (Emmelkamp and Vedel, 2006; Emmelkamp and Kamphuis, 2007). The median reported prevalence rates of substance abuse disorder in borderline patients is 67% and the median prevalence rate of borderline personality disorder within substance abuse populations is 18% (van den Bosch et al., 2002). The emphasis in this chapter is on a substance abusing patients diagnosed with antisocial personality disorder. Patients with antisocial personality disorder tend to disregard the rights of others and are prone to unethical behavior. Antisocial personality disorder is often associated with drug and alcohol problems and criminal behavior. People with antisocial personality disorder tend to be irresponsible, and do not “learn” from previous mistakes. For example, they are often unable to keep a job or meet adult financial responsibilities. As a result of low frustration tolerance, antisocial personality disorder patients can be quite aggressive and impulsive. More often than not, remorse and guilt are absent for the negative consequences their behavior may have for others; in fact, some may derive pleasure from the suffering of others. There are also more cunning and planful variations, however the formal classification of antisocial personality disorder in the Diagnostic and Statistical Manual of Mental Disorders (American Psychiatric Association, 2000) heavily emphasizes the antisocial behaviors, which sets it apart from the concept of psychopathy. There is an extremely high comorbidity between substance abuse and antisocial personality disorder, higher than with any other mental disorder (Emmelkamp and Vedel, 2006). It should be noted that in a number of substance dependent patients the diagnosis of antisocial personality disorders is not warranted. Although these patients formally fulfill the necessary criteria, their antisocial behaviors may be the result of their substance use disorder, rather than the result of a personality disorder (Emmelkamp and Vedel, 2006). Many clinicians hold that substance dependent patients with antisocial personality disorder have poorer treatment outcomes compared than patients without antisocial personality disorder. Earlier studies supported this notion, but poor prognosis and response to treatment may have been confounded by pretreatment severity. Substance abusing patients with antisocial personality disorder often use more alcohol and drugs and have more associated legal and psychiatric problems at baseline compared to substance dependent patients without antisocial personality disorder (Verheul et al., 1998). Results of studies with substance dependent patients indicated that those with co-occurring antisocial personality disorder were
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
156
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
more disabled at baseline and follow-up, but improved as much as the patients without antisocial personality disorder (Cacciola et al., 1995; McKay et al., 2000). In a review of treatment outcome studies of opiate dependent drug users meeting the criteria for comorbid antisocial personality, few differences were observed for those with and without antisocial personality disorder receiving treatment for opioid dependence with respect to retention, reductions in HIV risk behaviors, and drug use (Havens and Strathdee, 2005). However, in those continuing to use cocaine and benzodiazepines while enrolled in a methadone program, response to psychotherapy and pharmacological therapy was lower among opiate users with comorbid antisocial personality disorder.
Assessments in Substance Use Disorder Screening Instruments A number of screening instruments for substance use problems have been developed. For alcohol use disorders, the Michigan Alcoholism Screening Test (MAST; Skinner and Sheu, 1982) or the Alcohol Use Disorder Identification Test (AUDIT, Babor et al., 1992) can be used. The AUDIT has also been adapted to include drug use (AUDIT-ID, Babor et al., 2001; Campbell et al., 2004). An alternative drug screening instrument is the Drug Abuse Screening Test (DAST, Skinner, 1982). In adolescents, the Drug Use Screening Inventory (DUSI-R; Tarter, 1990) can be used. One subscale of this instrument, the Violence Proneness Scale (VPS), predicts aggressive events three years (Kirisci, Tarter and Reynolds, 2009) to seven years later (Tarter et al., 2002).
Structured Interviews The Time-line follow-back interview (TLFB; Sobell and Sobell, 1996) is a semistructured interview and uses a retrospective method for assessing alcohol and drug use patterns and related events and possesses adequate reliability. The TLFB procedure estimates daily alcohol/drug use through the employment of various memory aids, such as a daily calendar, key dates (e.g., birthdays, personal events, clinic appointments, visits from family), anchor points ([un]employment, illnesses, holidays), and other memory aids to facilitate recall. The TLFB provides useful information about antecedents and consequences of heavy drinking and illicit drug use and of high-risk situations for relapse, which may be of help for treatment planning. The Addiction Severity Index (ASI; McLellan et al., 1992) is a semistructured interview and provides a comprehensive assessment of substance use (history, frequency, and consequences of alcohol and drug use). In addition, it assesses family history, psychological symptoms, health problems, and legal issues. ASI scores on the major domains may be used to plan treatments targeting these domains. The ASI is available free of charge, and takes roughly 45–60 min to administer.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
157
Measuring Dependence and Impaired Control To evaluate the degree of alcohol dependence, the Severity of Alcohol Dependence Questionnaire (SADQ, Stockwell et al., 1979) may be used. A score of 31 or higher on the 20-item SADQ correlated with clinicians’ ratings of severe dependence, while a score of 30 or less correlated with ratings of mild to moderate dependence. Related to the degree of alcohol dependence is impaired control over drinking, which can be assessed with the Impaired Control Scale (ICS: Heather et al., 1993). This instrument showed good evidence of reliability, concurrent validity (Heather, Booth and Luce, 1998; Marsh et al., 2002) and predictive validity (Heather et al., 1998). The clinical value of the ICS and the SADQ include their use in deciding the recommended drinking goal of treatment: moderate drinking versus abstinence.
Measures to Assess High-risk Situations, Motives, and Coping A number of paper and pencil measures are particularly useful for the purpose of conducting functional analyses and understanding patterns of substance use. These instruments offer the clinician a profile of high-risk situations, and motives or reasons patients may endorse for substance abuse. This information can provide guidelines tailored to each individual patient for how to manage high-risk situations. The Inventory of Drinking Situations (IDS; Annis, 1982) and the Inventory of Drug-Taking Situations (IDTS; Annis and Martin, 1985; Turner, Annis and Sklar, 1997), which assess specific antecedents of alcohol and drug use. Motives or reasons for drinking may also provide important information for treatment planning. Measures include the 14-item Reasons for Drinking Scale (Farber, Khavari and Douglass, 1980), and the 15-item Drinking Motives Measure (Cooper et al., 1992), which contains subscales for social, coping, and positive affect enhancement motives. The following measures can be helpful in selecting inadequate specific coping skills, which may be targeted in treatment: the Situational Confidence Questionnaire for alcohol abusers (Breslin et al., 2000), and for substance abusers (Barber, Cooper and Heather, 1991). These measures were developed to assess substance abusers’ confidence in their ability to resist urges to use. The Drug-Taking Confidence Questionnaire (DTCQ; Sklar, Annis and Turner, 1997) assesses coping self-efficacy for a number of different types of drug and alcohol use.
Assessment of Comorbid Axis I and Axis II Disorders The Structured Clinical Interview for DSM–IV Axis I (SCID-I; First, Spitzer, Gibbon and Williams, 1995a) can be used to confirm substance use disorder diagnosis and to establish whether there are comorbid disorders. Two measures can be used for assessing personality disorders: (1) the Personality Diagnostic Questionnaire-4 (PDQ-4; Hyler, 1994) and (2) the Personality Assessment Inventory (PAI; Morey, 2007). Both measures, however, have the tendency to overrate, thus it is clinically wise to confirm a positive diagnosis using the Structured Clinical Interview for DSM–IV Axis II (SCID-II; First et al., 1995b). It is important to note that many psychiatric symptoms can be the direct result of intoxication/substance abuse. It
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
158
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
is advisable to assess comorbidity after a prolonged period of abstinence (rule of thumb: at least four weeks abstinence). In sum, when not working in the field of substance abuse treatment intake should always including screening for problematic alcohol and/or drug use. If a patient screens positive, quantity, frequency and the negative consequences of his or her substance use should be assessed. Preparing for treatment several further analyses of the problem behavior are needed. For example, assessing concepts such as selfefficacy and impaired control can give valuable additional information. Regarding the assessment of other Axis I and Axis II disorders, caution should be exercised when the patient is still active abusing substances given the likelihood of the psychiatric symptoms being substance induced.
CASE FORMULATION The case presented here describes the treatment of a patient referred through the criminal justice system to our substance abuse treatment center. The patient received a regular intake of two sessions, which is standardized practice at out institution. The intake was used to gather information to allocate the patient to the right substance abuse treatment intensity (outpatient, day hospital or inpatient treatment) based on stepped care principles. After this intake, the patient was referred to his therapist who used the first phase of treatment to increase motivation for change and to gather further information about the problem behaviors to enable a case conceptualization. Planning and practicing behavioral change (decreasing levels of substance use) often parallels the phase of gathering information about the problem behavior (substance use and craving) and is not postponed until after the making of a functional analysis. Tom (44 years of age) was referred through the criminal justice system to receive treatment for his substance use disorder after having been arrested in connection to armed robbery. The treatment in the substance abuse treatment center was mandatory, in exchange for early parole. Tom had a very long substance abuse history; he started using cannabis and alcohol at the age of 12 and by the time he was 17 he was using heroin, speed and cocaine as well. Parallel to his substance abuse, Tom started his criminal career. He dropped out of school at the age of 15, engaging in all kind of criminal activities. At the age of 20, Tom was arrested for robbery and assault and went to prison for the first time. After being released, Tom continued his criminal career and continued abusing substances. At the age of 26, Tom met Paula, who was also a substance user, with whom he tried to settle down. During this period Tom and Paul had two children. Both Tom and Paula were unsuccessful in cutting down on their substance use. Eventually, pressured by child welfare services, Paula’s sister and her husband got custody of the children. During the following years, Tom’s substance use became more severe and he became more and more aggressive. After a severe incident of domestic violence, Tom and Paula split up. A few months after their split-up, Tom – under the influence of drugs and alcohol – tried to commit suicide by driving his car off the road. At the hospital where he was admitted, Tom was seen by psychiatrist and diagnosed with
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
159
antisocial personality disorder. In the years to follow, Tom repeatedly got arrested for assault and other criminal activities.
Information Gathered at Intake The Addiction Severity Index was used to assess Tom’s history, frequency and the negative consequences of his alcohol and drug use. In addition, family history, psychological symptoms, health problems, and legal issues were assessed. Before his last incarceration, Tom’s substance use was characterized by daily use of large amounts of alcohol and cocaine (snorting) and incidental use of cannabis. Tom had stopped his use of heroin and speed (amphetamine) several years ago. He reported problems with anger management, and financial and housing problems: after being released from prison he was staying with friends and was unable to find work. Tom was interviewed in detail about his angry outbursts in the past. In all incidents, he had been heavily under the influence of substances; during periods in which his substance use was significantly less (e.g., in jail) he reported no physically violent behavior. Tom’s probation officer was present at one of the two intake sessions and Tom was given clear information to what extent his probation officer would be informed about disclosures during treatment. The meeting together with the probation officer was also used to explore to what extent the probation officer and/or our outpatient treatment facility could support Tom in finding a job, vocational training, and stable housing. It was decided that Tom and his probation officer, together with one of our social workers, would focus on housing and finding a job. Treatment would primarily focus on changing Tom’s substance use and on his anger management.
Increasing Motivation for Change In many psychiatric conditions, like anxiety or mood disorders, the treatment goal is often clear from the start – the patient wants to get rid of his/her panic attacks or depressed mood. Although the treatment interventions to achieve change are often negotiated (as in the case of exposure in vivo), the treatment goal itself is seldom a topic of discussion. In the treatment of substance use disorders, patients often enter treatment without the wish to get rid of the problem behavior (i.e., stop drinking or drug taking behavior); rather, they desire to moderate their substance use, or reduce the harmful/negative consequences of their use without stopping use altogether. This ambivalence about behavior change is in many ways comparable with the ambivalence or low motivation observed in patients entering forensic treatment facilities. The “old school” therapeutic way of dealing with “unmotivated” substance abusing patients was a confrontational approach, which was thought to be necessary to overcome the resistance of the patient, the ‘pathological denial’ of substance abuse, and the perceived inherent lack of motivation about changing substance abuse. These characteristics were often seen as inherent qualities of the patients themselves. Around 1980, a new style of interviewing substance abusing people who were ambivalent about change emerged (Miller, 1983). The clinical
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
160
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
method of Motivational Interviewing was developed specifically to work through this ambivalence and to enhance intrinsic motivation for change. Motivational Interviewing combines a supportive and empathic counseling style with a directive method for resolving ambivalence in the direction of change. Thus, this therapeutic approach integrates relationship building principles of nondirective therapy (Rogers, 1961) with active behavioral strategies directly related to the patient’s stage of change according to the model of Prochaska and DiClemente (1982) and has been found moderately effective in motivating patients to change substance abuse (Emmelkamp and Vedel, 2006). In the case of Tom, it may not come as a surprise that he was not very enthusiastic about his mandatory substance abuse treatment. During the first phase of treatment, motivational interviewing techniques were used to increase intrinsic motivation for change. We focused on the three critical components of motivation: being ready, willing and able. “Willing” stands for the extent to which a person (in this case Tom) wants, desires or wills changes in substance use and aggressive behavior. “Able” stands for the confidence the person has that change is possible (general efficacy) and the confidence a person has that he is able to make that change (self-efficacy). “Ready” refers to the extent to which the person gives change priority above other issues. One of the motivational techniques used to explore Tom’s ambivalence was the ‘the balance sheet’. Tom was asked to list out the advantages and disadvantages of change versus the advantages and disadvantages of not changing (Miller and Rollnick, 2002). Because the advantages and disadvantages of change may vary per substance, Tom was instructed to make separate balance-sheets, one for cocaine and one for alcohol. Tom’s balance sheet also helped to explore the relationship between his substance use and his criminal lifestyle and how changing his substance use may influence his criminal lifestyle (see Tables 7.1 and 7.2).
Table 7.1 Balance sheet – alcohol. Advantages of using
Disadvantages of using
Short-term: - Better able to relax - Makes me feel ‘alive’ Long-term:
Short-term - Getting into trouble with you and my probation officer - Having hangovers Long-term: - Not able to lead a normal life - Bad for my health
Advantages of change
Disadvantages of change
Short-term: - Keeps the probation officer of my back - Saves me money Long-term: - Better for my health - Will be able to have stable contact with my children
Short-term: - Not able to blow off steam - Can’t hang out with my friends Long-term: - Life will become boring
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
161
Table 7.2 Balance sheet – cocaine. Advantages of using
Disadvantages of using
Short-term: - Gives a tremendous Kick/very nice feeling - Don’t feel anxious while being on the job/ doing my thing. Long-term: - None
Short-term - Paranoia - Need other substances to counter the negative side effects of cocaine - Getting in trouble with you/the treatment clinic and probation officer - Expensive Long-term: - Eventually it will kill me - Will not be there for my children - Very expensive, have to resort to crime to finance this way of living - Getting into jail
Advantages of change
Disadvantages of change
Short-term: - Keeps the probation officer of my back - no paranoia Long-term: - Will be able to get a new lifestyle - Saves money
Short-term: - None Long-term: -Have to leave my old lifestyle behind
Building a Therapeutic Relationship The first phase of treatment was not only used to increase motivation for change but was also use to increase the quality of the working alliance/therapeutic relationship between Tom and this therapist. A number of patients with antisocial personality traits are characterized by a fearful attachment style (Timmerman and Emmelkamp, 2006). This attachment style is characterized by avoidance of close relationships because of fear of rejection, a sense of personal insecurity and a distrust of others. Most of them have a history of early traumatic experiences (Zanarini and Gunderson, 1997) in which parental figures often play a causal role in violating their trust in them. Further, in an addictive and often criminal milieu, distrusting others might be of major importance to ‘survive’. Further, many patients with a criminal history feel unfairly treated by the criminal justice system. Difficulties in the treatment of individuals with antisocial personality traits often have to do with fear of trusting therapists and fear of showing vulnerabilities. Individuals with antisocial personality traits are, in general, reluctant to be open about their emotions and the problems they encounter. Acting in a brutal, manipulative, or avoidant manner is often done in order to keep others at an emotionally safe distance. For clinical practice, this implies that effort will be needed to build a good therapeutic relationship and a safe environment in which the patient with antisocial personality traits dares to be open about the difficulties he is facing. Even the smallest suspicion of unreliability on the part of therapists can immediately undo the trust that was built (Timmerman and Emmelkamp, 2006).
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
162
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
During the first phase of treatment, we tried to enhance the working alliance between Tom and his therapist by keeping sessions primarily focused on current substance use and associated problems. Tom was not prompted for emotional disclosure nor was he extensively questioned about his background (being a child of alcoholic parents), his life with Paula, and his children.
Monitoring the Problem Behavior Tom’s willingness to change his substance use and aggressive outbursts increased as a result of the motivational techniques and the problem-focused working alliance, so the time was right to introduce the self-monitoring diary. Monitoring problem behavior helps to elucidate the conditions under which the behavior occurs. Diary information can illuminate crucial associations between the problem behavior and critical events (antecedents and consequences of the problem behavior). In the treatment of substance use disorders, self-monitoring of daily craving and substance use enables the patient and the therapist to identify specific recurring situations, thoughts, or feelings that elicit craving and to identify the positive and negative consequences of the problem behavior. This information can later be use to construct a functional analysis. In general, such diaries address the following questions: 1. 2. 3. 4. 5. 6. 7.
Which day and what time was it (external cue)? What was the situation you were in (external cue)? What were your thoughts while you were in this situation (internal cue)? How did these thoughts make you feel (internal cue)? Did you experience any bodily sensations (internal cue)? What was the intensity of your craving, between 0–100 (internal cue)? What happened next, how did you cope (did you use substances and if yes what and how much?) 8. What were the consequences of this coping behavior? Because of Tom’s anger issues, we added a second diary, one specifically monitoring feelings of anger (between 1 and 100) and monitoring aggressive behaviors (e.g., yelling, throwing things, fighting). Contrary to our expectations, Tom was quite motivated to monitor his craving, substance use and anger. In addition, we asked Tom to complete three questionnaires: the Inventory of Drinking Situations (IDS; Annis, 1982), the Inventory of Drug-Taking Situations (IDTS; Annis and Martin, 1985;Turner, Annis and Sklar, 1997), and the Drug-Taking Confidence Questionnaire (DTCQ; Sklar, Annis and Turner, 1997).
Setting Goals for Treatment Together with Tom we discussed the specific goals for treatment. Tom had become increasingly motivated to changing his cocaine use (and recreational use of cannabis) and was prepared to strive for abstinence for at least the next 12 months;
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
163
however, regarding his alcohol use, Tom remained ambivalent and wanted to strive for controlled use. In the literature, there is much debate about whether controlled alcohol use is a realistic treatment goal (Emmelkamp and Vedel, 2006). The fact is that some patients are able to reduce their alcohol intake to nonharmful levels and that, from a motivational perspective, demanding abstinence as the only realistic treatment goal is therapeutically unwise but you may have to recommend abstinence, e.g., if you have a seriously physical ill dependent drinker. In the case of Tom we proposed a ‘behavioral experiment’, which is an effective way of letting patients discover from their own experience whether moderation is a realistic goal. We explained about acceptable limits to drinking behavior, both in terms of the number of drinks per day and the number of alcohol-free days a week. As a general guideline, we used the following limits: not more than three drinks a day and a minimum of three alcohol free-days per week. In addition, patients are discouraged from drinking in high-risk situations because this will interfere with learning new coping skills, as well as enhance the likelihood of lapsing. If a patient is unable to keep the agreed goals in assignments, the belief endorsed by the patient that he or she is able to control drinking is empirically rejected. Difficulties encountered in such a guided attempt at moderation can lead to increased motivation for abstinence. Tom wanted to be able to drink during the weekends, the limit of “only” three units (a standardized unit is defined as containing 10 milliliters of pure alcohol/ ethanol) per day, however, he found unacceptable. We negotiated a maximum of 5 units on Friday and 5 units on Saturday for a period of two weeks. During this behavioral experiment, Tom was asked to continue monitoring his craving and substance use using the diary. During these two weeks, Tom was unable to limit his alcohol use to 5 units per night, and, during several alcohol binges, Tom lapsed into cocaine use as well. Tom was very disturbed by the fact that even when “putting his mind into it” he was unable to control his drinking, and he was even more distressed by lapsing into cocaine use. We used the next session to renegotiate goals for treatment and Tom committed himself to abstinence from alcohol for a period of 12 months.
Functional Analysis Using the self-monitoring diary, with recordings of craving and substance use, and the two questionnaires, a functional analysis was constructed. A functional analysis is a hypothetical working model of the problem behavior (e.g., substance abuse) from which specific treatment interventions are derived. The key questions that we attempted to answer in the functional analysis were the following: (1) What are the situations in which craving and substance use occur? (2) Which responses (emotional, physiological, cognitive, overt behaviors) occur? and (3) What are the consequences of substance use that reinforce use? There are different kinds of consequences that can reinforce substance use. The most obvious are positive rewards, for example an increase in self-esteem, and negative reinforcement, for example a decrease in anxiety or depressed mood. Because antecedent and consequences
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
164
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION EXTERNAL CUES
Situation Being home alone Having been in contact with government officials/or agencies (authority figures) Specific days or times of the day Evenings and weekends most difficult
Specific places Specific pubs
Specific people
Other specifics
Some drinking pals
Alcohol ads of TV
⇓ INTERNAL CUES Thoughts They are out to make my life miserable They want to get me down
Feelings Angry
Bodily sensations Tense muscles
⇓ CRAVING Physical manifestations Dry throat
Cognitive manifestations I need a drink now!
Behavioral manifestations Restless
⇓ BEHAVIOR What substance Alcohol
How much Between 10–15 units
How is it used Oral
⇓ CONSEQUENCES
Short-term
Long-term
POSITIVE Decrease in tension Increase in feeling capable of facing the troubles ahead
NEGATIVE Increased craving for cocaine Loss of control
None
Health problems Disturbs relation with children
Figure 7.1 Functional analysis of alcohol use.
may vary by substance, we made a functional analysis by substance (see Figures 7.1 and 7.2). Tom’s diary revealed a close link between increased craving for alcohol and high levels of anger (internal cue) which was associated with interacting with specific people (external cue). Cocaine craving was clearly associated with being under the influence of alcohol, feeling bored and negative self-talk (internal cues).
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
165
EXTERNAL CUES Situation Going out for a walk with no specific plan of where I want to go / what I want to do Seeing other people surrounded by luxury (expensive cars, restaurants)
Specific days or times of the day Evenings Weekends
Specific places
Specific people
The inner-city
Meeting/talking to friends “from the old life”.
Other specifics Being under the influence of alcohol
⇓ INTERNAL CUES Thoughts This is not fair I won’t be able to change my ways and become a “normal” person
Feelings Anger
Bodily sensations Headache
⇓ CRAVING Physical manifestations Tense muscles Sensations in stomach
Cognitive manifestations Just using once won’t be a big problem
Behavioral manifestations Restlessness
⇓ BEHAVIOR What substance Cocaine
How much 1–3 grams
How is it used Snorting
⇓ CONSEQUENCES
Short-term
Positive Physical arousal Feeling superior No negative thoughts about the future/past
Long-term
Figure 7.2 Functional analysis of cocaine use.
Negative Start drinking even more alcohol
Increase paranoia Increase likelihood of aggressive behavior • Need money; will resort to stealing/dealing
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
166
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Treatment Plan Because Tom had no substance abuse treatment experience, we introduced an evidence-based cognitive behavioral treatment (CBT) program for substance use (for review see Emmelkamp and Vedel, 2006) and we added an anger management module. The treatment rationale of the CBT program was explained to Tom, namely that from a cognitive behavioral perspective, substance abuse and dependence is defined as a habitual, maladaptive method for attempting to cope with the stresses of daily living. This maladaptive way of coping is triggered by internal and external cues and reinforced by positive rewards and/or avoidance of punishment (Monti et al., 1989). The treatment thus focused on overcoming skill deficits. Different techniques are used to increase the ability to detect and cope with high-risk situations that commonly precipitate relapse. Treatment is characterized by monitoring problem behaviors and cognitions, behavioral practice, and homework assignments.
Avoiding Cues versus Learning to Cope Differently Overall there are two main strategies that target the antecedents of the problem behavior. Tom could try to avoid cues that trigger substance use or could learn to deal differently with high-risk situations. Avoiding strategies are in general easier to apply compared to learning new coping skills but, although being an effective short-term solution, not all high-risk situations can be avoided and some are so strongly intertwined with living conditions that avoiding them permanently is in fact unrealistic. Together with Tom, we listed all situations that were linked with an increase in craving or an elevated risk for lapsing to alcohol or cocaine use. We discussed which of these situations Tom was able and willing to avoid.
Coping Skills: Handling Craving Almost all patients experience craving in response to changing their drinking or drug taking habits, although its intensity and frequency differ greatly among individual patients (and substances). Tom reported frequent and intense craving in his diary recordings. Because craving is closely linked to (re)lapse and sometimes highly disturbing to patients, it is important to target craving early on in treatment by educating patients about the phenomenon, addressing distorted beliefs about craving (Tom: “If I experience craving, this is a sign of weakness and lack of motivation”), and teaching new coping skills. We introduced the five main strategies of coping with craving: distraction, social support, recalling the negative consequences of substance use, supportive self-talk, and mind-surfing. Tom was prompted to experiment with all these strategies under different circumstances and found distraction (taking a long walk, watching a movie, taking a shower) and recalling the negative consequences of substance use (thinking of his time in jail, how his substance use damaged the lives of his two children) the most effective. Although Tom effectively used these skills to handle craving, his level of alcohol
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
167
craving remained high. We discussed the option of anticraving medication as an add-on to the psychological treatment and Tom was referred to one of our medical doctors. Tom was prescribed naltrexone (anticraving medication) which had a positive influence on his level of craving.
Coping Skills: Handling Anger We used an adaptation of an anger management treatment protocol developed for substance abusing and/or mental health patients as an add-on to our regular CBT substance abuse treatment program (Reilly and Shopshire, 2002). Tom was asked to make a balance sheet of the pros and cons of aggressive behavior versus the pros and cons of change (not reacting in an aggressive fashion). We discussed with Tom the issue of when anger becomes a problem and the myths about anger (e.g., anger automatically leads to aggression, anger is an effective way of getting what you want). We used Tom’s diary recordings to identify events associated with anger build-up and identified Tom’s physical, behavioral, emotional and cognitive cues when feeling angry. We introduced the time-out procedure and other strategies to enable behavioral control when feeling angry. We also introduced relaxation training but Tom was not very fond of this particular intervention and refused to rehearse in-between treatment sessions.
Relapse Prevention During treatment, Tom encountered many high-risk-situations. Often he was quite able to cope effectively, but sometimes he lapsed into substance use. This is a common phenomenon; in fact, most patients lapse during the course of treatment and many experience relapses during or after finishing treatment. In 1985, Marlatt and Gordon presented their very influential cognitive behavioral model of the relapse process. The model gives a detailed classification of factors or situations that can precipitate or contribute to relapse episodes. The key elements of this model are seemingly irrelevant decisions, possible poor coping skills, possible low selfefficacy and the abstinence violation effect. The model postulates that entering a high-risk situation is often preceded by seemingly irrelevant decisions. These small decisions or rationalizations apparently have no direct link with direct alcohol or drug use, however, they enhance the likelihood of entering a high-risk situations. In response to entering a high-risk situation, a patient may effectively use coping skills that will increase self-efficacy and thus decrease the likelihood of relapse. If coping skills are not used or not used effectively, self-efficacy will decrease and the positive outcome expectancies for the effects of alcohol or drugs will increase. This will enhance the likelihood of lapse, which will trigger the abstinence-violation effect. The abstinence-violation effect refers to feelings of guilt, shame and failure in reaction to a lapse, which instead of strengthening the patient to remain abstinent triggers thoughts of the perceived positive effects of alcohol or drugs use and thus enhances the likelihood of renewed alcohol or drugs use in order to cope with these negative thoughts and emotions.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
168
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
We used Tom’s diary records to identify seemingly irrelevant decisions and Tom’s negative self-evaluation associated with the abstinence-violation effect. We constructed a relapse prevention plan to cope effectively in the case of a lapse to prevent a full relapse which consisted of: (1) things Tom could do during the lapse to prevent full relapse; (2) identify helping thoughts that would counter negative self-talk during the lapse; (3) plan whom Tom could ask for social support during and after the lapse to prevent full relapse; and (4) plan specific behaviors after the lapse to prevent relapse. We used lapses during treatment to test the effectiveness of the relapse prevention plan and to make adjustment.
Role Play We extensively used role-play during the treatment sessions. We wanted to prepare Tom for situations in which, were he to be offered alcohol or drugs, the likelihood that he would be able to withstand temptation or social pressure would be increased. During role-play, we not only rehearsed the verbal component of refusing but also paid attention to body language and the behavioral expression of refusing (for example, by walking away). In addition to skill enhancement, we also scanned for possible dysfunctional beliefs that would interfere with practicing refusal skills (e.g., “They’ll think I’m a loser”).We also used role play to practice nonaggressive assertiveness and – very important to Tom – how he effectively could interact with his teenage children and their caretakers (his ex-wife’s sister and her husband).
Treatment Progress Three months into treatment, Tom had a 4 day relapse into alcohol and cocaine use. This relapse was triggered after several frustrating interactions with a housing agency. The relapse paralleled a motivation crisis and Tom did not show up for several treatment sessions. We chose to set the CBT program on hold and return to motivational interviewing techniques to explore Tom’s renewed ambivalence. During these sessions, Tom opened up more about feeling down, being pessimistic about the future and feeling guilty about things from the past. Tom confirmed that dysphoric mood had frequently triggered craving. Subsequently, we focused on helping Tom to recognize and identify negative cognitions that were associated with his depressed mood and how his depressed mood was associated with specific behavior (inactivity). We combined cognitive restructuring with reinforcement for constructive and pleasant behaviors. We listed out activities Tom used to enjoy but had ceased doing, and scheduled these as homework assignments. After 6 months in treatment, Tom was finally accepted at a combined housing and vocation training program for ex-convicts. This special program had a zerotolerance policy regarding alcohol and drug use and, although Tom found this policy highly annoying, it stimulated him to remain abstinent, which he succeeded in doing. The anger management module helped Tom to cope more effectively when feeling angry: during treatment there were several incidents of anger build-up and verbal aggression but no incidents of physical aggression.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
169
At the end of treatment, Tom’s probation officer started raising questions about the validity of the diagnosis antisocial personality disorder. Our observations were that Tom was avoidant in disclosing emotions, to some extent bitter and cynical about life and the intentions of others, but Tom also felt genuine remorse about behaviors in the past, including not being able to take care of his children. We concluded that at least for the past year, Tom had not met criteria for antisocial personality disorder (using the antisocial personality disorder section of the Structured Clinical Interview for DSM–IV Axis II, SCID-II). We also asked Tom to fill out the Drug-Taking Confidence Questionnaire (DTCQ; Sklar, Annis and Turner, 1997). Compared to the beginning of treatment, Tom showed an overall strong increase in self-confidence to withstand tempting situations.
Follow-up We had three follow-up sessions with Tom, one, three and six months after treatment. We used the TLFB method to assess drinking, drug use and violent behavior. During the second follow-up period, Tom relapsed into alcohol use after leaving the housing project and moving into a new apartment of his own. Tom contacted us and we scheduled 4 booster sessions to help Tom quit drinking again and handle this new situation with more liberties compare to living in the ex-convict housing project. Tom successfully remained abstinent during the third follow-up period. During the total 10 month follow-up period, Tom reported two incidents of verbal aggression while under the influence of alcohol; in several other incidents Tom had been able to use self-control (leaving the situation) to prevent escalation.
CONCLUDING REMARKS A number of evidence-based treatments are currently available for the treatment of substance use disorders, motivational interviewing and cognitive behavioral interventions being the most thoroughly tested and most effective (Emmelkamp and Vedel, 2006). The present case formulation shows that these treatments were also effective in a forensic case. Given the high prevalence of substance abuse in offenders (Kraanen et al., 2010; Kraanen and Emmelkamp, 2011), it is surprising to see that well-controlled studies using these therapies in violent offenders are lacking. Further, although these treatments are effective in a number of substance abusing patients, results in the long run are mixed; unfortunately lapses and relapses in substance abuse is rather common. In violent offenders, relapse in substance abuse may also lead to relapse in violence (Hirschel, Hutchison and Shaw, 2010; Mignone, Klostermann and Chen, 2009). There is a clear need of studies into treatment for substance abuse including regular booster sessions for substance abusing offenders. As illustrated in the description of this case, a case formulation based on a thorough assessment of the problem behaviors and the functional analyses of these behaviors is of paramount importance in the treatment of substance abuse in forensic cases. Although a number of treatments for substance-abuse are empirically supported, results cannot be generalized to forensic patients. Here, a thorough case conceptualization is a condition sine qua non.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
170
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
REFERENCES American Psychiatric Association (2000) DSM-IV-TR: Diagnostic Statistical Manual of Mental Disorder (5th edn). Washington, DC: APA. Annis, H.M. (1982) Inventory of Drinking Situations (IDS). Toronto: Addiction Research Foundation. Annis, H.M. and Martin, G. (1985) Inventory of Drug-Taking Situations (IDTS). Toronto: Addiction Research Foundation. Arseneault, L., Moffitt, T.E., Caspi, A., Taylor, P.J. and Silva, P.A. (2000) Mental disorders and violence in a total birth cohort: Results from the Dunedin Study. Archives of General Psychiatry, 57, 979–86. Babor, F., Higgins-Biddle, J.C., Saunders, J.B. and Monteiro, M.G. (2001) The Alcohol Use Disorders Identification Test. Guidelines for Use in Primary Care (2nd edn). World Health Organization Department of Mental Health and Substance Dependence. Babor, M., Hofmann, F.K., Del Boca, V. et al. (1992). Types of alcoholics: I. Evidence for an empirically derived typology based on indicators of vulnerability and severity. Archives of General Psychiatry, 49, 599–608. Barber, J.G., Cooper, B.K. and Heather, N. (1991) The situational confidence questionnaire. International Journal of Addiction, 26, 565–75. Boles, S.M. and Miotto, K. (2003) Substance abuse and violence. A review of the literature. Aggression and Violent Behavior, 8, 155–74. Breslin, F.C., Sobell, L.C., Sobell, M.B. and Agrawal, S. (2000) A comparison of a brief and long version of the situational confidence questionnaire. Behaviour, Research and Therapy, 38, 1211–20. Burnette, M.L., Ilgen, M., Frayne, S.M. et al. (2008) Violence perpetration and childhood abuse among men and women in substance abuse treatment. Journal of Substance Abuse Treatment, 35, 217–22. Cacciola, J.S., Alterman, A.I., Rutherford, M.J. and Snider, E.C. (1995) Treatment response of antisocial substance abusers. Journal of Nervous and Mental Diseases, 183, 166–71. Campbell, T.C., Catlin, L., Bentzler, J. et al. (2004) Test-retest reliability of the Alcohol Use Disorders Identification Test: Revised to include other drugs (AUDIT-ID). Alcoholism: Clinical and Experimental Research, 28, S110. Chermack, S.T., Murray, R.L., Winters, J.J. et al. (2009) Treatment needs of men and women with violence problems in substance use disorder treatment. Substance Use and Misuse, 44, 1236–62. Cooper, M.L., Russell, M., Skinner, J.B. and Windle, M. (1992) Development and validation of a three-dimensional measure of drinking motives. Psychological Assessment, 4, 123–32. Corrigan, P.W. and Watson, A.C. (2005) Findings from the national comorbidity survey on the frequency of violent behavior in individuals with psychiatric disorders. Psychiatry Research, 136, 153–62. Daughters, S.B., Sargeant, M.N., Bornovalova, M.A. et al. (2008) The relationship between distress tolerance and antisocial personality disorder among male inner-city treatment seeking substance users. Journal of Personality Disorders, 22, 509–24. Dumais, A., Lesage, A.D., Alda, M. et al. (2005) Risk factors for suicide completion in major depression: A case-control study of impulsive and aggressive behaviors in men. American Journal of Psychiatry, 162, 2116–24. Emmelkamp, P.M.G. and Kamphuis, J.H. (2007) Personality Disorders. New York: Psychology Press/Taylor & Francis. Emmelkamp, P.M.G. and Vedel, E. (2006) Evidence-based Treatment for Alcohol and Drug Abuse. Routledge/Taylor & Francis, New York Erickson, P.G., Macdonald, S. and Hathaway, A.D. (2009) Profiling violent incidents in a drug treatment sample: a tripartite model approach. Journal of Drug Issues, 9, 741–58. Farber, P.D., Khavari, K.A. and Douglass, F.M. (1980) A factor analytic study of reasons for drinking: Empirical validation of positive and negative reinforcement dimensions. Journal of Consulting and Clinical Psychology, 48, 780–1.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION, SUBSTANCE ABUSE DISORDERS, AND ANGER
171
First, M.B., Spitzer, R.L., Gibbon M. and Williams J.B.W. (1995a) Structured Clinical Interview for DSM-IV Axis Disorders – Patient Edition (SCID-I/P, Version 2.0). New York: New York State Psychiatric Institute. First, M.B., Spitzer, R.L., Gibbon M. et al. (1995b) Structured Clinical Interview for DSM-IV Axis II Personality Disorders – Patient Edition (SCID-II, Version 2.0). New York: New York State Psychiatric Institute. ¨ Fridell, M., Hesse, M., Meier Jæger, M. and Eckart Kuhlhorn, E. (2008) Antisocial personality disorder as a predictor of criminal behavior in a longitudinal study of a cohort of abusers of several classes of drugs: Relation to type of substance and type of crime. Addictive Behaviors, 33, 799–811. ¨ N. and Moller, ¨ H˚aggard-Grann, U., Hallqvist, J., L˚angstrom, J. (2006) The role of alcohol and drugs in triggering criminal violence: A case cross-over study. Addiction, 101, 100–8. Havens, J.R. and Strathdee, S.A. (2005) Antisocial personality disorder and opioid treatment outcomes: A review. Addictive Disorders and their Treatment, 4, 85–97. Heather, N., Booth, P., and Luce, A. (1998) Impaired Control Scale: Cross-validation and relationships with treatment outcome. Addiction, 93, 761–71. Heather, N., Tebbutt, J.S., Mattick, R.P. and Zamir, R. (1993) Development of a scale for measuring impaired control over alcohol consumption: a preliminary report. Journal of Studies on Alcohol, 54, 700–9. Hirschel, D., Hutchison, I.W. and Shaw, M. (2010) The interrelationship between substance abuse and the likelihood of arrest, conviction, and re-offending in cases of intimate partner violence. Journal of Family Violence, 25, 81–90. Hyler, S.E. (1994) PDQ-4 – Personality Diagnostic Questionnaire. New York: New York State Psychiatric Institute. Inskip, H.M., Harris, E.C. and Barraclough, B. (1998) Lifetime risk of suicide for affective disorder, alcoholism and schizophrenia. British Journal of Psychiatry, 172, 35–7. Kirisci, L., Mezzich, A.C., Reynolds, M. et al. (2003) Prospective study of the association between neurobehavior disinhibition and peer environment on illegal drug use in boys and girls. American Journal of Drug and Alcohol Abuse, 35, 145–50. Kirisci, L., Tarter, R. and Reynolds, M. (2009) The violence proneness scale of the DUSI-R predicts adverse outcomes associated with substance abuse. American Journal on Addictions, 18, 173–7. Kraanen, F.L. and Emmelkamp, P.M.G. (2010) Substance misuse and substance use disorders in sex offenders: A review. Clinical Psychology Review, 31, 478–89. Kraanen, F.L., Scholing, A. and Emmelkamp, P.M.G. (2010) Substance use disorders in perpetrators of intimate partner violence in a forensic setting. International Journal of Offender Therapy and Comparative Criminology, 54, 430–40. McKay, J.R., Alterman, A.I., Cacciola, J.S. et al. (2000) Prognostic significance of antisocial personality in cocaine-dependent patients entering continuing care. Journal of Nervous and Mental Diseases, 188, 287–96. McKinney, C.M., Caetano, R., Rodriguez, L.A. and Okoro, N. (2010) Does alcohol involvement increase the severity of intimate partner violence? Alcoholism: Clinical and Experimental Research, 34, 655–8. McLellan, A.T., Kushner, H., Metzger, D. et al. (1992) The fifth edition of the Addiction Severity Index. Journal of Substance Abuse Treatment, 9, 199–213. Marsh, A., Smith, L., Saunders, B. and Piek, J. (2002) The Impaired Control Scale: confirmation of factor structure and psychometric properties for social drinkers and drinkers in alcohol treatment. Addiction, 97, 1339–46. Mignone, T., Klostermann, K. and Chen, R. (2009) The relationship between relapse to alcohol and relapse to violence. Journal of Family Violence, 24, 497–505. Miller, W.R. (1983) Motivational interviewing with problem drinkers. Behavioural Psychotherapy, 11, 441–8. Miller, W.R. and Rollnick, S. (2002) Motivational Interviewing: Preparing People to Change Addictive Behavior. New York: Guilford Press. Monti, P.M., Abrams, D.B., Kadden, R.M. and Cooney, N.L. (1989) Treating Alcohol Dependence: A Coping Skills Training Guide in the Treatment of Alcoholism. New York: Guilford Press.
P1: OTA/XYZ P2: ABC JWST073-07 JWST073-McMurran
172
May 28, 2011
8:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Moore, T.M., Stuart, G.L., Meehan, J.C. et al. (2008) Drug abuse and aggression between intimate partners: A meta-analytic review. Clinical Psychology Review, 28, 248–75. Morey, L. (2007) Personality Assessment Inventory Professional Manual (2nd edn) Lutz, FL: Psychological Assessment Resources. Prochaska, J.O. and DiClemente, C.C. (1982) Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy, Theory, Research, and Practice, 19, 276–88. Reilly, P.M. and Shopshire, M.S. (2002) Anger Management for Substance Abuse and Mental Health Clients: A Cognitive Behavioral Therapy Manual. DHHS Pub. No. (SMA) 02-3661. Rockville, MD: Center for Substance Abuse Treatment, Substance Abuse and Mental Health Services Administration. Rogers, C. (1961) On Becoming a Person. London: Constable. Skinner, H.A. (1982) The Drug Abuse Screening Test. Addictive Behaviors, 7, 363–71. Skinner, H.A. and Sheu, W. (1982) Reliability of alcohol use indices: The Lifetime Drinking History and the MAST. Journal of Studies on Alcohol, 43, 1157–70. Sklar, S.M., Annis, H.M. and Turner, N.E. (1997) Development and validation of the drugtaking confidence questionnaire: A measure of coping self-efficacy. Addictive Behaviors, 22, 655–70. Sobell, L.C. and Sobell, M.B. (1996) Timeline Follow Back User’s Guide: A Calendar Method for Assessing Alcohol and Drug Use. Addiction Research Foundation, Toronto. Stockwell, T.R., Hodgson, R.J., Edwards, G. et al. (1979) The development of a questionnaire to measure severity of alcohol dependence. British Journal of Addiction, 74, 79–87. Tarter, R. (1990) Evaluation and treatment of adolescent substance abuse: A decision tree method. American Journal of Drug and Alcohol Abuse, 116, 1–46. Tarter, R, Kirisci, L, Vanyukov, M. et al. (2002) Prediction of violent behavior in high risk males between early adolescence and young adulthood. American Journal of Psychiatry, 159, 1541–7. Thornberry, T.P., Henry, K.L., Ireland, T.O. and Smith, C.A. (2010) The causal impact of childhood-limited maltreatment and adolescent maltreatment on early adult adjustment. Journal of Adolescent Health, 46, 359–65. Timmerman, I.G.H. and Emmelkamp, P.M.G. (2006) The relationship between attachment styles, Cluster B personality disorders in prisoners and forensic inpatients. International Journal of Law and Psychiatry, 29, 48–56. Turner, N.E., Annis. H.M. and Sklar, S.M. (1997) Measurement of antecedents to drug and alcohol use: Psychometric properties of the Inventory of Drug-Taking Situations (IDTS). Behaviour, Research and Therapy, 35, 465–83. Van den Bosch, L.M.C., Verheul, R., Schippers, G.M. and van den Brink, W. (2002) Dialectical behavior therapy of borderline patients with and without substance use problems, implementation and long term effects. Addictive Behaviors, 27, 911–23. Verheul, R., van den Brink, W. and Hartgers, C. (1998) Personality disorders predict relapse in alcoholic patients. Additive Behaviors, 23, 869–882. Wilcox, H.C., Conner, K.K. and Caine, E.D. (2003) Association of alcohol and drug use disorders and completed suicide. Drug and Alcohol Dependence, 76, S11–S19. Zanarini M.C. and Gunderson, J.G. (1997) Differental diagnosis of antisocial behavior and borderline personality disorder. In D.M. Stoff, J. Breiling and J.D. Maser (eds), Handbook of Antisocial Behavior. New York: John Wiley & Sons, Inc.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
PART III
SEXUAL OFFENDING
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
Chapter 8
SEXUAL OFFENSES AGAINST CHILDREN JAMES VESS Deakin University, Australia
TONY WARD Victoria University of Wellington, New Zealand
Assessment of individuals who commit sexual offenses requires knowledge of both the causes of sexual offending and effective means of reducing the risk of reoffending. Assessment involves the systematic collection of relevant information in order to detect clinically significant phenomena or problems in functioning, in order to provide clear treatment targets and risk management strategies. Without accurate assessment it is impossible to determine the nature of the risk an offender presents, nor the suitability and focus of intervention. Accurate assessment also allows for the subsequent determination of whether treatment has had a positive impact. In conducting assessments, practitioners must bring evidence based knowledge of sexual offenders as a population together with knowledge about a particular offender. In contemporary clinical practice this means taking into account a sex offender’s array of stable and acute risk factors. In essence, dynamic risk factors can be construed as distal, proximal, and contextual casual factors of sexual offending (Ward and Beech, 2004). The focus of this chapter is on psychological assessment and case formulation in the rehabilitation and risk management of individuals who have been convicted of sexual offenses against children. It begins with a review of the research literature on sex offender risk assessment, followed by a discussion of the relationship between risk, etiology and case formulation. Finally, a risk etiology framework for case formulation is applied to a clinical case example.
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
176
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
RISK ASSESSMENT WITH SEX OFFENDERS There is now a large and growing professional literature on risk assessment with sex offenders. As the field of risk assessment with sex offenders continues to develop, there is a need to empirically identify the best measures and methods to use (Abracen et al., 2004; Borum, 1996; Miller, Amenta and Conroy, 2005). Although there is continuing debate over the optimal utilization of static and dynamic risk factors in the detection and management of risk (Quinsey et al., 2006, vs. Hanson and Harris, 2001; Craig, Browne and Stringer, 2004), actuarial measures have demonstrated a statistically significant level of predictive accuracy regarding the risk of sexual reoffending, and consistently outperform unstructured clinical judgement (Hanson, 1998; Hanson and Thornton, 2000). While their limitations need to be recognized, the predictive accuracy of these standardized measures for sexual reoffending is now well-established and widely utilized in judicial decision-making. The primary value of risk assessment using actuarial procedures is to convey the relative likelihood of specific types of reoffending, against specific types of victims, over specified periods of time, for subgroups of offenders sharing similar levels of specific risk factors. In addressing the statistical properties of such assessments, Hanson and Thornton (2000) have noted, “most decision makers are not particularly concerned about the ‘percent of variance accounted for’. Instead, applied risk decisions typically hinge on whether offenders surpass a specific probability of recidivism” (pp. 129–30). The probability for sexual reoffending based on available empirical measures of risk may vary depending on offender characteristics such as offender age and the types of victims in their offense history. One of the limitations of most current measures in common use is that separate recidivism rates for different age groups and for child versus adult victim sex offenders are largely unavailable. Hanson and Bussi`ere (1998) reported an average sexual recidivism rate for rapists of 18.9% and a rate for child molesters of 12.7%, in contrast to Harris and Hanson (2004) who reported that rapists and child molesters had similar rates of sexual recidivism, although subgroups within the child molesters had significantly different reoffense rates. Recently reported research findings suggest that those who sexually offend against children and older offenders, as well as incest offenders and those with male victims, may sexually reoffend at different rates than younger offenders with adult victims or those that offend only against females (Skelton and Vess, 2008). A related issue is whether most sex offenders “specialize” in a specific victim type or even in sexual offending. Research suggests that there is a substantial degree of crossover in victim age between adult and child victims (Vess and Skelton, 2010).
Dynamic Factors One limitation to an actuarial measures that utilize only static factors is that they are insensitive to factors which may moderate the risk for an individual offender. Static measures also provide little information about the immediacy of
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
177
the risk for reoffending, and therefore are not helpful in developing an ongoing risk management plan. Dynamic, changeable factors which have demonstrated an empirical association with sexual reoffending should therefore also be considered in the assessment and management of individual cases. These dynamic factors arguably constitute psychological vulnerabilities and can provide a more individualized understanding of risk, identify those factors that should be monitored over time in order to detect changes in risk, and indicate areas of focus for supervision and intervention efforts. There are two types of dynamic risk factors: Stable dynamic risk factors are defined by Hanson et al. (2007) as, “personal skill deficits, predilections, and learned behaviors that correlate with sexual recidivism but that can be changed” through intervention (p. i). Acute dynamic risk factors are defined as highly transient conditions that only last hours or days. These factors include “rapidly changing environmental and intrapersonal stresses, conditions, or events that have been shown by previous research to be related to imminent sexual re-offense (Hanson et al., 2007, p. i). The first commonly used measure for dynamic variables was the Sex Offender Need Assessment Rating (SONAR), an actuarially based measure of dynamic risk factors empirically related to rates of sexual recidivism (Hanson and Harris, 2001). The SONAR was developed to measure and evaluate changes in dynamic risk factors over time. The SONAR has more recently evolved into two separate measures to assess variables across the two domains of stable dynamic factors and acute dynamic factors. The STABLE-2007 identifies 13 stable dynamic factors (Hanson, et al., 2007). The ACUTE-2007 (Hanson et al., 2007) is designed to assess these factors, including items such as victim access, rejection of supervision, collapse of social supports, and substance abuse. Acute dynamic risk factors exert an influence in the immediate environment in close temporal proximity to potential sexual offending. It is therefore typically not possible to assess these factors meaningfully while an offender remains incarcerated, although it is sometimes possible to anticipate factors that may be present in the post-release environment and extrapolate from current functioning.
Combining Measures The assessment of dynamic risk factors using structured measures such as the STABLE-2007 results in a standardized score that is used to categorize an offender’s risk as low, moderate, or high risk. These results are then used to adjust the level of risk as measured by the static actuarial risk factors (e.g., using the STATIC-99) to yield an overall categorization of the offender’s assessed risk for sexual reoffending. Recent research on sex offenders has shown that risk predictions made by static actuarial measures can be enhanced by incorporating dynamic variables to give a fuller picture of individualized prediction of risk (Craig et al., 2004; Craissati and Beech, 2005). Studies have given empirical support to the hypothesis that including an assessment of dynamic factors can strengthen the utility of static actuarial measures designed to measure sexual recidivism (Beech et al., 2002; Thornton, 2002; Hanson and Morton-Bourgon, 2005).
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
178
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Reliability and Accuracy of Risk Assessment For the purposes of applied risk assessment with sex offenders, the first concern is interrater reliability. The key issue here is whether different assessment experts arrive at similar findings regarding risk, using standardized assessment measures. Actuarial measures, such as the Rapid Risk Assessment for Sex Offender Recidivism (RRASOR) and Static-99, have demonstrated consistently high levels of interrater reliability, with reliability coefficients of .90 and higher, largely because of highly specific scoring criteria involved in this form of standardized assessment measure (Barbaree et al., 2001; Doren, 2002; Sjostedt and Langstrom, 2001). Standardized measures of dynamic risk factors such as the STABLE-2000 and STABLE2007 have also demonstrated adequate interrater reliability when administered by trained assessors according to specified scoring criteria, with an interclass correlation coefficient of .89 for STABLE-2000 total scores (Hanson et al., 2007). As La Fond (2005), has stated, “Actuarial instruments reduce much of the variability in predictions that can generate very different results when individual clinicians bring their different training, experience, and normative preferences to the task of assessing sex offenders for risk.” La Fond goes on to note, “No serious dispute exists about the fact that these instruments only allow experts to conclude that a particular individual belongs to a group with certain risk factors; however, they cannot be used to state authoritatively that an individual has a certain probability of reoffending” (p. 53). The predictive accuracy of risk assessment measures for sex offenders has been reviewed in detail elsewhere, and is beyond the scope of the current chapter (Hanson and Morton-Bourgon, 2009; Vess, 2009). It is important to recognize that none of the statistical indices yet developed can completely answer the question of how accurate a risk assessment measure is (Gottfredson and Moriarty, 2006). Some researchers (Quinsey et al., 2006) have argued that Relative Operating Characteristic (ROC) analysis offers the best index of statistical accuracy because it is independent of the base-rate variations in different samples of offenders. Yet others argue that the base rate of sexual recidivism cannot be ignored in considering the accuracy of a given measure in a particular application. Knowing the relevant base rate, and its impact on the accuracy of risk predictions based on various measures, is important for making sense of statements made about the risk measures used in a particular case (Gottfredson and Moriarty, 2006). An important issue related to the accuracy of actuarial risk assessment measures is whether they can be appropriately applied with individual offenders. This issue was highlighted by Hart, Michie and Cooke (2007), who concluded that the margin of error for actuarial risk assessment instruments is far too great to be used to estimate an individual’s risk for future offending, and should be used with great caution or not at all; however, this position has been criticized on several points, as recently summarized by Craig and Beech (2010) who cite evidence that the replication studies for the Static-99 provide empirical evidence that is more consistent than would be the case if the measure were as inaccurate as Hart et al. (2007) assert, and suggest that the internal logic of their argument regarding risk in terms of probability is inconsistent. Additional criticisms have focused on the mathematical basis for the margin of error figures, whereby the substitution of the
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
179
number “1” in place of the group figure of “n” in the formula used for calculating the margin of error renders the results uninterpretable (Mossman and Sellke, 2007; Harris, Rice and Quinsey, 2007). The idea that it is wrong to use group data as a basis for decisions about individuals has been refuted previously (Grove and Meehl, 1996; Harris et al., 2007; Quinsey et al., 2006). In light of the margin for error inherent in current actuarial measures, more individualized risk assessments are desirable, preferably those that take into account dynamic and aetiological factors (Ward and Beech, 2004). In our view, such assessments should culminate in a case formulation that depicts the etiological significance of the different dynamic risk factors that are present for the specific offender being assessed. The clinical use of these measures in a mechanical, formulaic way does not result in an explanatory formulation of risk that accounts for the idiosyncratic pathway to sexual offending, in other words the how and why of offending, for the particular individual who is being assessed. Such a formulation requires a detailed understanding of the specific behaviors, cognitions and affects that result in sexual offending (Vess, 2008).
Case Formulation and Risk Assessment Psychological assessment involves a systematic process of collecting, evaluating, and integrating relevant information about a client’s problems to arrive at conclusions about their nature, etiology, and implications (Ward and Haig, 1997; Ward, Vertue and Haig, 1999). A clinical assessment is said to be complete when the assessor arrives at a clear formulation of the client’s difficulties which enables the referral questions to be answered, at least provisionally. Relevant questions include: What are the main presenting problems or issues? How are these problems interrelated and what etiological explanations account for their occurrence? What options for modifying these difficulties are most likely to be efficacious for this person? Assessment is also an integral part of treatment in that practitioners must monitor and evaluate the effectiveness of their work with clients. This requires the establishment of rapport with the client. A forensic case formulation derived from the assessment process is a conceptual model representing an offender’s various problems, the hypothesized underlying mechanisms, and their interrelationships. It should be clearly linked to contemporary theory and research on offending behavior. A forensic case formulation is in effect a clinical explanation in the form of a testable hypothesis that specifies how an offender’s symptoms or problems are generated by psychological mechanisms, for example, dysfunctional core beliefs or behavioral deficits (Ward et al., 1999). Furthermore, a case conceptualization provides a rational basis for determining treatment needs that can be used to tailor interventions to individual offenders with the aim of achieving optimal outcomes, including but not limited to reduced reoffending. Optimal risk assessment and management requires the extension of case formulation skills, whereby risk factors are identified within an etiological framework (Ward and Beech, 2004). While it is important to be aware of the static and dynamic factors associated with a given type of risk, these factors
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
180
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
are most useful for risk management purposes when they are formulated into a coherent set of interrelated causal mechanisms (Collie, Ward and Vess, 2008; Vess, Ward and Collie, 2008). This requires examination of several categories of contributing factors, including historical (e.g., offense history, past episodes of violence, previous treatment compliance and response, performance under supervision or parole), developmental (e.g., adverse developmental events, nature of family relationships, attachment style), cognitive (e.g., level of intelligence, cognitive distortions, attitudes supportive of criminality or violence), personality (e.g., psychopathy, or traits such as impulsivity and hostility) and clinical (e.g., psychiatric diagnosis, level of functioning, substance abuse). In light of these individual factors, it is important to recognize that risk is contingent upon current situational or contextual variables (Doren, 2002). Even high risk cases will not be at imminent risk at all times, but will vary in their likelihood of reoffending depending on such factors as access to victims, current degree of alcohol or drug use, access to and compliance with treatment and supervision services, the nature of interpersonal relationships and support systems, and current mood states. Thus, different individuals who have similar profiles in terms of their scores on various risk assessment measures will not necessarily respond in a similar way to the same interventions or risk management plan. The recidivism risk at any given time will emerge from an etiological process determined by the interaction of individual characteristics and contextual factors. A case formation is in essence a circumscribed theoretically-grounded hypothesis and provides an explanation of why an offender acted as he did, while pointing to areas of vulnerability and harmful behavior that ought to be the focus of treatment and management (Vess, Ward and Collie, 2008). The following example illustrates this approach to case formulation.
A CLINICAL EXAMPLE: THE CASE OF JOHN DOE John Doe was a 38 year old Caucasian male currently detained in a secure treatment facility under indefinite civil commitment as a dangerous sexual offender. He was committed to this treatment program after serving a seven year prison sentence for the sexual assault of three victims, two male and one female, under 12 years of age. This was his second prison sentence for sexual offending, with prior convictions for indecent assault of a male child and rape of a female child when Mr Doe was 24 years old, for which he served 4 years in prison. He has now been civilly committed for two years, and is due for a legally mandated review of his response to treatment and current risk for sexual reoffending.
ASSESSMENTS Risk Assessment Static Factors Mr Doe’s risk of reoffending based on static historical variables was assessed using the Static-99. His score of 5 on this instrument places him in what has been
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
181
labeled the moderate-high risk category. Offenders in this category have been reported in previous research findings to show sexual reoffending rates of 33% at five years following release from prison, 38% at ten years post-release, and 40% at 15 years (Harris et al., 2003). More recent research with larger samples of sex offenders provide new norms for sexual recidivism rates associated with specific scores on the Static-99 (Helmus, Hanson and Thornton, 2009). Using what the researchers refer to as routine correctional samples of sex offenders, those with a score of 5 showed sexual recidivism rates of 10.2% at five years and 23.1% at the years post-release. Those from preselected high risk samples showed sexual recidivism rates of 11.8% at five years and 32.1% at ten years. These rates provide an empirically derived range of recidivism risk for those with Mr Doe’s score on the Static-99. By virtue of his civil commitment as a dangerous sexual offender, he is considered to more closely resemble the high risk samples of the most recent norms. Another way to represent Mr Doe’s risk for sexual reoffending is to report the rate for those with his score relative to the typical sex offender (i.e., someone with the median score of 2 in the most recent normative samples). Someone with Mr Doe’s score shows approximately two and a half times the rate of sexual recidivism relative to the average sex offender. The variables that place Mr Doe in this category include his prior sexual convictions, having male victims, unrelated victims, and lack of a stable intimate relationship.
Dynamic Factors Mr Doe’s risk based on dynamic factors was assessed using the STABLE-2007. The item on negative social influences assesses the nature of the offender’s primary interpersonal relationships. Mr Doe identified several family members as his closest relationships, and the information he provided suggests that none of them represent clearly antisocial influences on Mr Doe’s behavior (e.g., condoning sexual abuse, supporting alcohol or drug abuse, criminal activities, or defiance of the conditions of supervision). Mr Doe has not had a stable intimate relationship of at least two years duration, and therefore has not demonstrated a capacity for relationship stability. He shows clear signs of emotional identification with children but no significant hostility toward women. He demonstrates evidence of general social rejection and lack of concern for others. Impulsivity is currently less of a problem for Mr Doe than in the past, but he still occasionally displays impulsive behavior, particularly when he is under emotional stress. Similarly, he has shown improvements in his problem solving skills and negative emotionality, although in recent months he has had episodes demonstrating problems in both of these areas of functioning. Based on the current assessment of Mr Doe’s recent functioning, there is little evidence that he continues to experience the high levels of sexual preoccupation and sex drive that were evident in descriptions of his behavior when he was younger. The observations of the Prison Officers also suggest that Mr Doe is not a sexually preoccupied or sexually active prisoner. His self-reported sexual fantasy and masturbatory behavior suggest that he still occasionally uses sex for coping when faced with environmental or interpersonal challenges, such as when he has
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
182
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
experienced conflict with other inmates or has been confronted about his behavior in during treatment. His past sexual offenses suggest the presence of a deviant sexual preference for sexual activity with pre- and postpubescent children of both sexes. The results of penile plethysmograph assessment demonstrates arousal to pre- and postpubescent children of both sexes. The approach recommended by the internationally recognized expert Karl Hanson and his colleagues (Hanson et al., 2007) for combining the results of actuarial risk measures such as the Static-99 and dynamic risk measures such as the STABLE2007 to arrive at an adjusted level of risk for sexual reoffending suggests that a moderate-high level of static risk combined with a high level of stable dynamic risk yields a high overall risk. As previously described, actuarial approaches to risk assessment work by placing an individual offender in a group of offenders with similar characteristics for whom specific sexual reoffending rates over time have been observed. In order to better understand the risk presented by the offender under consideration, an individualized formulation is needed, including other factors known to contribute to increased risk.
Psychopathy Mr Doe was assessed using the revised Psychopathy Checklist (PCL-R; Hare, 2003). Mr Doe’s score on the PCL-R placed him below the usual cut-off used to classify someone as a psychopath. His total score placed him slightly above the mean (average) for male offenders, in other words slightly above the 50th percentile of male offender populations. While Mr Doe does not score in the range of the PCL-R that is properly classified as highly psychopathic, and therefore at increased risk based on this factor, he does, however, show certain psychopathic features which will likely influence his response to treatment and externally imposed supervision requirements. These characteristics, such as shallow affect, lack of empathy, manipulation, impulsivity and irresponsibility, may make it more difficult for him to internalize the goals of a treatment program, and to form the trusting therapeutic relationships necessary to progress in treatment and reduce his risk of reoffending.
Conclusions Regarding Risk The results of the current risk assessment indicate that Mr Doe remains at high risk for sexual reoffending if released to the community. Based on his pattern of previous sexual offending, Mr Doe is most at risk of reoffending when he is experiencing environmental and interpersonal distress such as conflict with peers or perceived criticism from those in authority positions, resulting in negative affect such as loneliness and depression, accompanied by feelings of being worthless, incompetent, and rejected by others. If he has unsupervised access to children at such times, especially those with whom he has established a sense of familiarity and trust, these children become his most likely potential victims.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
183
ASSESSMENT FOR CASE FORMULATION In order to develop a case formulation for Mr Doe, information is needed beyond a list of the static and dynamic risk factors known to be associated with risk for sexual reoffending. What is required is a more comprehensive understanding of his developmental history and current functioning in a variety of domains, so that an integrated etiological, explanatory framework for his offending can be provided. Recalling the categories of factors identified earlier in this chapter, the following sources of information are relevant for his case formulation.
Developmental History Mr Doe was born into an extremely dysfunctional family environment. His father is reported to have had significant drug and alcohol abuse problems, and had several convictions for violent offending. Mr Doe and his sisters were sexually abused by their father from an early age. His mother was ineffectual at stopping the abuse, and in fact did not believe Mr Doe’s reports of the abuse. The abuse came to light after one of the sisters informed authorities at school, whereby Mr Doe’s father was arrested, convicted and imprisoned. Mr Doe was eight years old at this time. From this dysfunctional family history it must be concluded that Mr Doe suffered severe problems with early attachment and trust, demonstrated by his vacillation during childhood between clinging dependence and emotional withdrawal from others, which subsequently interfered with the development of a stable and functional sense of self. At a behavioral level, his primary models for interpersonal and sexual relationships were severely disturbed. His psychological and psychosexual development was severely compromised by the ongoing experiences of sexual abuse by his father. Added to this was Mr Doe’s lack of compensating physical or mental abilities, whereby he has been described as physically awkward and socially immature. He lacked the interpersonal skills to fit in with others, and resorted to provocative behavior in his attempts to make social connections. As a result he was taunted, bullied, and ultimately rejected by peers. This further eroded his sense of adequacy and worth. Mr Doe’s childhood was marked by enduring emotional and behavioral problems, poor school performance, and developmental delays. At 14 years of age he began using cannabis and abusing alcohol that he would steal from his mother. His poor performance and behavioral problems at school progressively escalated, and he left school at the end of year ten (age 15 years), having failed all his academic subjects. Mr Doe reports an unstable and largely unsuccessful work history after leaving school. He has held several unskilled jobs as a labourer, janitor, and dishwasher. The longest he has held a job is 9 months, and he typically quits or is dismissed following conflicts with fellow workers and supervisors. His job performance has also occasionally been compromised by his cannabis and alcohol use.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
184
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Offense History Mr Doe has a total of 14 criminal convictions, including public intoxication, disorderly conduct, wilful damage of property, shoplifting, and cannabis possession. Mr Doe’s first sexual offending convictions were for indecent assault of a male child and two counts of rape of a female child when Mr Doe was 24 years old. These offenses occurred in the context of having returned to live with his mother following her separation from her second husband. The victims were a 12-year-old niece and 10-year-old nephew who were also residing at his mother’s house at this time. Mr Doe engaged in penetrative sexual intercourse with the female victim on two occasions and induced the male victim to touch his erect penis. Mr Doe has consistently denied that he used force in his sexual contact with these victims, which appears consistent with available court reports. Mr Doe’s index sexual offending consisted of the indecent assault of two 8year-old boys and unlawful sexual contact with an 11-year-old girl. These victims resided in the neighbourhood where Mr Doe was living following release from prison for his earlier sentence. Mr Doe enticed the victims into his apartment with offers of video games and candy, after which he engaged in touching the genitals of the male victims and performed oral sex on the female victim. He denies that there was any penetrative sexual contact with any of these victims, but has acknowledged in treatment that he would masturbate to fantasies of such activities with these children, and probably would have attempted to progress to greater sexual contact if given more time with the victims. During treatment, he disclosed additional sexual offenses against children of both sexes beyond those for which he has been convicted.
Treatment for Sexual Offending The most intensive and focused treatment Mr Doe has received appears to have occurred in the context of his current confinement in the dangerous sexual offender treatment program. This program has emphasized the usual components of treatment currently considered by most professionals to reflect the best standard of practice in the area of sex offender rehabilitation. These components include increasing the offender’s understanding of the chain of events that led to prior sexual offenses, identifying and understanding the role of deviant sexual fantasies and cognitive distortions, increasing empathy for victims, improving skills necessary for appropriate adult relationships, effective mood management, attempts to modify sexual arousal responses to inappropriate stimuli such as children, and the identification of high risk situations along with relapse prevention plans to deal with such situations. Mr Doe has made limited progress in these standard areas of treatment. His response to these initial treatment efforts provides additional information for the clinical formulation. He has been described as being able to verbalize an understanding of his chain of offense behaviors and high risk situations, but has failed to adequately demonstrate this understanding in his daily functioning and general
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
185
self-management. This suggests that he requires additional guidance and support to enact his intellectual understanding in his behavior. Concerns remain about his deviant sexual arousal to children and the potential for inappropriate sexual fantasies involving both male and female children. This will require sufficient rapport with treatment staff to openly explore his sexual arousal patterns. This is related to areas of functioning which appear not to have been thoroughly addressed thus far in treatment that involve his distrust of authority figures, including treatment staff, his lack of appropriate adult relationship experiences, and general reintegration issues such as employment and independent living skills. Previous reports indicate that Mr Doe has at times been reluctant or ambivalent about addressing some of these sexuality issues in the context of available treatment.
Cognitive Factors Mr Doe has been assessed previously as low average to average in his intellectual functioning, with test results on three occasions since adolescence indicating Full Scale IQ scores ranging from 82 to 98, and no substantial deficits in adaptive functioning. He tends to be somewhat concrete in his thinking, but does not show overt intellectual impairment to a degree that will significantly compromise his response to cognitive-behavioral treatment modalities. He does appear to maintain cognitive distortions related to his implicit theories about self and others, especially his perceptions of children as legitimate partners who can responsibly enter into intimate relationships to meet his emotional and sexual needs. These cognitive distortions are particularly apparent during times of interpersonal distress, when Mr Doe finds adult interactions threatening.
Clinical and Personality Features The influences of Mr Doe’s developmental history are seen as contributing to his current personality features. One feature noted evident in the current assessment is Mr Doe’s profound lack of connection to emotional experiences, both his own and those of others. This quality is potentially suggestive of a schizoid personality traits, whereby an individual has a lack of emotional reactions and is not interested in relationships with others. Mr Doe, however, shows more dependent and avoidant personality features, whereby he appears to crave attention from and a connection to others, but feels inadequate to engage in trusting intimate relationships and routinely anticipates abuse, ridicule or rejection from others. These personality features are also suggested by the results of a recent Millon Clinical Multiaxial Inventory (MCMI-III) (Millon, 1997). Mr Doe had primary elevations on the scales for Dependent, Avoidant, and Self-Defeating personality patterns, along with a significant level of anxiety. Individuals with this profile desperately want to be accepted and involved with other people but this desire is blocked by intense fear of being rejected or ridiculed. They typically perceive themselves as socially inept, inferior, and inadequate. Because of their fear of social situations and close relationships, they may rely heavily on fantasy to gratify
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
186
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
their needs for affection and to cope with their anger. This has the potential to restrict them to a solitary life where they are more likely to reactivate memories of past social rejections, rather than risk forming new, more satisfactory relationships. These dependency traits augment the core avoidant tendencies and add a desire to be supported and guided by others. The passive aggressive elements suggest the presence of moodiness and resentment, with significant difficulty trusting others. Such individuals may vacillate between being friendly and cooperative and then being hostile. They will subsequently often apologize in an attempt to be accepted again by others. As they are often uncomfortable with their anger, they may resort to covert expressions of hostility, such as passive obstructionism. A compounding factor for Mr Doe is his tendency to engage in excessive consumption of alcohol and cannabis as a way of coping with stress and moderating his negative affect.
Case Formulation Mr Doe was born into an extremely dysfunctional and damaging home environment, where he had an inadequate opportunity for secure attachment to a nurturing parent figure. It appears that he has never experienced an enduring relationship with an attachment figure which did not involve abuse, neglect, or interpersonal boundary violations, often of a sexualized nature. This resulted in deep-seated problems with trust, empathy and self-worth, leading to a confused and disturbed sense of self. His experiences have provided no foundation for healthy adult relationships that are free from the threat of abandonment, violence or sexual exploitation. Mr Doe appears to have learned from the beginning of his life that the world is largely an indifferent or hostile place where others will not protect or consistently care for him, and may in fact abuse him. In response to this he has learned at a very deep and fundamental level to deny or distance himself from his own emotional experiences as a form of self-protection, and to try to present as whatever someone else wants him to be in order to obtain the acceptance and approval he craves. Alternating with this is a deep resentment, which can be expressed as either direct aggression, or more frequently in passive-aggressive and self-defeating behavior. He appears to maintain a very fragile self-image with a massive underlying sense of inadequacy and unworthiness. His occasionally grandiose or inflated presentation seems to reflect a desperate need to be perceived by others and to perceive himself as adequate in at least some areas. Mr Doe’s sexual offending can be seen as one manifestation of his attempts to obtain the basic human needs for security, acceptance, affection, and control in his life. Mr Doe emerged from his adverse developmental experiences with a pervasive and generalized view of others, especially adult authority figures, as rejecting and untrustworthy, and himself as worthless and incompetent. Mr Doe initially learned to manage these negative feelings by withdrawing into isolated fantasies, further slowing his development and distorting his reality testing in social relationships. He subsequently looked to relationships with those much younger than himself to meet his emotional needs and bolster his sense of adequacy. With the onset of adolescence, he also managed his isolation and negative emotions through smoking
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
187
cannabis, drinking alcohol, and masturbation, focusing on deviant fantasies involving children, with whom he felt less vulnerable. These deviant fantasies were strongly reinforced through masturbation, and eventually through their enactment with children he came to know. Mr Doe’s sexual offending can also be understood in terms of the schemas or implicit theories he has developed. He learned through his early experiences to sexualize interpersonal relationships, and to associate affection with sexual contact, albeit in a conflicted way due to the negative emotions associated with his sexual abuse. His sense of self is also grossly deficient and fragile, so that his negative schemas are especially activated at times of interpersonal conflict or environmental stress. This dynamic is demonstrated in the context of his index offending, when he was living with his mother in a conflictual relationship that reactivated his anger toward her, along with his sense of vulnerability and isolation. He had also recently been dismissed from his job, increasing his sense of inadequacy and rejection. In these circumstances his need to cope with the attendant negative affect through sexual contact with children was particularly strong. He has reported that he began to see the children as potential partners who would be less rejecting than adults, and serve as outlets for his sexual needs. His judgement and sense of interpersonal boundaries were further compromised by his daily use of cannabis during this period. He also reported that he had developed what he experienced as an intimate bond with the children, and that the sexual activity was to him a natural expression of that intimacy.
The Formulation Process Developing a forensic case formulation such as Mr Doe’s is an ongoing process of identifying, testing, and integrating clinical hypotheses into a coherent explanatory theory or model of the individual based on the clinical information that is obtained. In cases of child victim sex offenders, the primary focus is typically an understanding of the sexual offending, with the goal of reducing and managing the risk for sexual recidivism. The forensic clinician starts with the known history of sexual offending, and bases the initial estimates of the probability of reoffending on empirically validated static and dynamic risk measures such as the Static-99, STABLE-2007, and ACUTE-2007. Specific attention should also be paid to the factors of psychopathy and deviant sexual arousal, as these are important risk factors identified in the research literature. Beyond these factors, case formulation involves a detailed examination of the process by which prior sexual offending has occurred, including the cognitions, emotions, interpersonal and environmental factors involved. This will depend on the cooperation of the offender to obtain a specific and accurate account of his or her offending process, sometimes called an offense cycle or behavior chain. In the case of Mr Doe, this aspect of the assessment and formulation was compromised and delayed by his reflexive distrust of others, particularly authority figures associated with the criminal justice system. A process of rapport building over many months was necessary, with several ruptures in the therapeutic alliance that required repair before additional disclosures were forthcoming to inform the formulation.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
188
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
One initial false start was the hypothesis that Mr Doe had significant narcissistic elements to his personality, based on the perception by some staff that he displayed a sense of entitlement and grandiosity, and that narcissistic defenses would be a challenge to the therapeutic process; however, results of the MCMI-III indicated an unusually low score on the narcissism scale. Behaviorally, it soon became apparent that Mr Doe was struggling much more overtly with a massive sense of inadequacy, and displayed more dependent and avoidant personality patterns in his relationships rather than the inflated self regard typical of narcissism. Once this was understood, the attendant needs to offer reassurance and not to restimulate the punitive and rejecting dynamics that have marked most of his important adult relationships became apparent. The therapists in the treatment program were able to directly and consistently address Mr Doe’s sensitivities to criticism, while still holding him accountable to follow program rules and address the issues necessary for treatment progress. This issue led to another revision in the treatment approach with Mr Doe. Like many sex offenders, an aggressively confrontive approach intended to break down denial and encourage taking responsibility for his offending had the effect of exacerbating Mr Doe’s sense of shame and personal inadequacy, with a corresponding increase in defensiveness and reversion to the maladaptive behavior patterns associated with his offending. It was found that a more collaborative, supportive approach, with specific and ongoing attention to the effects of the content of the treatment process on the therapeutic alliance, has begun to enable Mr Doe to undertake a more direct and less distorted examination of his offense cycle.
Treatment Implications In light of Mr Doe current psychological characteristics, there will be challenges to achieving significant therapeutic gains. A primary factor in any treatment effort will be the challenge of creating and maintaining a strong therapeutic alliance between Mr Doe and the clinician. Mr Doe’s problems with trust have been noted. His pervasive and largely unconscious internal models concerning others lead to resilient beliefs that adult males are victimizing and females are unreliable to protect him. These problems will be compounded by the potentially adversarial nature of treatment within the correctional system, whereby staff are explicitly responsible for both treatment and for protecting the public safety. To the degree that Mr Doe feels vulnerable to negative consequences in response to disclosures he may make, this will inhibit the development of a therapeutic process that would allow for the exploration of these issues. Mr Doe also shows evidence of significant emotional and personality difficulties that bear upon his responses to treatment. He lacks the communication and social skills to develop effective interpersonal relationships, including those with treatment staff. His attempts to manage his interpersonal anxiety and make social contact with others continue to involve awkward and immature attempts at humour, often resulting in experiences of rejection. He is largely disconnected from his own emotional experiences, and has limited recognition of the emotional reality
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
189
of others. This limits his capacity for empathy, and will make it difficult for him to experience the emotions that normally inhibit acts that inflict harm on others. A final treatment implication concerns the approaches or models of treatment that may be most effective. It appears that Mr Doe has thus far received a primarily cognitive-behavioral approach, which in the context of the current treatment program has been applied to the areas that are usually associated with effective sex offender treatment. While such an approach has shown some success in reducing reoffending with a portion of the sex offender population, and undoubtedly has content that is relevant for Mr Doe, it appears likely that his emotional and personality characteristics may prevent this approach from being optimally effective. Specifically, the formulation for Mr Doe suggests that his early developmental experiences have resulted in a pervasive sense of his own inadequacy, a distrust of others that is especially pronounced with those in authority positions, and a sense of vulnerability in relationships that involve intimate personal disclosures. All of these factors present obstacles to the therapeutic process. In light of Mr Doe’s abusive and damaging early developmental history, and the idiosyncrasies of his personality and sense of self, it may prove beneficial augment further treatment efforts with a approach that emphasizes these early experiences and their deepseated sequelae. This will require a patient and sensitive approach by the therapist, with ongoing attention to ruptures in the therapeutic alliance resulting from Mr Doe’s responses to clinical material that he experiences as threatening. Mr Doe will also benefit from having other needs met, needs which may assist him in establishing a more stable, rewarding and offense-free life. Mr Doe’s formulation suggests that he has deficits in multiple areas of age-appropriate functioning. These include employment, education, socialization, appropriate adult sexual experiences and relationships, as well as spiritual and personal identity needs (Ward and Maruna, 2007). Because of his deeply held beliefs about the world and others as indifferent or hostile, and the corresponding defenses that have limited his social development, Mr Doe will require a slow and supportive treatment process in these areas. He does not have the normal maturational experiences typically encountered in an individual of his age, so that his progress in these areas must begin from their current underdeveloped state. Although these issues can be identified and addressed in a preparatory way while he is in the secure treatment setting, they will be most effectively dealt with through ongoing treatment and support in the community. It is therefore crucial that an integrated aftercare plan be developed based on his individual case formulation that will address his needs for successful reintegration into the community.
Measuring Outcomes The ultimate outcome measure from the criminal justice perspective is the absence of sexual reoffending. Other, intermediate or associated outcomes can also be defined as measures of the adequacy of the case formulation and resulting interventions. In the inpatient environment, progress can be measured by progression through the identified stages of the program. Mr Doe’s treatment program, like many of its kind, is designed to facilitate advancement through various distinct
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
190
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
stages, each associated with progressively more internally directed change and risk management (i.e., internal to the offender rather than imposed and directed from others). To the extent that Mr Doe is self initiating in his pursuit of understanding and modifying the factors that contribute to his offending, this is a definable treatment outcome. Specific indicators for Mr Doe are increased self-disclosure, decreased defensiveness, fewer incidents of manipulation or deceit, and improved relationships with peers and staff. Behavioral indicators for each of these areas should be defined in light of Mr Doe’s past behavior so that his progress can be assessed. He will be assisted in making the desired changes in his behavior by providing clear examples of what is needed, such as avoiding instances of lying to staff or peers, responding with nondefensive consideration of criticism offered by others, and self-initiated disclosure of personal information in appropriate interpersonal contexts. There are also standardized measures of functional skills that can be used for treatment planning and outcome measurement with forensic patients, including sexual offenders (Vess, 2001a, 2001b). Such measures can be used repeatedly over time to assess the degree of change in areas that are a focus of intervention. Measures of dynamic risk factors such as the STABLE-2007 can also be used in a secure inpatient setting, and the STABLE and ACUTE measures were specifically designed to assess areas of functioning directly related to the risk of sexual recidivism in the community. Where the assessment resources are available, penile plethysmograph assessment can be used to both provide a baseline assessment of deviant arousal for case formulation and treatment planning purposes, and to assess changes in deviant arousal in response to treatment. It should be recognized that not all offenders respond adequately to the stimuli used in such assessments, and the assessment of deviant arousal should also rely on other sources of self-report and behavioral observation data. This is best achieved in the context of a strong therapeutic alliance in which candid self-disclosure regarding sexual fantasies and behavior is possible. Finally, a functional analysis of the offense pathway for the individual offender provides a behaviorally anchored map of the cognitive, emotional, and contextual factors associated with sexual offending (Vess, 2008; Ward, Yates and Long, 2006). The results of such an analysis not only informs the case formulation for the individual, but can serve as a set of observable precursors to sexual reoffending that can be used as an ongoing measure of treatment outcome. This would require a close working relationship with the offender, especially in the community setting, to allow for adequate ongoing assessment of the offender’s functioning in the key areas identified in the functional analysis of his offending pattern.
SUMMARY AND CONCLUSIONS An important implication of the case formulation approach presented here is that risk is heavily contingent on the postrelease environment. The elements of the risk management plan will depend on the specific needs of the individual offender. Group and individual treatment modalities have the potential to reduce risk, if they are well matched to the needs of the offender and delivered with fidelity to
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
191
the original clinical design and purpose of the intervention. The frequency and intensity of treatment and supervision activities should be based on an ongoing assessment of risk, which can increase or decrease in frequency over time depending on current contingencies in light of the offender’s individual risk factors. High risk offenders may initially need frequent and intense supervision, including scheduled office appointments and unannounced home visits by the parole agent to minimize the opportunity for acute risk of reoffense to develop. This has implications for staffing levels and case load sizes, as well as the availability of treatment and ancillary support services in the community. Offenders will vary in the set of needs they present, their level of risk and their responsivity to treatment. Some offenders will have fundamental life skills deficits which contribute to the etiological process leading up to sexual offending, and others will not. Some offenders will have clearly established patterns of deviant sexual preference, such as paedophilia, and others will be nonspecialized, opportunistic career criminals for whom sexual offenses are just one aspect of general criminal offending (e.g., the immature, impulsive and antisocial adolescent who sexually offends against a child while disinhibited by drugs, but who has no enduring sexual interest in children). These different offenders will present different needs, so it is unlikely that a “one size fits all” approach will effectively manage risk. Targeted interventions should be based on thorough, individualized assessment and case formulations. Another important implication of the contingent nature of risk for clinicians contributing to case management is to combine current empirical knowledge with a clear and realistic understanding of the local environment and resources. One must have an individualized, context specific risk formulation, and then be able to communicate the assessment findings effectively to others involved in the management of an individual case. This is best done when the clinician is also thoroughly familiar with risk hazards present in the offender’s community environment, including the nature of his intimate and social relationships, potential substance abuse, emotional and mood management, and so on, as delineated in a functional analysis of the individual’s offending. What is to be avoided in clinicians’ assessment reports is a formulaic adherence to known risk factors, combined in a mechanical fashion and resulting in a routine list of treatment recommendations. Formulaic assessments of risk are less useful than formulations of risk for the individual offender. A case formulation provides an etiological framework with which to understand the unfolding process which results in a sexual offense for this particular individual. When this understanding is combined with an accurate, detailed picture of the individual’s current life circumstances, and a knowledge of the available supervision, support and treatment resources, then a dynamic and responsive risk management framework can be developed.
REFERENCES Abracen, J., Mailloux, D.L., Serin, R.C. et al. (2004) A model or the assessment of static and dynamic factors in sexual offenders. Journal of Sex Research, 41, 321–9.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
192
May 26, 2011
4:59
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Barbaree, H. E., Seto, M. C., Langton, C. M. and Peacock, E. J. (2001) Evaluating the predictive accuracy of six risk assessment instruments for adult sex offenders. Criminal Justice and Behavior, 28, 490–521. Beech, A., Friendship, C., Erikson, M. and Hanson, R. K. (2002) The relationship between static and dynamic risk factors and reconviction in a sample of U.K. child abusers. Sexual Abuse: A Journal of Research and Treatment, 14, 155–67. Borum, R. (1996) Improving the clinical practice of violence risk assessment: Technology, guidelines, and training. American Psychologist, 51, 945–56. Collie, R., Ward, T. and Vess, J. (2008) Assessment and case conceptualization in sex offender treatment. Journal of Behavior Analysis of Offender and Victim – Treatment and Prevention, 1, 65–81. Craig, L.A. and Beech, A.R. (2010) Towards a guide to best practice in conducting actuarial risk assessments with sex offenders. Aggression and Violent Behavior, 15, 278–93. Craig, L. A., Browne, K. D. and Stringer, I. (2004) Comparing sex offender assessment measures on a UK sample. International Journal of Offender Therapy and Comparative Criminology, 48, 7–27. Craissati, J. and Beech, A. (2005) Risk prediction and failure in a complete urban sample of sex offenders. Journal of Forensic Psychiatry and Psychology, 16, 24–40. Doren, D. (2002) Evaluating Sex offenders: A Manual for Civil Commitments and Beyond. London: Sage Publications. Gottfredson, S.D. and Moriarty, L.J. (2006) Statistical risk assessment: Old problems and new applications. Crime and Delinquency, 52, 178–200. Grove, W.M. and Meehle, P.E. (1996) Comparative efficiency of informal (subjective, impressionistic) and formal (mechanical, algorithmic) prediction procedures: The clinical – statistical controversy. Psychology, Public Policy, and Law, 2, 293–323. Hanson, R.K. (1998) What do we know about sex offender risk assessment? Psychology, Public Policy, and Law, 4, 50–72. Hanson, R.K. and Bussi`ere, M.T. (1998) Predicting relapse: A meta-analysis of sexual offender recidivism studies. Journal of Consulting and Clinical Psychology, 66, 348–62. Hanson, R.K. and Harris, A.J.R. (2001) A structured approach to evaluating change among sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 13, 105–22. Hanson, R.K., Harris, A.J.R., Scott, T.L. and Helmus, L. (2007) Assessing the Risk of Sexual Offenders on Community Supervision: The Dynamic Supervision Project. Retrieved January 25, 2008 from Public Safety Canada website at http://www.publicsafety.gc.ca/ res/cor/rep/_fl/crp2007-05-en.pdf Hanson, R.K. and Morton-Bourgon, K.E. (2005) The characteristics of persistent sexual offenders: A meta-analysis of recidivism studies. Journal of Consulting and Clinical Psychology, 73, 1154–63. Hanson, R.K. and Morton-Bourgon, K.E. (2009) The accuracy of recidivism risk assessments for sexual offenders: A meta-analysis of 118 prediction studies. Psychological Assessment, 21, 1–21. Hanson, R.K. and Thornton, D. (2000) Improving risk assessment for sex offenders: A comparison of three actuarial scales. Law and Human Behaviour, 24, 119–36. Hare, R.D. (2003) Hare Psychopathy Checklist – Revised (PCL-R) (2nd edn). Toronto: MultiHealth Systems, Inc. Harris, A.J.R. and Hanson, R.K. (2004) Sex offender recidivism: A simple question. Retrieved August 23, 2005, from Public Safety and Emergency Preparedness Canada’s Web site: http://www.psepcsppcc.gc.ca/publications/corrections/200403-2_e.asp Harris, A.J.R., Phenix, A., Hanson, R.K. and Thornton, D. (2003) Static-99 Coding Rules: Revised – 2003. www.sgc.gc.ca. Harris, G.T., Rice, M.E. and Quinsey, V.L. (2007) Shall evidenced-based risk assessment be abandoned? British Journal of Psychiatry, 192, 154. Hart, S.D., Michie, C. and Cooke, D.J. (2007) Precision of actuarial risk assessment instruments: Evaluating the ‘margins of error’ for group v. individual predictions of violence. British Journal of Psychiatry, 190 (suppl. 49), s60–s65. Heilbrun, K. (2001) Principles of Forensic Mental Health Assessment. New York: Plenum.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST CHILDREN
193
Helmus, L., Hanson, R.K. and Thornton, D. (2009) Reporting Static-99 in light of new research on recidivism norms. The Forum, 21, 38–45. La Fond, J.Q. (2005). Preventing Sexual Violence: How Society Should Cope with Sex Offenders. Washington DC: American Psychological Association. Miller, H., Amenta, A. E. and Conroy, M. A. (2005) Sexually violent predator evaluations: Empirical evidence, strategies for professionals, and research directions. Law and Human Behavior, 29, 29–54 Millon, T. (1997) Millon Clinical Multiaxial Inventory-III Manual (2nd edn). Minneapolis, MN: National Computer Systems. Mossman, D. and Sellke, T.M. (2007) Avoiding errors abut ‘margins of error’. British Journal of Psychiatry, 191, 561. Quinsey, V.L., Harris, G.T., Rice, M.E. and Cormier, C.A. (2006) Violent Offenders: Appraising and Managing Risk (2nd edn). Washington D.C.: American Psychological Association. Sjostedt, G. and Langstrom, N. (2001) Actuarial assessment of sex offender recidivism risk: A cross-validation of the RRASOR and the Static-99 in Sweden. Law and Human Behavior, 25, 629–45. Skelton, A. and Vess, J. (2008) Risk of sexual recidivism as a function of age and actuarial risk. Journal of Sexual Aggression, 14, 199–209. Thornton, D. (2002) Constructing and testing a framework for dynamic risk assessment. Sexual Abuse: A Journal of Research and Treatment, 14, 139–53. Vess, J. (2001a) Development and implementation of a functional skills measure for forensic psychiatric inpatients. Journal of Forensic Psychiatry, 12, 594–611. Vess, J. (2001b) Implementation of a computer assisted treatment planning and outcome evaluation system in a forensic psychiatric hospital. Psychiatric Rehabilitation Journal, 25, 124–33. Vess, J. (2009) Fear and loathing in public policy: Ethical issues in laws for sex offenders. Aggression and Violent Behavior, 14, 264–72. Vess, J. (2008) Risk formulation with sex offenders: Integrating functional analysis and actuarial measures. Journal of Behavior Analysis of Offender and Victim – Treatment and Prevention, 1(4), 29–41. Vess, J. and Skelton, A. (2010) Sexual and violent recidivism by offender type and actuarial risk: Reoffending rates for rapists, child molesters and non-contact offenders. Psychology, Crime and Law, 16, 541–554. Vess, J., Ward, T. and Collie, R. (2008) Case formulation with sex offenders: An illustration of individualized risk assessment. Journal of Behavior Analysis of Offender and Victim – Treatment and Prevention, 1, 284–93. Ward, T. and Beech, A. (2004) The etiology of risk: A preliminary model for sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 16, 271–84. Ward, T. and Haig, B. D. (1997) Abductive method and clinical assessment. Australian Psychologist, 32, 93–100. Ward, T. and Maruna, S. (2007) Rehabilitation: Beyond the Risk Paradigm. New York: Routledge. Ward, T., Vertue, F.M. and Haig, B.D. (1999) Abductive method and clinical practice. Behaviour Change, 16, 49–63. Ward, T., Yates. P.M. and Long, C. (2006) The Self-regulation Model of the Offense and Relapse Process – Volume 2: Treatment. Victoria, Canada: Trafford Publishing.
P1: OTA/XYZ P2: ABC JWST073-08 JWST073-McMurran
May 26, 2011
4:59
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
Chapter 9
SEXUAL OFFENSES AGAINST ADULTS STACEY L. SHIPLEY North Texas State Hospital, USA
BRUCE A. ARRIGO University of North Carolina–Charlotte, USA
INTRODUCTION This chapter examines forensic case formulation for adult sex offenders who rape adult victims. This includes an overview of a range of issues relevant to male and female perpetrators and male and female victims. Along these lines, definitional, statistical, and background information are provided; classification schemas, including rapist profiles, are discussed; and the importance of assault prevalence, motive, method, victim selection, and offender disposition are delineated. For exemplification purposes, the instance of an adult male rapist who targets adult female victims is featured. Although its details are largely fictitious, presenting the clinical case makes it possible to then explore the contours of sexual violence risk assessment, case formulation, and treatment planning for this offender type. The final section of the chapter presents an illustrative case formulation and corresponding treatment plan.
LITERATURE REVIEW Sexual Assault of Adult Females Rape is typically defined as “the penetration of the anus or vagina by a penis, finger or object or the penetration of the mouth by a penis” (McCabe and Wauchope, 2005, p. 241). When a man forcibly penetrates someone else in any of the aforementioned ways, this is also considered rape. More broadly speaking, rape can be Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
196
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
defined as sexual contact with a nonconsenting person (Sbraga, 2003). Rapists have been categorized based on their distinct behavioral, motivational, and cognitive characteristics (McCabe and Wauchope, 2005). Theorists have indicated that an early criminogenic environment or child rearing in a subculture where criminal behavior is the norm can help to explain the onset and maintenance of sexually deviant behavior or offending in some cases (Lussier et al., 2007). This is especially the case when coupled with few external constraints, for example, lack of parental vigilance, limited attachment to conventional, prosocial institutions, such as school, church, family, and/or prolonged exposure to antisocial behaviors (e.g., antisocial parents and conduct disordered peers). Some opportunistic sexual offenders commit rape impulsively with victims of chance while others are clearly preferential in their victim type with rape as their primary criminal activity. Those who commit rape do not easily fit in a one size fits all category and their violence pathways, subsequent case formulations, and treatment and disposition planning can be quite distinct. Attachment theorists have suggested that early negative parent–child interactions damage the child’s ability to cope with stressful events and difficulties (Lussier, et al., 2007; Ward and Beech, 2006), while also failing to provide a healthy and sufficiently complete model for future intimate relationships (Marshall and Marshall, 2000). Moreover, sex offenders are often distinguished by early and enduring deficits in emotional regulation and a lack of social competence or an inability to foster healthy relationships with peers (Lussier et al., 2007). This condition is further exacerbated when raised in a sexually deviant environment and when subjected to early experiences of physical and/or sexual abuse where both the modeling of aberrant sexuality and the disinhibition of sexuality are gateways to adult criminality (Knight and Sims-Knight, 2003). Adult victimization accounts for approximately one-third of all rapes that occur in the United States each year with children representing the remainder of sexually assaulted victims (Salter, 2003). The US Department of Justice funded a national study of 8000 Americans for the Violence Against Women Survey, and their findings indicated that 17.6% of women and 3% of men reported having been raped (Tjaden and Thoennes, 1998). While the majority of all victims of rape or attempted rape are attacked by someone they know, stranger rapists are far more likely to avoid detection and to reoffend, and they are more likely to be represented in instances of serial rape. According to Salter (2003), only about 5% of all rapists ever spend a day in jail and still fewer fail to serve any significant time behind bars. A large number of rapes are perpetrated by a small number of serial offenders (Warren et al., 1991). The belief that serial rapists are all sadistic or even psychopathic is not supported by research (Pardue and Arrigo, 2008; Shipley and Arrigo, 2008). Psychopathy is the most common among opportunistic and pervasively angry rapists and infrequent in nonsadistic rapists (Richards et al., 2004). Psychopathy is far more common in rapists than child molesters (Serin et al., 1994). For many psychopathic rapists, the sexual assaults are an extension of criminal versatility. For example, the assailant may break into a home to commit a burglary and if the female resident happens to be home he will rape her without remorse (Shipley and Arrigo, 2008); however, the majority of serial rapists premeditate their assaults and utilize strategies to avoid detection (Kocsis, Cooksey and Irwin, 2002).
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
197
Clinicians who evaluate rapists or suspected serial rapists should pay close attention to charges such as Burglary of a Habitation or Attempted Burglary of a Habitation. Kocsis et al. (2002) found that theft is common after a rape; however, it is often an opportunistic crime in that the rape, as an expression of sexual violence, is the primary motive. Psychopathic serial rapists are particularly dangerous as they are at a higher risk to reoffend than nonpsychopaths, will recidivate sooner, and will likely reoffend more violently (Arrigo and Shipley, 2001; Serin, et al., 1994; Shipley and Arrigo, 2001, 2004, 2008). The sadistic rapist is sexually aroused by the physicality and emotionality of the pain and suffering, wherein the torture or harm is primary and the sex act is secondary (Shipley and Arrigo, 2008). The psychopathic offender is aware of the pain he causes but disregards or ignores the harm as it is irrelevant to sating one’s own needs (Salter, 2003; Shipley and Arrigo, 2001; Shipley and Arrigo, 2004, 2008). While all serial rapists are not abused as children, research has demonstrated that the majority experience sexual victimization (Burgess et al., 1988; Hazelwood and Burgess, 1987). In their original sample of 41 serial rapists, Burgess et al. (1988) found that the repetition of sexually aggressive behaviors began at adolescence subsequent to the onset of rape fantasies and rape behaviors. Indeed, in regard to juveniles’ compulsive sexual fantasies, the investigators noted that: “The repetition of the fantasies did not result in the resolution of . . . inner tension; rather, repetition became a prototype for the early rape behaviors” (p. 290). Motives for rapists vary widely. Some assailants principally seek to cause pain, as well as to degrade and defile the victim; others are motivated largely by the sexual, relational, and aggressive aspects of the assault (Warren et al., 1991). The rapist typologies and profiles described below address the aforementioned components and differ considerably based on the offender’s personality and intention.
Serial Rape Typologies According to Kocsis et al. (2002), while there have been many typologies offered relevant to the psychological profiling of rape, essentially they have been elaborations of the original ideas put forth by Groth, Burgess and Holmstrom (1977). Their typologies indicated that the actual sexual act of rape was secondary to the primary motivations of power and aggression. Three main patterns of rape were classified: anger; power; and sadistic. The anger rapist uses more force than is necessary for control of the victim and is motivated by rage toward the injured party or who the victim symbolizes. The assailant uses considerable physical violence, subjects his victims to a series of degrading sexual acts, and typically causes significant harm (Kocsis et al., 2002). The power rapist is motivated by a need to assert authority and dominance over his victim, and the level of force will likely depend on the degree of victim compliance (Pardue and Arrigo, 2008). The sadistic rapist eroticizes violence. Sexual and aggressive urges are fused and the offender is aroused by the degradation, pain, suffering, torture, and possible mutilation of the victim (Shipley and Arrigo, 2008). The pain and suffering of the victim is not incidental; it is central. This offender typically uses restraints, bondage, engages in paraphilias, torture, mutilation, and significant physical injury to the victim, which can escalate to death (Purcell and Arrigo, 2006). The three categories were converted into a fourfold
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
198
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
schema and include: power-reassurance, power-assertive, anger-retaliatory, and anger excitation rapists. The absence of adequate empirical validation remains the chief criticism of the reformulation; however, these categories are frequently used among those who investigate sexual crimes. Hazelwood and Burgess (1987) further developed the categories into four rapist profiles. The first typology is the power-reassurance rapist who has significant feelings of inadequacy and wants to restore feelings of masculinity. This type of rapist frequently is concerned about avoiding harm to the victim, is apologetic, makes efforts to reassure the injured party, and uses less intrusive forms of rape behavior (Warren et al., 1991). He typically relies on the threat of a weapon and may not even have one. The assailant is often known as the “gentleman rapist” (Hazelwood, 2005). He is looking for reassurance of his masculinity, is less aggressive than other rapists and his objective is to have sexual intercourse. His attacks are premeditated and fueled by persistent rape fantasies. He enjoys pretending that the rape is consensual and that the victim takes pleasure in the assault. The assailant spends a great deal of time with his victim and attempts to engage in pillow talk, asking personal questions and talking about himself. The second typology is the power-assertive rapist who uses the assault as confirmation of his virility, machismo, and ability to dominate women. He does not need reassurance; rather, he needs to display power and control (Hazelwood and Burgess, 1987). His attacks may be impulsive and unplanned. His use of force is moderate and dependent on victim resistance. He frequently uses his fists as weapons and the assault is short-lived. He is arrogant, hot headed, and has a history of conflict with women. This offender sees himself as very socially competent; however, he likely has a history of failed marriages (National Center for Women and Policing (NCWP), 2004). The third typology is the anger-retaliatory rapist who is motivated to victimize as an extension of anger and rage (Warren et al., 1991). He is likely to overtly express anger by using excessive force, relying on a knife, and projecting a macho image (McCabe and Wauchope, 2005). The rage is not sexualized and these rapes do not appear to be motivated by compulsive, sadistic fantasies. Women are viewed as whores or as dirty and cannot be trusted (NCWP, 2004). The assailant likely uses a “blitz” style of attack or quickly overpowers the victim with surprise and extreme force. The attack usually results in a severe beating and can end in murder, but this is not typical. The victim’s resistance can result in a more savage assault as it fuels the assailant’s anger. The attacker’s rape may represent displaced anger in that the assailant is reacting to a previous fight with a girlfriend, lover, or wife and causes harm to a victim to symbolically even the score with a woman perceived to have wronged him. The rape can be used to punish or humiliate and often represents a general disdain for females. The final typology is the anger-excitation rapist who is likely sadistic and enjoys the pain and suffering of his victim. This type of rapist also has been described as a sadistic-ritualistic assailant (NCWP, 2004). He is more likely to bind and transport victims, to not reassure them, and will likely try to terrorize those on whom he preys. He accomplishes this by telling his victims how he will hurt them. His interactions with his victims are detached, and he takes satisfaction from their reactions. These rapes entail greater planning and the behavior surrounding the offense is less impulsive (Warren et al., 1991). Hazelwood and Burgess (1987)
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
199
indicated that this type of rapist is sexually stimulated or gratified by the victim’s response to or suffering from physical or emotional pain. The primary motivation is to inflict harm and to induce fear and total submission. Hazelwood and Burgess (1987) further described the offender as using a brutal level of force that often resulted in death. Rather than a blitz style of attack, this rapist is manipulative, often confidently using verbal tactics or tricks to gain victim access. Victims are often strangers but are symbolic of those in the assailant’s compulsive sadistic fantasies. He is likely to be married but his wife is firmly under his control and may be a compliant victim herself (NCWP, 2004).
Sexual Assault of Adult Males In 2006, a total of 260, 940 rape/sexual assault victimizations were reported to United States law enforcement, and approximately 26.2% or 68,366 of the victims were males (US Department of Justice, 2008). According to the 2000 National Incident-Based Reporting System (NIBRS), males make up about nine out of every ten adult sexual assault perpetrators, totaling about 26, 878 incidents within this reporting period; however, men can become victims of sexual assault, and the perpetrator’s tactics can vary from violent attacks with weapons to deliberate intoxication or verbal coercion. Those who engage in sex crimes against men are almost always male themselves. The majority of perpetrators are heterosexual and Caucasian, and most studies indicate that the majority (58% to 100%) of male victims are also Caucasian and relatively young (Frazier, 1993). Sexual offenders who assault stranger male victims are more likely to use weapons and physical violence (Stermac, Del Bove and Addison, 2004; Frazier, 1993). Studies reveal that male victims of sexual assault tend to be young, single adults who are more likely to be vulnerable given psychosocial issues such as homelessness, as well as physical, psychiatric, and cognitive disabilities (Stermac et al., 2004). These investigators reported that male victims were more likely to be subjected to anal/or fellatio assault. Moreover, male stranger perpetrator victims were more likely to be attacked by multiple assailants than were female victims, and they were more likely to be assaulted outdoors or in a park setting. Injuries suffered were similar for male and female victims, and studies have found that most male victims are known to their perpetrators as the offender typically is an acquaintance (Stermac et al., 2004). Prior research on rape myths for male victims focused on beliefs such as “getting raped doesn’t really upset men” and “male rape cannot happen” (Anderson, 1999, p. 390) or that men are less affected by sexual assault than women (Stermac et al., 2004). Kassing, Beesley and Frey (2005) described and critiqued rape myths that focus on sexual orientation; specifically, that male perpetrators are gay and that victims bring the attack upon themselves by presenting themselves as homosexual, and by engaging in high risk or reckless behavior.
Female Perpetrators with Adult Victims Limited information is available in the literature on female perpetrated sexual offenses against male or female adults (Struckman-Johnson and Struckman-Johnson,
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
200
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
2006). Almost all available research on female sexual offenders involves child or teenage victimization. In part, this is due to victim accessibility and the perpetrator’s ability to control them. For example, Strickland (2008) studied a sample of 130 incarcerated females (60 sex offenders and 70 nonsex offenders) in Georgia prisons. She found that female sex offenders encountered significantly higher rates of total childhood trauma than their male counterparts. Most sadistic offenders are male. While mostly rare, sadistic females typically are exposed to this dysfunctional behavior in adolescence and, as they mature, derive pleasure from it through an adult relationship with a masochistic man (Scott, 1983). There are some adult females who are coerced into sexually assaulting adult victims by a dominant, sadistic, male partner. These women tend to be very dependent and have experienced significant abuse from their partners whom they fear will abandon them if they do not participate in acquiring and assaulting victims. The probability of these women independently undertaking sexual assaults is low. Despite evidence that the sexual victimization of males by female assailants does exist (Banbury, 2004), society typically perceives the rape of men by women as largely implausible (Davies, 2002). Rape myths and gender role stereotypes support the notion that a woman cannot force a man to have sex (StruckmanJohnson and Struckman-Johnson, 2006). Moreover, as Davies (2002) explained, “It is also difficult to believe that men can become sexually aroused and even ejaculate during a sexual assault, but still report that they did not want the situation to take place, tried to stop it, and felt fear and disgust during and after the assault” (p. 206).
ASSESSMENTS Risk Assessment for Sex Offender Recidivism Once a sexual offender has been prosecuted, the probability of rehabilitation and the appropriate place for it to occur are discussed. Issues such as the degree and type of deviant sexual arousal, the frequency and severity of the known offenses, the presence or absence of major mental illness, and the presence and degree of prosocial deficits are examined by clinicians with the requisite expertise in these areas. According to Sbraga (2003), forensic mental health questions at this juncture include the following: Is there a paraphilia? How severe is it? At what stage in the punishment/ rehabilitation process should the inmate receive treatment? And, does the paraphilia render the inmate so dangerous that treatment is mandatory prior to release? Once the inmate has entered treatment, the queries turn toward treatment gains, less restrictive alternatives for treatment, and risk of reoffending. (p. 434).
Rape and sexual sadism correlate with a high risk of reoffending and greater risk of physical harm to victims (Marshall et al., 2008). Forensic clinicians are called upon to evaluate an offender for dispositional evaluations, at intake prior to treatment, or to conduct treatment with those who may benefit. Forensic ethics stress the importance of avoiding dual relationships (Shipley and Arrigo, 2004).
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
201
In other words, those who conduct evaluations for the court should not be those who provide the court-ordered treatment. Sbraga (2003) indicated that the separation of these functions is necessary in order to maintain objectivity and to avoid damaging the therapeutic relationship. Prior to beginning treatment, there is often some assessment to gauge where the offender is with regard to thinking errors, sexual deviance, and the like. The degree of sexual deviance is also used to determine postconviction disposition and treatment (e.g., release, parole, the degree of monitoring, or civil commitment). When forensically assessing an adult sex offender, the evaluator should conduct a detailed psychosocial interview in addition to administering any instruments that can help organize the evaluation. Information that has been provided by the offender should be identified as such in the report. The historical information provided by this self-report data should be viewed with skepticism, given the secondary gains in denying or minimizing one’s offense behavior and deviant sexual fantasies. A thorough review of collateral information (e.g., police reports, victim statements, mental health records, etc.) is essential. Records such as police reports should be reviewed to determine a history of criminal activity and to avoid the pitfalls associated with using a single source to formulate an opinion or to make recommendations. Psychological testing can also provide data about personality characteristics, the presence or absence of major mental illness, and additional information about thinking errors and the like. Assessments of various types of sexual offenders in both outpatient and inpatient treatment programs often use the polygraph and penile plethysmograph testing; however, the literature is mixed regarding physiological assessments with some reviews suggesting unclear reliability, and the validity on its use for rapists is questionable (Marshall et al., 2008). Sex offender risk assessment is typically pursued by forensic clinicians to help predict the probability of future sexual violence based on risk factors that relate to sexual offending. As with nonsexual violence, the best predictor of future sexual offending is past sexual violence. The risk assessment addresses the characteristics, frequency, and severity of prior offenses, and estimates the likelihood, imminence, and seriousness of potential future offenses. The standard practice and the preferred approach in the assessment of risk for sex offenders is to use a combination of evaluative strategies. These include structured clinical judgment guided by a strong understanding of the relevant literature and professional experience with this population, and the use of actuarial instruments. Some actuarial instruments focus on static risk factors or those that are based on past history and cannot be changed, and others consider both static and dynamic (changeable) features (Marshall et al., 2008). Whenever possible, an actuarial approach is strongly recommended given the lack of predictive accuracy of clinical judgment alone (Quinsey et al., 1998). While the list is not exhaustive, the following assessment tools may be used in the prediction of sexual reoffending:
r The Sex Offender Risk Appraisal Guide (SORAG) is a 14 item instrument that uses both static (historical or unchangeable) and dynamic (changeable) risk factors (Quinsey, et al.. 1998). One criticism is that this instrument was designed using a limited sample of sexual offenders.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
202
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
r The STABLE-2007 consists of a guided interview process covering 13 major ar-
r r r r
r r r
eas of stable dynamic risk factors in the offender that are amenable to change; for example, deviant attitudes and sexual preoccupation (Hanson et al., 2007). Marshall et al. (2008) indicated that this instrument identifies features useful for treatment. The ACUTE-2007 (Hanson et al., 2007) identifies changing factors that increase immediate risk of reoffense such as the sudden onset of negative moods, recent interpersonal conflicts, and immediate access to a victim (Marshall et al., 2008). The Minnesota Sex Offender Screening Tool-Revised (MnSOST-R) is composed of 16 items that include both static and dynamic risk factors. The Rapid Risk Assessment for Sexual Offense Recidivism (RRASOR) was created by Hanson (1997) to predict sex offender recidivism. The STATIC-99 is a static 10-item actuarial assessment instrument with a weighted key. It was developed by Hanson and Thornton (1999) for use with adult male sexual offenders who are at least 18 year of age at the time of release to the community. The Sex Offender Needs Assessment Rating (SONAR) was designed in order to apply actuarial methods to dynamic variables (Hanson and Harris, 2000). This 9-item scale was designed to measure change in risk level for sexual offenders. The Sexual Violence Risk-20 (SVR-20) is a 20 item guide for assessing violence risk in sex offenders (Boer et al., 1997). The Hare Psychopathy Checklist-Revised (PCL-R) 2nd edition (2003) is used to assess for psychopathy. The PCL-R was not designed to predict recidivism or violence; however, good empirical evidence supports that psychopathy is a strong predictor of violence to include sexual violence. The PCL-R 2nd edition is either used on its own to assess for psychopathy or as part of a risk assessment battery (e.g., with the HCR-20). Psychopathy should be considered in all sexual violence risk assessments.
Other Common Assessment Tools As well as assessment tools for risk assessment, a number of other assessments are also commonly used. The decisions about what assessment instruments are to be used by the evaluating psychologist or by an appropriately trained forensic clinician should be guided by the referral question and by evaluative tools that have the best validity and reliability for what they intend to measure. Widely used psychological tests that evaluate psychopathology or personality traits, such as the Minnesota Multiphasic Personality Inventory-2 (MMPI-2), the Millon Clinical Multiaxial Inventory-III (MCMI-III), or the Personality Assessment Inventory (PAI), will not be detailed in this chapter. They are frequently employed in psychological evaluations to aid in the differential diagnosis of psychiatric disorders and/or personality profiles. Additionally, there are various instruments that can be used specifically in the assessment of adult sex offenders. These include the Multiphasic Sex Inventory – Adult Male Form (MSI II), which is designed to measure the sexual deviance characteristics of an adult male alleged to have engaged in sexual misconduct or a sex crime; the Sexual Fantasy Questionnaire (SFQ), which is a self-report
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
203
measure that examines the type and degree of sexual fantasy content (O’Donohue, Letourneau and Dowling, 1997), and the Aggressive Sexual Behavior Inventory (ASA), which examines both sexual and aggressive factors.
CASE FORMULATION In order to illustrate the interplay of risk assessment and case formulation of sexual offenses against adults we present the cases of Mr Smith, a 50-year-old Caucasian male, who was assessed in prison after conviction for the offense of Aggravated Sexual Assault. Mr Smith had been sentenced to a maximum-security correctional facility and his treatment needs were to be evaluated and services provided as deemed appropriate. A routine intake evaluation was conducted to quickly assess whether there was a need for further evaluation and for mental health follow-up. Due to his reported symptoms, documented past accounts of mental illness, and history of serious sex offenses, Mr Smith was referred for an evaluation to determine his current level of risk for sexual violence, assess issues that relate to his sexually deviant behavior, identify his goals for treatment, and determine if he was appropriate for placement in sex offender treatment group(s). He had a significant history of sexually deviant behavior in both institutional and community settings, as evidenced by his current charge, other prior arrests for sexual offenses, and behaviors during his incarceration. The latter included nonconsensual touching/groping of female staff and an actual attempt to sexually assault a female staff member whom he later disclosed he intended to rape.
Assessments Mr Smith’s treatment and level of security were based on his risk factors, ability to benefit from treatment, and the degree of risk the evaluator has predicted the offender would pose to the institution and eventually the community.
Risk Assessment Psychopathy Mr Smith’s overall score on the PCL:SV was 20 (91.7th percentile), which suggested that he may be psychopathic and should be further evaluated with the full PCL:R. Scores of 18 or higher offer a strong indication of psychopathy. His score on Part 1 was 9 (83.3rd percentile). This represents the affective and interpersonal characteristics associated with psychopathy such as malignant narcissism, a lack of empathy, and the selfish, callous, and remorseless use of others. His score on Part 2 was 11 (95.8th percentile). This represents a chronically unstable and antisocial lifestyle or social deviance. Out of 100 individuals in a forensic psychiatric population, only 4 would score higher on Part 2. While Mr Smith could not be classified as a psychopath with the PCL:SV, he possessed many psychopathic traits and psychopathy could not be ruled out. Mr Smith’s hospital and criminal records were utilized but they were available for only a specific period of time. Moreover, there were insufficient records to accurately
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
204
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
complete the PCL-R. When additional collateral information was gathered, his scores on the PCL:SV strongly suggested that the PCL:R should be completed. SVR-20/Risk Assessment The SVR-20 is an assessment method used to organize risk factors that correlate with sexual violence. Sexual violence is defined as actual, attempted, or threatened sexual contact with a person who is nonconsenting or unable to give consent. The SVR-20 consists of 20 items organized around 11 Psychosocial Adjustment variables, 7 Sexual Offenses variables, and 2 Future Plans variables. Mr Smith was evaluated to be in the high-risk category. He had 11 out of 11 Psychosocial Adjustment risk factors, 7 out of 7 Sexual Offenses risk factors, and 2 out of 2 Future Plans risk factors. Mr Smith had the following risk factors: (a) Risk Factors (Psychosocial Adjustment): Sexual Deviation, Relationship Problems, Victim of Child Abuse, Past Nonsexual Violent Offenses (“serious”), Past Nonviolent Offenses, Substance Use Problems, Employment Problems, Psychopathy (possible/traits), Major Mental Illness, Suicidal/Homicidal Ideation and Past Supervision Failure; (b) Risk Factors (Sexual Offenses): High Density Offenses, Multiple Offense Types, Physical Harm to Victim(s), Uses Weapons or Threats of Death, Escalation in Frequency/Severity, Extreme Minimization/Denial and Attitudes Condone Offenses; and (c) Risk Factors (Future Plans): Negative Attitude Toward Intervention and Lacks Realistic Plans.
Summary of Risk Assessment Based on the results of the current risk assessment, should Mr Smith have been released to the community or into a facility or situation with little supervision, he would have posed a high risk for sexually violent recidivism. Mr Smith had a previous history of sexual violence including Aggravated Sexual Assault, Indecent Exposure, an attempted sexual assault in the prison, numerous incidents of exhibitionism, and threats to engage in deviant sexual behavior. Based on his past history, Mr Smith’s level of risk increases when he is under the influence of alcohol or other illicit substances, when he experiences auditory hallucinations or symptoms of mania, or when he becomes depressed or angry, particularly with women. Risk may be reduced with additional environmental controls, such as an increased level of structure, the lack of an available victim pool (e.g., no one-to-one contact with female staff), no access to weapons, and preventing drug acquisition. Medication management also helps to improve his impulsivity and symptoms of mental illness.
Assessment for Case Formulation A clinical interview was conducted consisting of a sexual biography, mental status examination, and a records review. In order to effectively formulate his case, understand his offense pathways, identify their possible causes, and develop appropriate treatment options, a comprehensive examination of Mr Smith’s relevant history was essential. Awareness of his sexual autobiography and developmental
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
205
history assisted the evaluator and, eventually, the therapist in recommending and undertaking offender therapy. Moreover, this insight enabled both the treatment provider and the client to recognize those factors that led to sexual violence, as well as to specify the warning signs that occurred along his offense pathway. The mental status examination identified current symptoms of psychosis and mania to be treated with psychotropic medication. The records reviewed provided collateral information on criminal history, witness and victim statements, prior symptoms, and other relevant past behaviors.
Developmental History Mr Smith’s father was a chronic alcoholic and was physically abusive toward his wife and son. His mother was emotionally distant and Mr Smith did not feel close to either of his parents. His parents divorced when he was about 16 years old. Following the divorce, he lived with his mother in a single-parent household. Mr Smith reported that he was sexually abused, including penetration, when he was 4 or 5 years of age by a paternal uncle. Mr Smith indicated that he continues to hear the voice of the uncle who molested him saying derogatory things. When asked how the sexual abuse affected Mr Smith, he stated, “sad, angry, embarrassed”. He said he did not report the abuse as a child. Subsequently, Mr Smith had very insecure and poor attachments with his parents who either physically abused or neglected him. Mr Smith stated that he had problems relating to others and fought with his peers in school. Mr Smith was married for four years immediately after high school and had two children. Mr Smith has had no contact with either his ex-wife or his children for numerous years. He reported that he was living with his mother prior to his most recent arrest and has been violent with his mother in the past. He showed considerable lack of interest in relating to others, as well as a general tendency to harbor anger toward women. His father modeled physical aggression and his mother did nothing to protect him from violence or to comfort him. He was quite impaired in his ability to connect with and trust others and does not appear to be motivated to engage in meaningful relationships. Cognitive distortions or thinking errors typical of sexual offenders were present. These were exacerbated by symptoms of mental illness, which further disinhibited his behavior and increased the likelihood that he would impulsively and aggressively reoffend. On the other hand, when his symptoms of mental illness responded to medication management and treatment, he became a more organized sexual offender and his offenses were more predatory in nature. Mr Smith attended school until he was in the 10th grade (age 16 years). His prior intellectual testing indicated that his IQ functioning fell in the Low-Average range. Mr Smith has been unable to maintain gainful employment for any significant period of time, working sporadically in construction or as a day laborer. Mr Smith previously disclosed using alcohol, cocaine, marijuana, and LSD.
Legal History Mr Smith’s legal history included multiple arrests. His record specifies that he was first charged with Burglary of a Residence when he was 24 years old. He had spent
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
206
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
a large portion of his adult life imprisoned or hospitalized in structured, supervised settings. He had prior convictions for Aggravated Robbery with a Deadly Weapon, Burglary of a Habitation (twice), Indecent Exposure, Weapons Offense, Possession of Cocaine, and the current offense of Aggravated Sexual Assault. One of his convictions for Burglary of a Habitation involved the following circumstances: Mr Smith attempted to open the door of a female stranger’s residence. When he was unable to do so, he went to the side gate but her dogs alerted her at 2:45 a.m. He then returned to the front of the house and was arrested after it was discovered that he had attempted to remove the screen from the woman’s bedroom window. Mr Smith was most recently arrested for Aggravated Sexual Assault. He approached the unknown, 23-year-old, female victim at night as she was about to enter her home carrying groceries. He reportedly asked her if she would like some help carrying her things into her house. She said “no”, but Mr Smith forced her in at knife point. He shoved her over a couch and punched her in the face a couple of times while pushing her toward the bedroom. The victim stated that he knocked her head into the wall and then he picked her up and threw her onto the bed. She screamed and he told her that he would “shut her up if he had to cut her up”. Mr Smith vaginally raped her while making derogatory statements. The victim reported that he acted dazed and “high” during the assault. He reported that he was not taking his medicine and was using cocaine at the time of the offense. Mr Smith told her that if she reported the offense he would kill her. He eventually left her home, insisting that she close her eyes and lie face down on the bed. When she was sure he was gone, the police were called and a sexual assault examination was performed at the hospital. Mr Smith stood trial and was sentenced to 10 years in the state Department of Corrections. In regard to Mr Smith’s sexual history, he indicated that his first sexual experience was “my family doing weird shit”. When asked about how he learned about sex, he stated, “watching my father and mother”. He denied having current sexual fantasies about underage persons or nonconsenting persons. Based on his history and previous reports of fantasizing about exposing himself to adult women, this statement was suspect. He denied currently masturbating to sexually deviant fantasies. He reported several other instances of exhibitionism for which he was not charged. When asked about his preferences with regard to sexual partners or potential victims, he stated, “skinny, thin, Caucasian women . . . age 20 or so . . . I’m not a child molester . . . I don’t mess with kids . . . just adults”.
Issues for Treatment Mr Smith has a history of mental illness and substance abuse. When he initially arrived at the correctional facility, he presented with paranoid delusions, labile and angry affect, illogical thinking, hypersexuality, and confusion of thought. He had a history of being diagnosed with Schizoaffective Disorder, Bipolar Type and Antisocial Personality Disorder. He exposed himself to female correctional officers and exhibited aggressive and threatening behaviors. He eventually responded adequately to his medication regimen, although, all symptoms had not resolved entirely. His thinking became more organized, his emotions were better regulated, and his demeanor was more relaxed.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
207
Mr Smith readily admitted his offense(s) and was willing to participate in a comprehensive sex offender treatment program. He expressed a significant number of criminal thinking errors or cognitive distortions combined with having command hallucinations that were congruent with his sexually deviant thoughts and behaviors. This put him at a higher risk of reoffense. Medication improved his disorganized thinking but redirection and offering alternatives were still frequently needed. Mr Smith attended a sex offender treatment group based on a cognitivebehavioral framework, as well as incorporating the relapse prevention model, and psycho-educational modules on issues such as anger management and social skills training. His treatment program is empirically based in that it includes standard of practice elements of sex offender treatment such as: (1) acceptance of responsibility (e.g., denial/minimization, cognitive distortions, and victim empathy); (2) offense pathways (e.g., early warning signs, dangerous situations, and relapse prevention plans); (3) coping skills; (4) intimacy and attachments, (5) sexual arousal, fantasies, and preferences, and (6) mood and symptom management. Initially, Mr Smith sometimes appeared internally preoccupied. He responded to redirection and he improved in this regard as his psychotropic medications continued to be adjusted by his psychiatrist. Mr Smith indicated that he was experiencing auditory hallucinations in the form of voices but he felt that he was better able to cope with them and did not have to do what the voices told him to do. He also disclosed that substance abuse caused an increase in his anger and was a significant risk factor for reoffense of nonsexual and sexual violence. Initially, he described having sexual thoughts almost every day but indicated that he had not acted on them or masturbated to them. He also reported that he continued to hear voices at times telling him to commit a sexual offense (e.g., “to touch a woman”), but he continued to indicate that he was able to resist acting on those commands or other deviant impulses. While in the group, Mr Smith participated but his progress was limited by the intermittent reemergence of psychiatric symptoms and/or a pattern of disregarding the impact of his actions on others. It was particularly challenging to differentiate and address criminal thinking errors or cognitive distortions that are present independent of psychosis and those symptoms of psychosis that were present with similar themes.
FORMULATION Mr Smith was a 50-year-old, divorced, Caucasian male with a history of physical and sexual abuse, poly-substance abuse, and an extensive psychiatric and legal history. His current incarceration was precipitated by the instant offense of Aggravated Sexual Assault. Mr Smith’s attack was premeditated and predatory. Based on the means by which he gained control over his victim, the degree of physical force, how he interacted with his victim, and other motives and behavioral patterns, he could be classified as an anger-retaliatory rapist. Mr Smith’s childhood was marked by fear and abuse. He did not form healthy attachments and he learned to identify with the aggressor(s). His sexual
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
208
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
victimization at an early age coupled with the violence in the home created a very unstable sense of self, maladaptive relationships with others, and dysfunctional sexual development. He described a protracted history of inappropriate sexual fantasies to which he masturbated. It is apparent he developed a preference for sexual activity with nonconsenting, adult females. His history of substance abuse; academic and social failures; criminal behaviors; and rejection from parents, spouse, and peers have all contributed to his offense pathway. Mr Smith views his world as a hostile place where you take what you want and you “get them before they get you”. Moreover, he appeared to use coercive sex as a coping strategy for negative mood states or when he feels powerless or humiliated in order to reestablish his feelings of control and power. He was likely to be very opportunistic about who he chose as a victim (e.g., easiest target); although, if his preferred victim type were available, this would trigger him to act. Mr Smith felt sexual arousal or a high from violent fantasy and behavior. His psychosis, impulsivity, and deviant sexual urges were currently so extreme that his daily functioning was greatly impaired and his risk for nonsexual and sexual violence was high. He has escalated from indecent exposure to a violent, potentially sadistic rape and another attempted rape that was premeditated. It appeared that Mr Smith’s sexual assaults increased in the degree of violence used during the attack. There was a compulsive element in his behavior but he appears to have an absence of feeling and no desire to conform. There was a lack of psychological attachment or closeness to others. Drinking heavily or using cocaine made him even more likely to become selfish, unreasonable, and forceful in order to regain control of the victim. The exercise of verbal intimidation to gain power and secure dominance also spoke to his motivation and was self-revelatory. His derogatory, profane, and hostile language during the assault indicates anger toward women and his use of brute sexual behavior signified an urge to punish and degrade them. Mr Smith has a history of being aggressive with his mother. Mr Smith likely saw his victims as objects but also has a proclivity for using sex as a weapon to harm and humiliate. This mirrors what he felt and experienced as a young boy when abused sexually by an uncle and physically by his father. Mr Smith has an explosive temper, lower intelligence, does not trust others, lacks social skills, has a low socioeconomic status and leads a highly impulsive lifestyle.
Criteria for Success/Expected Outcome The ultimate goal is for Mr Smith not to reoffend in the correctional setting or in the community. Given the longstanding nature of Mr Smith’s focus on deviant sexual behavior, his resistance to treatment, and his history of offending, the most effective way to manage his risk for sexual violence is to deny him access to potential victims and to house and treat him in a highly structured environment with careful monitoring. His deviant sexual fantasies are not likely to remit, but he may become less impulsive with regard to acting on these fantasies or behaviors if his psychiatric symptoms improve. Tracking and monitoring the frequency of inappropriate sexual behaviors with the expectation that any such behaviors will cease is one expected treatment outcome. Both in an institutional setting and
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
209
in the community, Mr Smith’s risk management strategies should focus on intensive supervision, psychiatric intervention, and environmental separation from any potential victims. Female staff should never engage him alone. A measure for success is Mr Smith’s ability to identify and then avoid or reframe any triggers, thoughts, feelings, behaviors, and high-risk situations that might ready him to sexually reoffend. As a result of his dual diagnosis and history of mania and psychotic symptoms, Mr Smith requires psychiatric services to include psychotropic medications and possible hospitalization if his symptoms become more severe. Mr Smith has traits of psychopathy but he was not definitively classified as such. To deny him certain aspects of his treatment without more evidence could cause more harm than good. If additional data were provided or if his actions conveyed the same, several treatment approaches in his plan would need to be revisited and some might need to be changed altogether. Individuals clearly identified as psychopathic or sadistic are not appropriate candidates for sex offender group therapy, particularly in a correctional environment. These individuals can mimic emotions, such as empathy or remorse, but if you scratch below the surface the fac¸ade crumbles and it quickly becomes apparent that they are going through the motions without any real foundation for what it genuinely means to express or experience these feelings (Shipley and Arrigo, 2008). The formulation identified a number of important proximal antecedents to aggression, exposure to females and rape including: hearing voices, manic symptoms, alcohol and cocaine abuse, and angry and depressed mood. Thus, his treatment program addressed these antecedents in a number of ways. For example, psychiatric symptoms were addressed by medication, promoting medication compliance and cognitive behavior therapy. Cognitive behavior therapy also was used to teach him alternate strategies when dealing with angry and depressed mood. Stable antecedents, such as cognitive distortions and indifference to other people, were addressed by cognitive restructuring and attempts to promote empathy, intimacy and attachment to others. Risk management also dictated reduction of risk by removal of opportunities to harm through supervision and reduced access to trigger situations, such as being alone with women. Thus, intervention required a balance of strategies to promote other, appropriate behavior and limiting opportunities. In evaluating Mr Smith’s response to treatment goals and in amending, if necessary, treatment planning, it is important to incorporate relevant measures of behavioral change including improvement in depression or hypersexuality stemming from mania, and demonstrated use of effective coping skills in managing sexual behavior and in managing other environmental stressors without resorting to exhibitionism or brute force.
CONCLUSIONS The sexual assaults of adult women and men are traumatic and destructive as evidenced by the enormous cost they exact from victims, their families, the mental health and justice systems, and society in general. The range of assessment, diagnostic, treatment, and programming issues as developed in this chapter amply demonstrate the complexities of forensic case formulation for adult sexual
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
210
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
offenders. These significant concerns notwithstanding, McGrath’s (1991, p. 346) caution is worth noting. As he indicated, Disposition plans should not be influenced by misinformation, politics, or fear, but should be solidly grounded in clinical experience, empirical knowledge, and availability of specialized resources. Rehabilitation of offenders who show potential for change is imperative, while protection of the community must remain a continual priority (p. 346.)
In the final analysis, this is the purpose of any sound clinical case formulation and forensic evaluation. This includes those individuals whose underlying behavior is sexual, deviant, and criminal in nature.
REFERENCES Anderson, I. (1999) Characterological and behavioral blame in conversations about female and male rape. Journal of Language and Social Psychology, 18(4), 377–94. Arrigo, B.A. and Shipley, S.L. (2005) Introduction to Forensic Psychology (2nd edn): Issues and Controversies in Law, Law Enforcement, and Corrections. San Diego, CA: Elsevier/Academic Press. Arrigo, B.A. and Shipley, S.L. (2001) The confusion over psychopathy (I): Historical considerations. International Journal of Offender Therapy and Comparative Criminology, 45(3), 325–44. Banbury, S. (2004) Coercive sexual behaviour in British prisons as reported by adult exprisoners. The Howard Journal, 43(2) May, 113–30. Beere, C.A. (1990) Sex and Gender Issues: A Handbook of Tests and Measures. Westport, CT: Greenwood Press. Boer, D.P., Hart, S.D., Kropp, P. R. and Webster, C.D. (1997) Manual for the Sexual Violence Risk – 20: Professional Guidelines for Assessing Risk of Sexual Violence. Vancouver, British Columbia: British Columbia Institute on Family Violence and Mental Health, Law, and Policy Institute, Simon Fraser University. Brand, P. and Kidd, A. (1986). Frequency of physical aggression in heterosexual and female homosexual dyads. Psychological Reports, 59, 1307–13. Burgess, A.W., Hazelwood, R.R., Rokous, F.E. et al. 1988) Serial rapists and their victims: Reenactment and repetition. Annals of New York Academy of Sciences, 528, 277–95. Burt, M.R. (1980) Cultural myths and supports for rape. Journal of Personality and Social Psychology, 38, 217–30. Davies, A. (1992) Rapists’ behavior: A three aspect model as a basis for analysis and the identification of serial crime. Forensic Science International, 55, 173–94. Davies, M. (2002) Male sexual assault victims: a selective review of the literature and implications for support services. Aggression and Violent Behavior, 7, 203–14. Elliott, D.S. (1994) Serious violent offenders: Onset, developmental course and termination. The American Society of Criminology 1993 presidential address. Criminology, 32, 1–21. Epperson, D.L., Kaul, J.D. and Hesselton, D. (1998, October). Final report of the development of the Minnesota Sex Offender Screening Tool-Revised (MnSOST-R). Presentation at the 17th Annual Research Conference of the ATSA, Vancouver, BC, Canada. Frazier, P.A. (1993) A comparative study of male and female rape victims seen at a hospitalbased rape crisis program. Journal of Interpersonal Violence, 8, 64–76. Groth, A., Burgess, A. and Holmstrom, L. (1977) Rape, power, anger and sexuality. American Journal of Psychiatry, 134, 1239–43. Hanson, R.K. (1997) The Development of a Brief Actuarial Risk Scale for Sexual Offense Recidivism. Report no. 1997-04. Ottawa: Office of the Solicitor General of Canada.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
211
Hanson, R.K. and Harris, A.J.R. (2000) Where should we intervene? Dynamic predictors of offense recidivism. Criminal Justice and Behavior, 27, 6–35. Hanson, R.K., Harris, A.J.R., Scott, T-L. and Helmus, L. (2007) Assessing the Risk of Sexual Offenders on Community Supervision: The Dynamic Supervision Project. User Report No. 2007-05, Corrections Research, Ottawa: Public Safety Canada. Hanson, R.K. and Thornton, D. (1999) STATIC 9: Improving Actuarial Risk Assessments for Sex Offenders. Report No. 1999-02. Ottawa: Office of the Solicitor General of Canada. Hazelwood, R. (2005) Key Motives and Behaviors in Sexual Homicide. Presentation at the Homicide: Behaviors, Motives and Psychology: A Gathering of Leading Experts Conference, San Diego, CA. Hazelwood, R. and Burgess, A. (1987) An introduction to the serial rapist: Research by the FBI. FBI Law Enforcement Bulletin, September, 16–24. Hazelwood, R. and Warren, J. (1989a) The serial rapist: His characteristics and victims (Part I). FBI Law Enforcement Bulletin, January, 11–17. Hazelwood, R. and Warren, J. (1989b) The serial rapist: His characteristics and victims (Part II). FBI Law Enforcement Bulletin, February, 19–25. Karabatsos, G. (1997). The Sexual Experiences Survey: Interpretation and validity. Journal of Outcome Measures, 1(4), 305–28. Kassing, L.R., Beesley, D. and Frey, L.L. (2005) Gender role conflict, homophobia, age, and education as predictors of male rape myth acceptance. Journal of Mental Health Counseling, 27(4), 311–28. Knight, D.A. and Sims-Knight, J.E. (2003) Developmental antecedents of sexual coercion against women: Testing of alternative hypotheses with structural equation modeling. In R.A. Prentky, E.S. Janus and M. Seto (eds), Sexual Coercive Behavior: Understanding and Management (pp. 72–85). New York: New York Academy of Sciences. Kocsis, R.N., Cooksey, R.W. and Irwin, H.J. (2002) Psychological profiling of offender characteristics from crime behaviors in serial rape offences. International Journal of Offender Therapy and Comparative Criminology, 46(2), 144–69. Kuhn, J.A., Arellano, C.M. and Chavez, E.L. (1998) Correlates of sexual assault in Mexican American and White non-Hispanic adolescent males. Violence and Victims, 13, 11–20. Laws, D.R. and Marshall, W.L. (1991) Masturbatory reconditioning with sexual deviates: An evaluative review. Advances in Behavior Research and Therapy, 13, 13–25. Levenson, J.S. and Prescott, D.S. (2007) Considerations in evaluating the effectiveness of sexual offender treatment: Incorporating knowledge into practice. In D.S. Prescott (ed.), Knowledge and Practice: Challenges in the Treatment and Supervision of Sexual Abusers (pp. 124–42). Oklahoma City, OK: Wood & Barnes. ¨ Losel, F. and Schmucker, M. (2005) The effectiveness of treatment for sexual offenders: A comprehensive meta-analysis. Journal of Experimental Criminology, 1, 117–46. Lussier, P., Leclerc, B., Cale, J. and Proulx, J. (2007) Developmental pathways of deviance in sexual aggressors. Criminal Justice and Behavior, 34(11), 1441–62. McCabe, M.P. and Wauchope, M. (2005) Behavioral characteristics of men accused of rape: Evidence for different types of rapists. Archives of Sexual Behavior, 34(2), 241–53. McGrath, R.J. (1991) Sex-offender risk assessment and disposition planning: A review of empirical and clinical findings. International Journal of Offender Therapy and Comparative Criminology, 35(4), 328–50. Marshall, W.L. and Barbaree, H.E. (1990) Outcome of comprehensive cognitive behavioral treatment programs. In W.L. Marshall, D.R. Laws and H.E. Barbaree (eds), Handbook of Sexual Assault: Issues, Theories, and Treatment of the Offender (pp. 363–85). New York: Plenum Press. Marshall, W.L. and Marshall, L.E. (2000) The origins of sexual offending. Trauma, Violence, and Abuse, 1, 250–63. Marshall, W.L., Serran, G.A., Marshall, L.E. and O’Brien, M.D. (2008) Sexual deviation. In M. Hersen and J. Rosqvist (eds), Handbook of Psychological Assessment, Case Conceptualization, and Treatment: Volume 1 Adults (pp. 590–615). New Jersey: John Wiley & Sons, Inc. Masho, S.W., Odor, R.K. and Adera, T. (2005) Sexual assault in Virginia: A population-based study. Women’s Health Issues, 15, 157–66.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
212
May 26, 2011
5:11
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
National Center for Women and Policing (2004). Suspect typology; profiling the sex offender. http://www.hawaii.edu/hivandaids/Suspect%20Typology%20%20%20Profiling%the% 20Sex%20Offen der.pdf. O’Donohue, W. and Letourneau, E. and Dowling, H. (1997) The measurement of sexual fantasy. Sexual Abuse: A Journal of Research and Treatment, 9, 167–78. Polaschek, D.L.L., Hudson, T. and Ward, S.M. (1997) Rape and rapists: Theory and treatment. Clinical Psychology Review, 17(2), 117–44. Prentky, R.A. and Knight, R.A. (1993) Age of onset of sexual assault: Criminal and life history correlates. In G.C.N. Hall, R. Hirschman, J.R. Graham and M.S. Zaragoza (eds), Sexual Aggression: Issues in Etiology, Assessment, and Treatment (pp. 43–62). Washington, DC: Taylor & Francis. Pardue, A. and Arrigo, B.A. (2008) Power, anger, and sadistic rapists: Towards a differentiated model of offender personality. International Journal of Offender Therapy and Comparative Criminology, 52(4): 378–400. Purcell, C.E. and Arrigo, B.A. (2006) The Psychology of Lust Murder: Paraphilia, Sexual Killing, and Serial Murder. San Diego, CA: Academic Press. Quinsey, V.L., Harris, G.T., Rice, M.E. and Cormier, C.A. (1998) Violent Offenders: Appraising and Managing Risk. Washington, DC: American Psychological Association. Ratner, P.A., Johnson, J.L., Shoveller, J.A., Chan, K. et al. (2003) Non-consensual sex experienced by men who have sex with men: Prevalence and association with mental health. Patient and Education Counseling, 49, 67–74. Richards, H.J., Washburn, J.J., Craig, R. et al. (2004) Typing rape offenders from their offense narratives. Individual Differences Research, 2(2), 97–108. Salter, A.C. (2003) Predators, Pedophiles, Rapists, and Other Sex Offenders: Who They Are, How They Operate, and How We Can Protect Ourselves and Our Children. New York: Basic Books. Sbraga, T.P. (2003) Sexual deviance and forensic psychology: A primer. In W. O’Donohue and E. Levensky (eds), Handbook of Forensic Psychology: Resource for Mental Health and Legal Professionals, London: Elsevier Academic Press. Scott, G.G. (1983) Dominant Women, Submissive Men. New York: Praeger. Serin, R.C., Malcolm, P.B., Khanna, A. and Barbaree, H.E. (1994) Psychopathy and deviant sexual arousal in incarcerated sexual offenders. Journal of Interpersonal Violence, 9(1), 3–11. Shipley, S.L. (2007) Perpetrators and victims: Maternal filicide and mental illness. In R. Muraskin (ed.), Prentice Hall’s Women in Criminal Justice Series (4th edn): It’s a Crime, Women and Justice, Upper Saddle River, NJ: Prentice Hall. Shipley, S.L. and Arrigo, B.A. (2001) The confusion over psychopathy (II): Implications for forensic (correctional) practice. International Journal of Offender Therapy and Comparative Criminology, 45(4), 407–20. Shipley, S.L. and Arrigo, B.A. (2004) The Female Homicide Offender: Serial Murder and the Case of Aileen Wuornos. Upper Saddle River, New Jersey. Shipley, S.L. and Arrigo, B.A. (2008) Serial killers and serial rapists: dichotomy or continuum- an examination of commonalities and comparison of typologies. In R. Kocsis (ed.), Serial Murder and the Psychology of Violence, Totowa, NJ: Humana Press, Inc. Stermac, L., Del Bove, G. and Addison, M. (2004) Stranger and acquaintance sexual assault of adult males. Journal of Interpersonal Violence, 19(8), 901–15. Strickland, S.M. (2008) Female sex offenders: Exploring issues of personality, trauma, and cognitive distortions. Journal of Interpersonal Violence, 23(4), 474–89. Struckman-Johnson, C. and Struckman-Johnson, D. (2006) A comparison of sexual coercion experiences reported by men and women in prison. Journal of Interpersonal Violence, 21(12), 1591–1615. Tjaden, P. and Thoennes, N. (1998, November) Prevalence, incidence, and consequences of violence against women: Findings from the National Violence against Women Survey. National Institute of Justice Centers for Disease Control and Prevention Research in Brief, 1–16. US Department of Justice (2008) Criminal Victimization in the United States, 2006 Statistical Table. Retrieved September 12, 2008, from http://www.ojp.usdoj.gov/bjs/pub/pdf/ cvus06.pdf.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
SEXUAL OFFENSES AGAINST ADULTS
213
Ward, T. and Beech, A. (2006) An integrated theory of sexual offending. Aggression and Violent Behavior, 11, 44–63. Warren, J.I., Reboussin, R., Hazelwood, R.R. and Wright, J.A. (1991) Prediction of rapist type and violence from verbal, physical, and sexual scales. Journal of Interpersonal Violence, 6(1), 55–67. Whitehead, P.R., Ward, T. and Collie, R.M. (2007) Time for a change: Applying the Good Lives model of offender rehabilitation to a high-risk violent offender. International Journal of Offender Therapy and Comparative Criminology, 51(5), 578–98. Willis, D.G. (2009). Male-on-male rape of an adult man: A case review and implications for interventions. Journal of the American Psychiatric Nurses Association, 14(6), 454–61. Zgoba, K.M. and Levenson, J. (2008) Variations in the recidivism of treated and nontreated sexual offenders in New Jersey: An examination of three time frames. Victims and Offenders, 3, 10–30.
P1: OTA/XYZ P2: ABC JWST073-09 JWST073-McMurran
May 26, 2011
5:11
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
PART IV
SPECIFIC POPULATIONS
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
Chapter 10
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS PHIL RICH Stetson School, Massachusetts, USA
INTRODUCTION AND LITERATURE REVIEW Case formulation involves the capacity to reduce and synthesize information to a more precise form that expresses broad ideas concisely. It represents an analysis and abstraction of key features of the case, and reduces history, facts, symptoms, and circumstances into a brief summary that identifies key treatment targets, provides meaning, conjectures causes, outlines current issues, and informs prognosis. To some degree, the formulation puts forth a clinical theory about the case, providing explanatory information, identifying hypothesized causes, and suggesting meaning and future action. It thus allows the basis for understanding the case, both historically and dynamically, and its likely trajectory, as well as interventions that may interrupt trajectory and create a new prognosis.
Formulation Is Not Formulaic Formulation provides an individualized approach to recognizing the idiosyncratic particulars of each case and understanding each case as unique, and does not equal a formulaic approach to or interpretation of cases. To this end, Drake and Ward (2003) write that formulation-based approaches to treatment require that evaluators understand psychological problems and vulnerabilities for individual clients, rather than utilizing a manualized or cookbook approach to understanding behavior. Despite the perspective that offenders share common dysfunctions and can thus be treated through a prescribed and common approach, Drake and Ward argue that this position is limited and results in weak and poorly targeted treatment that fails to meet individual needs. Instead, they note, case Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
218
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
formulation highlights specific developmental factors relevant to the case of each individual which converged to make that specific individual vulnerable to engaging in antisocial behavior. In case formulation, then, an understanding of the case is based on the individual, in which treatment interventions are not only individualized but emerge from the process of case formulation. Whereas gathering information and condensing it may follow a standard model, the process of understanding the individuals with whom we work is individually based. The data gathering process is therefore nomothetic, or based on general and universal principles, whereas the focus on the individual is idiographic, targeting only the case at hand and structured by and focused on the individual being assessed. Case formulation, then, can be thought of as idiographic practice framed within a nomothetic process (Houston, 1998). The process of case formulation helps clinicians understand the factors and motivations that shape and explain behavior, as well as prognosticate the further development or resolution of pathology. A forensic focus recognizes and seeks the presence of risk factors that contributed or led to the antisocial behavior, and perhaps the development of a troubled and/or antisocial or criminal personality structure in the individual. In particular, the forensic approach looks for the historical or current presence of criminogenic factors, or elements that produce, drive, or predict criminal behaviors. The related concept of criminogenic needs points to the presence of elements that need to be addressed if crime is to be reduced (Andrews, Bonta and Hoge, 1990). Offenders have many needs, but only some are functionally related to criminal behavior. Criminogenic needs are dynamic or changeable risk factors that are the intermediate targets in interventions that aim ultimately to reduce reoffending.
How, Why, and Where to from Here? In its forensic application case formulation addresses the intersection of legal and mental health processes and incorporates a full understanding of both. It includes the criminogenic factors that influence, shape, and drive antisocial and illegal behaviors and the mental health issues that affect, drive, or result from criminal behavior. A central question in forensic work, then, is not simply what happened, but how and why it happened, and what to do about it. We can thus think of forensic work in two areas: Evaluation and treatment. Simply put, as forensic evaluators our role is to evaluate the juvenile’s behavior based on the presence and influence historical and current psychosocial and psychological factors. In the case of forensic risk evaluation, we extend this to predictions, projections, or the likelihood of future antisocial or criminal behavior in juveniles who have already engaged in such behaviors. Most simply, Grisso (2006) has written that such evaluations are always forensic because they are performed in order to inform court decisions, although this is not always the case for children and adolescents. The contexts in which juvenile forensic evaluations occur extend beyond the court, as not every child or adolescent engaging in criminal behavior is legally prosecuted, even though the behavior itself involves the possibility of juvenile or adult criminal charges.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
219
In the case of forensic treatment, we can simply reframe this as the process of applying mental health and psychosocial treatment to children and adolescents who have engaged in and are at risk for re-engaging in significant antisocial and/or criminal behaviors. Such treatment includes both standard treatment interventions and specific interventions aimed at the criminogenic and other dynamic factors that specifically contribute to juvenile criminal behavior. In many respects, unlike nonforensic mental health treatment, forensic treatment for juveniles assumes, or should assume, a largely rehabilitative approach in which the goal is to build positive and prosocial attitudes, beliefs, expectations, ideas, social interactions, critical social skills, and mental representations of self and others, as well as change behavior. In the case of both evaluation and treatment, formulation is important if we are to understand the youth beyond the mere presence of historical factors and behaviors, and projections about future antisocial behavior. In either case, formulation provides the basis for understanding the juvenile and the myriad factors that have come together in his or her particular life to produce the historical behavior, the current level of psychological and psychosocial functioning, and the probability of future occurrences of problematic behavior, in which, as Kagan (2006) has noted, there is no single cause, rather, “A coherence of several factors is necessary to produce a particular phenomenon” (pp. 94–5).
Forensic Work with Juveniles: A Comprehensive, Developmental, and Contextual Framework Coffey (2006), Greenberg and Shuman (1999), Grisso (1998, 2006), and McCann (1998) describe the difference between forensic evaluation and treatment, and consider the two roles to be inherently different, incapable of being filled by the same practitioner. Nevertheless, the idea that forensic evaluation and treatment are separate processes that require separate practitioners assumes not only that the differences are irreconcilable but also that forensic work with children and adolescents is the equivalent of forensic work with adults. Acknowledging the differences between evaluation, or serving and informing the legal system, and serving and treating the client, Bluglass (1990) writes that it is the task of the forensic practitioner to reconcile these differences in order to be effective in both arenas. Indeed, in the treatment of juvenile offenders both roles must be consolidated, unless we are to separate the functions of evaluation and treatment completely, from the initial assessment of risk through all subsequent risk evaluations. Accordingly, the work of both the forensic evaluator and the forensic clinician must be synthesized into an overarching model of forensic mental health if we are to ensure an assessment and treatment model that is able to evaluate and treat children and adolescents in a single and seamless model capable of responding to youths holistically, as individuals, and as persons-in-development. Not only must we blend evaluation and treatment into a single framework, in a forensic model that evaluates and treats children and adolescents we must remain acutely aware of the developmental process and the developmental status of
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
220
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
juveniles and bring these ideas to bear in both our understanding of youthful offenders and at every stage of our treatment. In advocating for this developmental framework in our work with juvenile offenders, Steinberg and Scott (2003) describe adolescents and children as developmentally immature when compared to adults, not only in general brain maturation and psychological development, but also with particular respect to their decision making capacity, increased vulnerability to social circumstances, and still forming character and personality. Similarly, Zimring (2004) admonishes us to take into account the developmental status of juvenile sexual offenders, with respect to the moral significance of their current behaviors, predictions of future behavior, and implications for treatment. Grisso (1998), too, pays great attention to the rapidly moving development of juveniles, writing that “questions of growth and development are at the heart of all juvenile forensic evaluations” (p. viii) and suggests that it may be more relevant to refer to forensic evaluations for juveniles as “forensic developmental evaluations”. He writes that “nothing about the behavior of adolescents can be understood without considering it in the context of youths’ continued biological, psychological, and social development” (p. 27). From a developmental and contextual framework, Grisso (1998) also describes the necessity of paying attention to the social environment in forensic work, recognizing the juvenile in the context of the systems within which he or she lives, interacts, and functions, and that this systemic perspective is central in the translation of evaluation data into recommendations for responses that are in the best interests of the youth and society. Indeed, the social environment is the great mediator in many of the developmental tasks faced by the developing child and the context in which personal learning take place. It is within this developmental and learning environment that the vulnerabilities described by Marshall and Eccles (1993) develop and grow into risk factors, or in which children develop the assets and strengths that serve as protective factors. In this environment, children find, or fail to find, the elements necessary to function successfully, effectively, happily, and prosocially in the world; however, just as their presence provides the greatest opportunities for children, the absence of these ideals presents an environment in which a child will have great difficulty succeeding. In fact, the developmental pathway of each individual cannot be separated from the contextual social environment into which it is woven. An ecological perspective from this point of view relates to the interconnection between and the mutual influence of each part of the environment. Described by Elliot, Williams and Hamburg (1998), the ecological-developmental approach provides a framework by which human development is understood through the interactive social contexts which influence and shape behavior. Human development and human behavior – including the antisocial and criminal behavior of juvenile offenders – occurs within a complex and multiply nested, multiply interacting, and mutually transactional environment. Our ability to understand human development and behavior thus requires an understanding of the individual affected by all levels of the ecological system. To a great degree, this becomes part of a guiding model in both the assessment and treatment of juvenile offenders, in which we understand risk in children and adolescents in the contexts of both their social and psychological development and their social learning environment, or the developmental/learning environment in which they have been raised.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
221
Forensic Risk When we engage in forensic work and adopt a forensic approach we must think about risk, for at least two reasons and from at least two perspectives. In risk assessment, we consider risk from the perspective of recidivism, or the potential, likelihood, or probability of future recurrence. In this case, regardless of how we evaluate risk for recidivism, we are concerned with trajectory if things remain unchanged; however, understanding risk not only helps us to understand the direction of future behaviors but also the nature of risk, the conditions that give rise to risk factors, and the conversion, in some children and adolescents, of risk factors to actual harmful behavior. When we discuss risk in relationship to juvenile criminal behavior, we are focused upon risk factors believed to be related to re-engagement in criminal behavior, as well our understanding of what caused or contributed to the behavior in the first place. Nevertheless, despite focusing on risk factors specific to juvenile delinquency and adolescent criminal behavior, a broad range of psychosocial variables is associated with risk for conduct disordered behavior (Rutter, 1994), and even a cursory review of risk factors makes it clear that forensic risk (i.e., risk for criminal behavior) is linked to risks of many types for children and adolescents. Hence, many of the essential risk factors for juvenile crime are identical to those for other personal or social problems, as is true for the presence of protective factors, or the personal and social elements that buffer against or neutralize risk. This again reflects an ecological approach to understanding the development of juvenile offending, highlighting the fact that such behavior emerges from the same social conditions that give rise to other troubling conditions, and it is the presence, convergence, and multiplicative effects of individual and sets of risk factors, at the biological, psychological, social, and environmental levels, that finally come together to produce harmful and criminal behaviors in any given individual.
The Strength of Static Risk Factors: The Early Development of Risk Static risk factors are those elements of risk that are historical, and are not subject to change. Early adverse developmental experiences and prior criminal behaviors, although in quite different categories are both examples of static factors that serve as potential indicators of future risk. Static risk factors offer a great deal to risk prediction because history counts, not just as a predictor but also as a foundation upon which present ideation and behavior is built and as the basis for continued trajectory. Just as resiliency against adversity and stress is laid down in early positive development (Sroufe et al., 2005), so too is the foundation for antisocial behavior, as well as many other kinds of risk, laid in early life. This is easily seen by the number of general risk factors that make their appearance in childhood and before age 12 years. Thus, Henry et al. (1996) note that children who become serious criminal offenders are characterized by features that consistently bring them into conflict with their surroundings during early childhood. Individual characteristics that make individuals susceptible to risk are perhaps at their greatest during early childhood
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
222
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
as they come into contact and interplay with environmental conditions that may catalyze risk or build a backdrop out of which antisocial behavior may later emerge (Rutter, Giller and Hagell, 1998). Moreover, rather than unexpectedly emerging at a later point in adolescent or adult development, Loeber et al. (2005) write that risk factors for serious crime can be observed in earlier childhood, and later antisocial behavior is often based on processes that accumulate over many years, associated with multiple risk factors in multiple domains. It is thus important that we recognize the difference between what Moffitt (2003) calls life-course persistent antisocial behavior, in which persistent antisocial behavior clearly originates in childhood when the troubled behaviors of the high-risk young child are acted out in and amplified by a high-risk social environment, and adolescence-limited antisocial behaviors that are more transient and first appear during adolescence.
The Operation of Multiple Risk Domains During Childhood and Adolescence In focusing on risk factors for serious delinquency or violence, Lipsey and Derzon (1998) and Hawkins et al. (2000) identified risk residing within and spread through five essential domains: individual, family, school, peer group, and community. Figure 10.1 shows examples of risk factors found within each domain; however, of the 23 childhood risk factors identified in the US Surgeon General’s report on
Risk factor by domain Individual
• Aggressiveness • Early violent behavior • Antisocial attitudes and behavior
Family
• • • • • •
Parent criminality Child maltreatment, poor family bonding Family conflict Low parental involvement Antisocial parent attitudes Parent-child separation
School
• • • • •
Academic failure Poor school bonding Frequent school changes School truancy School dropout
Peer group
• Delinquent siblings • Delinquent peers
Community
• • • •
Poverty Community disorganization Crime ridden neighborhoods Exposure to violence and prejudice
Figure 10.1 Examples of risk factors for youth violence found in each of the five domains identified by Lipsey and Derzon (1998) and Hawkins et al. (2000).
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
223
youth violence largely based on the work of Lipsey and Derzon (US Department of Health and Human Services, 2001), no single risk factor alone is especially powerful; however, this does not mean that these risk factors are inconsequential. Instead, it suggests that it is a combination of risk factors, and no single factor alone, that best predicts risk. That is, although some risk factors are stronger in effect than others, and are therefore better predictors of later antisocial behavior, no single factor is itself necessary for nor sufficient to predict or produce antisocial behavior. In addition, not only is risk best predicted by multiple risk factors, but the likelihood that an individual will engage in antisocial behaviors is greatly increased by the number of risk factors to which the individual is exposed (Farrington, 1997; Garmezy, 1987; Hawkins et al., 2000). Further, risk that produces antisocial behavior is driven not just by multiple risk factors but, as noted, interactions among risk factors across multiple domains (Haggerty and Sherrod, 1996; Loeber et al., 2005). Further, many of the childhood risk factors take on a greater role during adolescence whereas others drop in significance, reflecting the idea that the same risk factors operate differently at different points during childhood and adolescent development and have more or less impact based on the age and developmental level of the child. For instance, Lipsey and Derzon (1998) found that the most significant early risk factors in childhood development (ages 6–11 years) included low family socioeconomic status and antisocial parents. These were followed by factors that, while less statistically significant, were nonetheless meaningful, including poor parent–child relationships, harsh or lax parental discipline, broken homes, poor attitude toward or performance in school, and exposure to television violence, as well as weak social ties. Of note here, the influence of risk factors associated with social connection increases in strength during early adolescence (ages 12–14 years), such as weak social ties which increases considerably in effect size, and by adolescence becomes the strongest indicator in Lipsey and Derzon’s inventory of risk factors, as well as an increase in the risk power of poor child–parental relationships. During adolescence, other factors not especially prominent during childhood take on more significance; in the case of serious and violent criminal behavior, these especially include association with antisocial peers, which is a factor that may be less relevant when we consider risk for sexual reoffense and especially among those who sexually abuse children. Nevertheless, important seeds of risk are often sown early in childhood development and take greater root in adolescence, along with other risk factors that may have held less importance at an earlier point in development. As attachment patterns are considered to develop early in childhood, setting the pace for later social connections (Rich, 2006), it is not surprising that the effect size of weak social ties as a predictor of risk in Lipsey and Derzon’s model increases by 160% during early adolescence.
Secure Attachments are Protective A discussion of attachment and its secure and insecure variants is well beyond the scope of this chapter; however, whereas insecure attachment, as the by-product of suboptimal or disrupted early experience, may be considered a general risk factor
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
224
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
for many difficulties, secure attachment almost certainly serves as a protective factor. In a nutshell, the attachment model proposes that early and on-going attachment experiences provide a foundation upon which identity is built and internalized. These experiences either contribute to secure attachment and resiliency, serving as a protective factor against negative or hostile life circumstances, or as a developmental vulnerability in which insecure attachment not only fails to protect but under adverse conditions, and particularly when combined with other risk factors, is quite possibly a risk factor in its own right. On the other hand, secure attachment strengthens the individual, buffering against adverse social conditions and helps neutralize and weakens both internal and external risk factors that may otherwise increase risk (Rich, 2006, 2007). Atkinson and Goldberg (2004) describe secure attachment serving a protective function because, throughout life, under adverse or anxiety provoking circumstances, it triggers distress-regulating and support seeking behaviors. The failure to internalize attachment security is assumed by attachment theorists to set in motion pathways that lead to behavioral, relational, and other functional difficulties. Fonagy (2001, 2004), for example, asserts that the capacity to adequately “mentalize,” or visualize, one’s own mental state and the mental states of others, develops directly out of the attachment experience and the child’s opportunity to observe and explore the mind of the caregiver, and that severe deprivation undermines the acquisition of this skill (i.e., metacognition). Fonagy (1999a, 1999b) describes the skills of metacognition as critical to effective social functioning, and proposes that crimes are committed by people with inadequate metacognitive capacities who instead engage in pathological attempts in order to adapt to a social environment in which metacognition is essential. Fonagy’s perspective, then, is that adolescent antisocial behavior results from a lack of both security and metacognitive skills, linked through the attachment experience. Travis Hirschi (2002), far from an attachment theorist, also considers risk for criminal behavior to be tied to attachment and social connection, writing that “the bond of affection for [prosocial] persons is a major deterrent to crime” (p. 82). He notes that we are moral beings to the extent that we have internalized the norms of society, and “that the essence of internalization of norms, conscience, or superego lies in the attachment of individuals to others” (p. 18). Simply speaking, then, secure and prosocial attachments serve as a protective rather than a risk factor. Secure attachment is implied in the capacity for close and connected relationships to parents, other adults, and adults, and further in the capacity to form bonded relationships to community organizations like schools, accept and connect with prosocial norms and values, demonstrate self-regulation, and experience self-efficacy. Secure and insecure attachment styles are thus key to the case study presented in this chapter.
Attachment, Social Connection, and Moral Behavior The Commission on Children at Risk (2003) describes a crisis in American childhood, based upon a lack of social connection, both to other people and to constructs
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
225
such as morality and personal meaning. Their perspective is that the task is to reconnect children, not just to peers and adults in the community, but to society itself. Directly linking social attachment to moral development, and hence social behavior, the Commission describes moral behavior stemming from attached relationships as much as the acquisition of rules. They take the position that we are genetically predisposed to connect with others and form attachments, and that our sense of right and wrong originates largely from a biologically primed need for attachment. From this perspective, moral behavior stems as much from relationships as from the acquisition of standards and rules; thus, the failure to form secure attachments necessarily means the failure to form a strong moral code. The idea that moral development and behavior is linked to social connection was postulated by Piaget (1997), who believed that moral development is based on values directly related to equality and reciprocity in social relationships. Hoffman (2000) also describes empathic connections with others as the basis for moral development, as does Vetlesen (1994) who asserts that perceptions of morality are built on the experience of empathy for others. Kagan (1984) too considers the child’s acquisition of standards to be facilitated by the recognition of feelings and thoughts in self and others, mediated through the development of empathy, and he makes moral development contingent upon the development of empathy. Similarly, Stilwell et al. (1998) describe the “moralization of attachment”, or the transformation of early attachment and social experiences into the values, attitudes, and beliefs that underlie relationships and behaviors, resulting in a moral conscience. They write that the development of empathy and morality occurs in the context of early attachment relationships and later expanding relationships with other family members, adults, and friends, and within the social organizations and institutions in which children are raised and grow to adulthood. Stilwell defines moral delay, arrest, and deviancy as developmental disruptions, interruptions, and derailments that result from disruptions in attachment, neglectful parenting, or trauma (Stilwell et al., 1994). Through their social environments, then, children and adolescents become attached to the norms and values of their societies, and incorporate these into their identities and character. In the social context, we find the role models who influence the development of behavior and personal identity, and the arena in which social lessons are taught and learned. Thus, the larger social context in which child rearing and child development occurs is not just an important, but passive, backdrop to the development of antisocial behavior, but an active ingredient in terms of attachment, social connection, and the development of empathy and moral behavior.
ASSESSMENT The Application of Ideas in Case Formulation More than an exercise in psychosocial assessment, case formulation helps us to recognize the influence of past and present factors that influence and shape antisocial thinking, attitudes, interactions, and drive antisocial behavior, and central issues or problems to be addressed through the treatment process. Equally, case
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
226
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
formulation provides an opportunity to think, not only about risk factors and their amelioration, but the presence and action of protective factors as well, and their central role in developing and delivering treatment interventions, not simply to contain behavioral and social deficits but to build on and develop strengths and engage in the process of rehabilitation.
Application of Structured Instruments in Case Formulation Case formulation is key to the use of the widely used Risk-Need-Responsivity model (Andrews, Bonta and Hoge, 1990; Andrews, Bonta and Wormith, 2006; Hoge and Andrews, 1996, 2003), and in the application of the model through the Level of Service/Case Management Inventory and its counterpart for adolescent offenders, the Youth Level of Service/Case Management Inventory (YLS/CMI). Simply, the RNR model provides the evaluator with a structure and guide by which to individualize the intensity, level, location, and type of case management and treatment services provided to both adult and juvenile offenders, based upon their assessed level of Risk for reoffense, an assessment of their individualized treatment Needs based on the presence of specific criminogenic factors, and an assessment of the likely Responsiveness of each individual to the process and format of case management and treatment services. Of importance, the RNR model is strongly based on case formulation, and final decisions are based on professional “override,” or the case formulated judgment of the evaluator. The RNR model not only embraces a case formulation approach, but case management and treatment programs that adhere to its model, which includes differentiated treatment based on individual case formulation, are more effective in reducing recidivism than programs that do not follow the model that are more standardized and “cookie cutter” in their approach (Andrews and Bonta, 2010; Hanson et al., 2009). Case formulation is also central, although perhaps to a lesser degree, in the use of the Structured Assessment for Violence Risk in Youth (SAVRY), which, like the YLS/CMI, is a structured instrument used to assist in the clinical evaluation of youths with respect to risk. Unlike the YLS/CMI, the SAVRY is only designed to assess risk and is not intended as a case management instrument. On the other hand, the SAVRY is a 30-item instrument that includes an evaluation of six protective factors, and thus recognizes the mediating influence of protection in predictions of recidivism, permitting the assessment of risk to be made “as a professional judgment, including consideration of . . . risk and protective factors” (Borum, Bartel and Forth, 2002, p. 17). The protective items mirror exactly those already described in this chapter: Prosocial involvement, strong social support, strong attachments and social bonds, commitment to school, and resilient personality, as well as positive attitudes toward treatment and authority. These instruments are forensic assessment tools that combine clinical judgment with empirically based structure. They both structure the assessment of risk by providing a list of risk factors to be evaluated for both presence and strength and, as noted, the SAVRY pays some attention to the presence of protective factors as well. Both instruments require a detailed understanding of the young person being assessed, but whereas the SAVRY helps formulate an assessment of risk the LSI
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
227
goes beyond risk in order to further formulate treatment, case management, and supervision needs for the youth.
CASE STUDY Mitchell is a 15-year-old male adolescent, short in height and slight in stature, looking significantly younger than his age, and of low average IQ. Raised by two unmarried drug addicted parents, Mitchell’s mother was 18 and his father 23. He and his younger brother were significantly physically, and possibly sexually, abused by their father, who also severely physically abused Mitchell’s mother. After being severely physically assaulted, Mitchell’s mother moved into the home of her own former foster mother, taking Mitchell, 61/2 , and his four-year-old brother with her. Soon after the move, his mother began to leave the home for weeks at a time, returning to periodically live with Mitchell’s father, re-engaging in active significant drug abuse, and was arrested when Mitchell was 8 and has not returned to his life since that time. His father was later arrested for serious drug and physical assault charges and is incarcerated today, and has had no contact with Mitchell since age 6. By age 5, Mitchell had begun to engage in sexual behavior with other children, steadily and progressively continuing until age 8, by which time he was behaviorally out of control in almost all environments. By age 8, he was intermittently violent toward peers in school and had been transferred to several different schools, assaulted an unrelated male adult on the street, was stealing from peers and stores, and in the home of his foster mother he was stealing money and other things, rummaging through garbage cans for food, stealing and hoarding food, and engaging in self-injurious behavior and violence against his foster mother. After aggressive and suicidal behaviors at home, at age 8 Mitchell was psychiatrically hospitalized, followed by temporary shelter care. He refused to return to his foster home, ran away without shoes and socks, and eventually remained in shelter care for two years before being placed in a long-term residential treatment facility from ages 10–13 where he continued to engage in noncompliant and oppositional behavior, assaults against staff, and ongoing sexual behavior with peers. During this time, he engaged in periodic suicidal behaviors, mostly involving suicidal ideation, mild self-injurious behavior, and suicidal gestures, resulting in three additional psychiatric hospitalizations. He was placed in a third residential program between ages 13–15, after aging out of the prior program, where the same behaviors occurred with more frequency. Despite repeatedly not doing well in residential care, at age 15 Mitchell was placed in a community-based group home where he remained for only three months. During that period, he was suspended from school three times, engaged in several fist fights at school, threatened teachers on three occasions, engaged in shoplifting, and broke into a car. Mitchell was placed back into residential care, and after sexually molesting a female staff member by touching her breasts and stealing property from staff and peers, all within three months of his admission, criminal assault charges were pressed against Mitchell. After running away from the program, Mitchell broke into a home and two cars, and was arrested while shoplifting. At 151/2 , he was placed into a staff secure residential correctional facility, by which time he had
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
228
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
been placed outside of his family since age 6, and, prior to that, had been seriously neglected and abused by his parents. He has been in residential treatment since age 81/2 , for seven of his 15 years. He has an IQ of 92, and although age 15 stands only 5 feet and 1 inch tall, physically appears to be about age 11 or 12, and has not yet reached puberty. He would like to take hormone shots for his height. In his new residential program, Mitchell almost immediately re-engaged in prior behaviors. By seven weeks postadmission, he had been physically restrained numerous times, had repeatedly reported suicidal ideation and engaged in selfinjurious behavior and, most significantly, following a minor incident, committed a serious and unprovoked assault on a female staff, telling her if he was going to get into trouble for the minor infraction it may as well be “worth it”. He repeatedly punched her in the face, head, and torso, leaving her physically harmed and feeling extremely vulnerable. Prior this, almost every day Mitchell had engaged in very troubling behavior and required much one-one staff attention. Three days prior to the staff assault, after cutting himself with a broken plastic cup and tying a shoelace around his wrist so tightly that his hand turned blue, Mitchell, while making and putting up Christmas decorations in his clinician’s office, casually discussed these self-injurious behavior with his therapist. He said he was angry at the time, but did not know how to tell staff he was upset and angry. He then clarified that he actually did not really know he was angry at the time of the self-injurious behavior.
Case Formulation Instrument-Based Formulation A cursory application of both the YLSI/CMI and the SAVRY in Mitchell’s case reveals both the presence of multiple static and dynamic risk factors and the absence of protective factors. Both instruments suggest a high level of risk and a wide range of criminogenic needs – again, personal needs and desires that propel the individual toward antisocial behavior in order to get these needs met. The YLSI also suggests the requirement for a strong level of supervision and management and a low level of responsiveness to treatment. A more detailed description of Mitchell’s case would provide fine detail, but even in a brief case one can see in his developmental history many of the elements that flag risk, point to the development of psychological vulnerabilities, and reflect a lack of age appropriate developmental protection. Not only are many risk factors obvious in Mitchell’s history, but the absence of protective factors is just as clear. Hence, we easily see a history of abuse and neglect with Mitchell as victim, exposure to significant domestic violence, parental absence and abandonment, parental substance abuse, instability in and multiple living conditions, pediatric/early onset depression, a lack of stable and adequate adult attachment figures and role models, a lack of cohesive social connection, and poor or arrested moral development. In addition, we recognize the early onset of Mitchell’s own violent behavior as a risk factor, as well as a more global early onset conduct disorder. In his current life, given Mitchell’s lack of ties to any individual or peer group, we also see the action of weak social connection at work, recalling its role in Lipsey and Derzon’s
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
229
(1998) model in which weak ties during early adolescence is flagged as one of the strongest indicators of risk. Together, these risk factors allow us to recognize the impoverished nature of Mitchell’s relationships, the developmentally primitive and reactive source of his behaviors, his lack of connection to social norms and values, and his experience of alienation from those around him, and thus helps us to recognize the elements that come together to produce, in Mitchell’s case, antisocial behavior and the ability to act against others in a harmful manner, with a limited sense of self-agency.
Developmentally Informed Formulation Let us go further, however, than the organized and structured process that the use of risk assessment instruments can provide, and take a more detailed look at Mitchell through the lens of developmental theory, and particularly from the perspectives of attachment, social connection, and the moralization of social connection and empathy. Mitchell is alexithymic in that he has no words to describe his feelings to staff, or even recognize his feelings as he is experiencing them. Not surprisingly, as the two concepts are related and are probably derived from the same source of early experience and social development, Mitchell shows a lack of metacognitive skill (or mentalization) with respect to self-reflection. He is unaware he is even having feelings, and certainly cannot identify or name them, so it is no wonder that he is unable to express these to staff. Instead, he responds to his feelings by going directly to a learned, nonreflective, and acting out mode of behavior in order to resolve a dysphoric state of mind, draining the emotionality out of the dysphoria or diverting it, dissipating the mood, gaining staff attention and proximity, and additionally getting to engage in an angry and hostile punishing relationship with adult staff, against whom he directs his feelings. The angry feelings that Mitchell was experiencing, which led to his self-injurious behavior just days earlier, escalated to an unprovoked and serious assault physical and emotional on a member of the staff. The discussion between Mitchell and his therapist while making and putting up decorations in her office is important. In this formulation, Mitchell was capable of having that discussion, and at that time, because he was in a safe, relaxed, and socially related state, connected to both his clinician at that moment and society, while engaged in holiday festivities that indirectly Mitchell was sharing with his larger society, and feeling neither stressed nor anxious. He was thus able to demonstrate some mild self-reflection at that time, as well as a relatively honest discussion with his therapist, who was neither pressing Mitchell into working hard as a patient nor seeking anything specific from him, other than engagement. She, too, was making decorations and them up Mitchell, engaging with him in an unspoken show of connection. Overall, and especially when stressed, Mitchell demonstrates poor metacognitive skills, a weak sense of self-agency, little sense of self-efficacy, and virtually no capacity for self-regulation, all skills and capacities believed to derive from early attachment experiences, and certainly on-going experience in the social developmental/learning environment. With a battered, drug addicted, and neglectful
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
230
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
mother who was unable to protect Mitchell from abuse, and who grew up in foster care herself, it was not likely that Mitchell would develop the skills of metacognition or self-regulation through his relationship with his mother. Further, she was completely out of his life by age seven, having effectively abandoned him to foster care, and to her own prior foster mother at that. We can surmise that his mother had few self-regulatory skills, given her drug addiction and her incapacity on unwillingness to remain away from a violent boyfriend, not to mention relinquishing motherhood of her two young children. By 9, even Mitchell’s foster mother was out of his life, with no contact with her or his brother since that time. Add to this Mitchell’s size and lack of physical maturation, for which he seeks hormonal remediation. We can reasonably hypothesize that this influences his self-esteem and sense of being unlike his peers, which he more than makes up for by his seriously conduct disordered behaviors. These gain him everyone’s respect through their fear of his unpredictable behavior, including an increasing capacity for violence. With respect to the development of social learning, Mitchell has not been responsible for, nor had the opportunity to demonstrate, truly independent behavior since age eight, largely growing up in the highly contained and artificial environment of institutional care. Mitchell fails to feel a significant level of empathy for others, given his own early experiences of maltreatment, and a pattern of insecure avoidant attachment that for many years has resulted in Mitchell maintaining his distance from others while actively pushing them away. In this regard, Mitchell has become the busy painter described by Anthony (1987), who writes that development is an interactive process in which children are not merely passive receptacles for experiences that simply write upon and shape them. He describes the child, from the start, as not a bucket waiting to be filled with experience but a searchlight exploring the horizon, actively creating and shaping his or her own world. That is, children and adolescents make decisions about the behaviors in which they choose to engage, even if their decision making process is different than that of adults, and different again based on their age and cognitive development or capacity. When high-risk juveniles are drawn to and actively seek out environments, peers, or conditions that allow or promote the expression of antisocial behavior, they themselves directly influence and increase the number or risk factors present in their lives, and thus the chances that antisocial behavior will actually occur. Building upon Erikson’s (1959, 1968) model of early development and the accomplishment of developmental tasks, Mitchell’s early experience of the world as unresponsive, unreliable, untrustworthy, and quite possibly a dangerous place, has contributed significantly in Mitchell’s experience of the world fitting that very schema. Mitchell’s active detachment from and avoidance of social connection, perhaps stemming back to his earliest experiences of victimization, helplessness, and abandonment, has been instrumental in creating the very circumstances in which he is now unable to connect with others and others feel little connection with him or desire to connect. In this world, Mitchell has not been able to recognize and hence experience the empathy that others may have experienced for him, and this has further limited his capacity to experience empathy for others or empathic distress at the results of his own victimization of others, and limited the development of moral judgment and behavior.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
231
Finally, Mitchell’s experience is one of alienation, social disconnection, and isolation. His sense of connection to a larger social environment is very much limited to his earlier chaotic and maltreated life, the world of social services and residential treatment where peers and staff come and go, and a world in which his only means to get his needs met are by force of character. With respect to those needs, given his lack of metacognition and reflectivity, Mitchell neither understands nor takes ownership for his needs. Moreover, Mitchell is showing signs of developing the callous and unemotional traits that are linked to psychopathy.
CASE FORMULATION AND TREATMENT PLANNING Summary and Process of the Formulation The formulation of Mitchell’s case and his prognosis, as well as his needs for treatment and case management, can be structured by the use of instruments such as the LSI, but a more informed and meaningful understanding and interpretation is driven by clinical insight and not instrumentation alone. This formulation and Mitchell’s behavior are both best understood by taking into account: (a) multiple risk factors and the absence of protective factors, and (b) the developmental pathways and ongoing social environment and contexts in which human development occurs and from which human behavior, beliefs, attitudes, social skills, social competence, and connections result. Tying these ideas to those introduced early in the chapter, forensic case formulation is about both prognosis and cause, and not simply risk alone, which otherwise tells us very little. Thus, in addition to risk assessment, we need case formulation to guide, shape, and define treatment and case management interventions. In the case of still developing children and adolescents, the application of a developmental approach to formulation and treatment is critical. In Mitchell’s case, in terms of assessment instrumentation, the formulation is very much influenced by a model that recognizes significant static and dynamic risk factors, and hence a high level of risk and needs when utilizing principles of the risk, need, and responsivity model. In terms of the second “r”, responsivity, we also see that Mitchell is not likely to easily or freely engage in or respond to treatment at the outpatient level, and so this formulation points to clearly identifying specific risk factors as targets for treatment through the treatment planning process (below), and providing treatment for Mitchell in a restrictive and closely managed inpatient setting, such as secure residential treatment. When also informed by an attachment-oriented perspective, formulation points to the need for the development of critical social skills, including those of social perspective taking, self-reflection and metacognition, and distress tolerance, described in more detail in the treatment planning section that follows. In terms of formulation, however, through an attachment-informed lens, Mitchell is a significantly undersocialized and highly detached individual, lacking social skills at both the intra- and interpersonal level, and hence in strong need of treatment aimed at the development of such necessary social skills.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
232
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Treatment Planning The beauty of an RNR-influenced case formulation approach to treatment is that it allows a view of the individual first built upon his or her risk for continuing the same behaviors, and then in terms of his or her treatment needs and likely responsiveness to treatment interventions. An expansion upon both needs and responsivity leads to the fine details of the treatment plan, as well as the details, intensity, and location and form of the treatment setting or environment. In Mitchell’s case, any risk assessment instrument is likely to indicate significant risk for continued difficulty, or recidivism. Both static/history and current/dynamic risk factors clearly indicate a high likelihood of continued troubled and harmful behaviors. An assessment of needs, especially from a developmental perspective, shows the need for the development of, not simply social connections, but the development and acquisition of social skills that will allow Mitchell to experience meaningful and personally satisfying social experiences and relationships, and provide a means for meeting personal goals prosocially. This includes the development of self-regulatory skills, the ability to understand his own needs and the needs of others, and to address the “syndrome of social disability” described by Barbaree, Marshall, and McCormick (1998), who describe the contributions of early abusive and adverse family experiences to the development of deficits in critical social skills and social competence in children and adolescents, as well as the development of antisocial behavior. Both the RNR and the attachment-oriented formulation identified risk factors and clear deficits in social skills, including lack of self-regulatory skills, metacognition, and social relatedness, each of which, in Mitchell’s case, increase the risk for reoffense and general problematic behavior. In terms of specific targets for treatment, key of course to any treatment plan, treatment goals will include the development of perspective taking, problem recognition, problem-solving and decision-making skills, the recognition of internal mental states and emotional experience, the recognition of mental states and emotional experiences of others, and the skills of teamwork and collaboration, both with peers and staff. Finally, Mitchell historically has been unresponsive to treatment interventions, and persistent staff supervision and management. Accordingly, as the RNRinfluenced formulation highlights the unlikelihood of Mitchell’s working on and developing the necessary treatment tasks, or even being contained, in an outpatient treatment setting, the treatment plan for Mitchell will highlight the need for a treatment environment that is able to provide a high level of behavioral supervision and management, and is thus most likely to call for treatment in a structured residential treatment setting.
CONCLUSION When we step from forensic evaluation to the treatment of forensic cases, we step into forensic mental health, or the application of mental health ideas and techniques to the treatment and rehabilitation of criminal behaviors and other
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
233
forms of behavior that are forensic in nature (Cordess, 2001; Rich, 2003; Welldon, 1997). When we extend these ideas to forensic work with children and adolescent, we recognize the treatment of juvenile offenders as a forensic subspecialty that crosses the lines between understanding criminal behavior, assisting the process of legal discrimination and decision making regarding the behavior, assessing the behavior for future occurrence (risk assessment), and treating the behavior. In working with juveniles, such work requires an additional understanding of the developmental and personal psychology of children and adolescents and surrounding social systems and social forces that shape and define the emotions, cognitions, and behaviors of the child. Beyond this, both to avoid formulaic treatment and to implement the ideas that drive models like risk, need, and responsivity, we must take on an approach that utilizes case formulation, and in so doing develop a model for the comprehensive assessment and treatment of juvenile offenders. Such a model will provide the basis for a deeper understanding of prior behavior, current attitudes and behaviors, and the likelihood of future behavior if things remain unchanged, as well as an individualized understanding of the treatment needs and targets of each youth with whom we work.
REFERENCES Andrews, D.A. and Bonta, J. (2010) Rehabilitating criminal justice policy and practice. Psychology, Public Policy, and Law, 16, 39–55. Andrews, D.A., Bonta, J. and Hoge, R.D. (1990) Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, 19–52. Andrews, D.A., Bonta, J. and Wormith, S.J. (2006) The recent past and near future of risk and/or need assessment. Crime and Delinquency, 52, 7–27. Anthony, E.J. (1987) Risk, vulnerability, and resilience: An overview. In E.J. Anthony and B.J. Cohler (eds), The Invulnerable Child (pp. 3–48). New York: Guilford Press. Atkinson, L. and Goldberg, S. (2004) Attachment and psychopathology: The integration of developmental and clinical traditions. In L. Atkinson and S. Goldberg (eds), Attachment Perspectives on Psychopathology and Intervention (pp. 3–26). New Jersey: Erlbaum. Barbaree, H.E., Marshall, W.L. and McCormick, J. (1998) The development of and sexually deviant behaviour among adolescents and its implications for prevention and treatment. Understanding, Assessing, and Treating Juvenile and Adult Sex Offenders. A Special Issue of the Irish Journal of Psychology, 19, 1–31. Belsky, J. and Nezworski, T. (1988) Clinical Implications of Attachment. New Jersey: Erlbaum. Bluglass, R. (1990) The scope of forensic psychiatry. Journal of Forensic Psychiatry, 1, 5–9. Borum, R., Bartel, P. and Forth, A. (2002) Manual for the Structured Assessment of Violence in Youth (SAVRY). Department of Mental Health Law and Policy, University of South Florida. Bronfenbrenner, U. (1979) The Ecology of Human Development: Experiments in Human Behavior. Harvard University Press, Massachusetts. Coffey, P. (2006) Forensic issues in evaluating juvenile sexual offenders. In D.S. Prescott (ed.), Risk Assessment of Youth Who Have Sexually Abused (pp. 75–86). Oklahoma: Wood & Barnes,. Commission on Children at Risk (2003) Hardwired To Connect. Institute for American Values, New York.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
234
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Cordess, C. (2001) Forensic psychotherapy. In C.R. Hollin (ed.), Handbook of Offender Assessment and Treatment (pp. 309–29). Chichester: John Wiley & Sons, Ltd. Drake, C.R. and Ward, T. (2003) Practical and theoretical roles for the formulation based treatment of sexual offenders. International Journal of Forensic Psychology, 1, 71–84. Elliot, D.S., Williams, K.R. and Hamburg, B. (1998) An integrated approach to violence prevention. In D.S. Elliot, B.A. Hamburg and K.R. Williams (eds), Violence in American Schools: A New Perspective (pp. 379–86). Cambridge: Cambridge University Press. Erikson, E.H. (1959) Identity and the Life Cycle. New York: Norton. Erikson, E.H. (1968) Identity: Youth and Crisis. New York: Norton. Farrington, D.P (1997) Early prediction of violent and nonviolent youthful offending. European Journal on Criminal Policy and Research, 5(2), 51–6. Fonagy, P. (1999a) Male perpetrators of violence against women: An attachment theory perspective. Journal of Applied Psychoanalytic Studies, 1, 7–27. Fonagy, P. (1999b) Psychoanalytic theory from the viewpoint of attachment theory and research. In J. Cassidy and P.R. Shaver (eds), Handbook of Attachment: Theory, Research, and Clinical Application (pp. 595–624). New York: Guilford Press. Fonagy, P. (2001) Attachment Theory and Psychoanalysis. New York: Other Press. Fonagy, P. (2004) The developmental roots of violence in the failure of mentalization. In F. Pfafflin and G. Adshead (eds), A Matter of Security: The Application of Attachment Theory to Forensic Psychiatry and Psychotherapy (pp. 13–56). London: Jessica Kingsley. Garmezy, N. (1987) Stress, competence, and development: Continuities in the study of schizophrenic adults, children vulnerable to psychopathology, and the search for stressresistant children. American Journal of Orthopsychiatry, 57, 159–74. Greenberg, S.A. and Shuman, D.W. (1999) Irreconcilable conflict between therapeutic and forensic roles. In: D.N. Bersoff (ed.), Ethical Conflicts in Psychology (2nd edn, pp. 513–20). Washington, DC: American Psychological Association. Grisso, T. (1998) Forensic Evaluation of Juveniles. Florida: Professional Resource Press. Grisso, T. (2000) Ethical issues in evaluations for sex offender re-offending. Invited address presented at Sinclair Seminars, 6 March 2000, Madison, WI. Grisso, T. (2006) Foreword. In S.N. Sparta and G.P. Koocher (eds), Forensic Mental Health Assessment of Children and Adolescents (pp. vii–x). New York: Oxford University Press. Haggerty, R.J. and Sherrod, L.R. (1996) Preface. In R.J. Haggerty, L.R. Sherrod, N. Garmezy and M. Rutter (eds), Stress, Risk, and Resilience in Children and Adolescents: Processes, Mechanisms, and Interventions (pp. xiii–xxi). Cambridge: Cambridge University Press. Hanson, R.K., Bourgon, G., Helmus, L. and Hodgson, S. (2009) The principles of effective correctional treatment also apply to sexual offenders: A meta-analysis. Criminal Justice and Behavior, 36, 865–91. Hawkins, J.D., Herrenkohl, T.I., Farrington, D.P. et al. (2000, April) OJJDP Juvenile Justice Bulletin: Predictors of Youth Violence. Washington, DC: US Department of Justice, Office of Justice Programs. Henry, B., Caspi, A., Moffitt, T.E. and Silva, P.A. (1996) Temperamental and familial predictors of violent and nonviolent criminal convictions: Age 3 to age 18. Developmental Psychology, 32, 614–23. Hirschi, T. (2002) Causes of Delinquency. New Jersey: Transaction Publishers. Hoffman, M.L. (2000) Empathy and Moral Development: Implications for Caring and Justice. Cambridge: Cambridge University Press. Hoge, R D. and Andrews, D.A. (1996) Assessing the Youthful Offender: Issues and Techniques. New York: Plenum Press. Hoge, R D. and Andrews, D.A. (2003) Youth Level of Service/Case Management Inventory: User’s Manual. New York: Multi-Health Systems. Houston, J. (1998) Making Sense with Offenders: Personal Constructs, Therapy and Change. Chichester: John Wiley & Sons, Ltd. Howell, D. (1995, May) Guide for Implementing the Comprehensive Strategy for Serious, Violent, and Chronic Juvenile Offenders. Washington, DC: US Department of Justice Office for Juvenile Justice and Delinquency Prevention, Kagan, J. (1984) The Nature of the Child. New York: Basic Books.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
FORENSIC CASE FORMULATION WITH CHILDREN AND ADOLESCENTS
235
Kagan, J. (2006) An Argument for Mind. Connecticut: Yale University Press. Lipsey, M.W. and Derzon, J.H. (1998) Predictors of violent and serious delinquency in adolescence and early adulthood: A synthesis of longitudinal research. In: R. Loeber and D.P. Farrington (eds), Serious and Violent Juvenile Offenders: Risk Factors and Successful Interventions (pp. 86–105). California: Sage. Loeber, R., Pardini, D., Homish, D.L. et al. (2005) The prediction of violence and homicide in young men. Journal of Counseling and Clinical Psychology, 73, 1074–88. McCann, J.T. (1998) Malingering and Deception in Adolescents: Assessing Credibility in Clinical and Forensic Settings. Washington, DC: American Psychological Association. Marshall, W.L. and Eccles, A. (1993) Pavlovian conditioning processes in adolescent sex offenders. In H.E. Barbaree, W.L. Marshall and S.M. Hudson (eds), The Juvenile Sex Offender (pp. 118–42). Guilford Press, New York, Moffitt, T.E. (2003) Life-course-persistent and adolescence-limited antisocial behavior: A 10-year research review and a research agenda. In B.B. Lahey, T.E. Moffitt and A. Caspi (eds), Causes of Conduct Disorder and Juvenile Delinquency (pp. 49–75). Guilford, New York. Piaget, J. (1997) The Moral Judgment of the Child. New York: Simon & Schuster. Rich, P. (2003) Understanding Juvenile Sexual Offenders: Assessment, Treatment, and Rehabilitation. John Wiley & Sons, Inc., New Jersey. Rich, P. (2006) Attachment and Sexual Offending: Understanding and Applying Attachment Theory to the Treatment of Juvenile Sexual Offenders. Chichester: John Wiley & Sons, Ltd. Rich, P. (2007) The implications of attachment theory in the treatment of sexually abusive youth. In M.C. Calder (ed.), Working with Young Children and People who Sexually Abuse: Taking the Field Forward (pp. 201–16). Lyme Regis: Russell House. Rutter, M. (1994) Family discord and conduct disorder: Cause, consequence, or correlate? Journal of Family Psychology, 8, 170–186. Rutter, M., Giller, H. and Hagell, A. (1998). Antisocial Behavior in Young People. Cambridge: Cambridge University Press. Sroufe, L.A., Egeland, B., Carlson, E.A. and Collins, W.A. (2005) The Development of the Person: The Minnesota Study of Risk and Adaptation from Birth to Adulthood. New York: Guilford Press. Steinberg, L. and Scott, E.S. (2003) Less guilty by reason of adolescence. American Psychologist, 58, 1009–18. Stilwell, B.M, Galvin, M.R., Kopta, S.M. and Norton, J.A. (1994) Moral-emotional responsiveness, a two factor domain of conscience functioning. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 130–9. Stilwell, B.M., Galvin, M.R., Kopta, S.M. and Padgett, R.J. (1998), Moral volition: the fifth and final domain leading to an integrated theory of conscience understanding. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 202–10. US Department of Health and Human Services (2001) Youth Violence: A Report of the Surgeon General. Maryland: US Department of Health and Human Services. Vetlesen, A.J. (1994) Perception, Empathy, and Judgment: An Inquiry into the Preconditions of Moral Performance. Pennsylvania: Pennsylvania University Press. Welldon, E.V. (1997) Forensic psychotherapy: The practical approach. In E.V. Welldon and Van Velsen, C. (eds), A Practical Guide to Forensic Psychotherapy (pp. 13–19). London: Jessica Kingsley Zimring, F.E. (2004) An American Travesty: Legal Responses to Adolescent Sexual Offending. Illinois: University of Chicago Press.
P1: OTA/XYZ P2: ABC JWST073-10 JWST073-McMurran
May 26, 2011
5:25
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
Chapter 11
FORMULATING OFFENDING BEHAVIOR WITH PEOPLE WITH MILD LEARNING DISABILITIES WILLIAM R. LINDSAY University of Abertay, Dundee, UK
The purpose of a formulation is to consider the various factors in the person’s life and in the incident or incidents which have led to referral so that we may understand the reasons why the person functions in the way they do and why the incidents occurred. Once we have identified or formulated why these incidents occurred and why the person is the kind of individual they are, this will allow us to address these factors through a treatment program. All of this work is bound up in research and theory on why people behave in the way they do. Theory, supported by research findings, draws our attention to factors that are likely to influence behavior and, for this chapter, specifically offending behavior. We can look to the research literature to identify variables that may have contributed to causing this individual to act in the way they have. There are a number of empirical findings and theories related to offending behavior that account for both distal/historical factors and proximal/immediate factors, in addition to the modeling and stimulus–response models that might count for immediate actions. I will review these in brief, pointing out the way in which they affect an individual clinician’s judgement on the formulation of an individual case.
BIOLOGICAL RESEARCH AND SOCIOLOGICAL THEORIES It may seem odd to group biological findings and sociological theories under one heading, but they are certainly not antithetical. A number of theories, based primarily on genetics, have promoted research to determine the extent to which biological mechanisms of inheritance affect the likelihood of criminal behavior. Many of these studies have looked at familial patterns with antisocial behavior in children or criminal behavior in adulthood associated with a relatively high Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
238
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
frequency of similar problems in parents. The risk of antisocial and criminal behavior in boys whose fathers have received one prison sentence has been found to be around five times greater than those boys whose fathers had not been registered with the police (Kandel et al., 1988). Twin studies have also found greater concordance for criminal behavior for identical pairs than fraternal pairs (Christiansen, 1997). Mednick et al. (1984, 1988) conducted studies on adopted twins on the register of 14 427 Danish adoptees. The main results were that, if neither the biological nor adoptive parents were criminal, then 13.5% of their sons were criminal. If the biological parents were not criminal and the adoptive parents were criminal, the figure was only marginally greater at 14.7%. If the biological parents were criminal and the adoptive parents were not criminal, the figure rose to 20%. Finally, if both sets of parents were criminal then the figure was 24.5%. The probability of a conviction for the boy rose with the number of convictions for the biological parents from 0 to 3 times or more. In this way, we know that criminality in parents is a risk factor for an individual being convicted of a crime (Quinsey et al., 2006) and this should be considered in a formulation. One of the obvious drawbacks is that this research is wholly related to male offenders and would not have direct implications for women. Around the 1950s, sociological theories began to emerge and these in turn began to direct the nature of research into criminal groups. Cohen (1955) suggested that boys entered into delinquency because they were conforming to the expectations of their delinquent subculture. Hirschi (1969) developed these notions with control theory, paying attention to both positive learning of criminal behaviors through association with criminal peers and also to the development of self-control through appropriate social learning, including being law-abiding. He felt that the success of social training was dependent on four factors: (1) attachment to the expectations and values of society; (2) commitment to society through an understanding of the loss that will be experienced should the individual be arrested and convicted for crime; (3) involvement in and engagement with the ordinary activities of the community, such as work and education; and, finally, (4) a belief in the accepted laws of society as a reasonable framework for community cohesion. A large number of empirical tests of Gottfredson and Hirschi’s general theory (1990) have shown low self-control to be one of the most consistent predictors of crime (DeLisi, 2005). Self-control and self- regulation have been shown to be germane to the development of sociable law abiding people while deficits in these skills are related to persistent criminality. People with low self-control are impulsive, insensitive, have poor emotional regulation, and are risk takers who tend to perform poorly and fail to meet the responsibilities toward school, work, family and, ultimately, society (Gottfredson and Hirschi, 1990).This has been shown to hold across gender, race and ethnicity, In addition people with poor self-regulation commit greater amounts of deviant and maladaptive behaviors people with higher levels of self-regulation. Recent validation by DeLisi and Vaughn (2008) found that poor self-control was by far the strongest predictor of criminal careers when compared to demographic variables such as age and socioeconomic status. Because of this work, in a formulation the forensic assessor would evaluate the extent to which the individual identified and conformed with society. Issues such as prosocial influences, isolation and occupation become important considerations.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
239
THEORIES OF SEX OFFENDING The following case illustrates the value of formulation for a sex offender with Intellectual Disabilities (ID). Lindsay (2009) has outlined a model for the treatment of sex offenders with ID that draws on theoretical variables from mainstream sex offenders and also postulates a number of variables specific to men with ID. One of the most critical developments in theory for mainstream sex offenders was that of Marshall and Barbaree (1990), who focussed on the importance of developmental experiences in men as a motivation and explanation for their wish to offend sexually. They viewed adolescence as a critical period in the development of personal understanding of aggressive and sexual impulses, the development of social and interpersonal skills, the development of problem-solving coping styles, and the development of self-regulation skills in relation to both aggression and sexual impulses. In this way, as with Hirschi’s control theory for general offending, they placed self-regulation as a central concept in the perpetration of incidents. A history of developmental adversity, such as physical or sexual abuse or oppressive parenting, may result in teenage males developing poor skills to cope with major biological and social changes. As a result, young men, who may have already been vulnerable, do not develop adequate coping or appropriate skills for developing social and sexual relationships in adulthood. If an individual has developed inadequate relationship skills, poor coping styles, low self-esteem and attachment difficulties, they are more likely to have significant problems in developing appropriate self-regulation skills in relation to sexuality and interpersonal relationships. This in turn primes certain men for a breakdown in social and sexual relationships and predisposes them to inappropriate and even violent sexuality. In terms of a formulation for individuals, this theory leads us to the consideration of a number of factors. Personal abuse in childhood is clearly important, as are developmental experiences through adolescence that may prevent the individual from developing appropriate interpersonal skills and effective self-regulation of emotion and impulses. If these various factors have not been adaptive, then the theory allows for the development of stable cognitive self-schemas, which are the beliefs the individual holds about him- or herself. Maladaptive schemas will be strengthened by chronic deficits in emotional regulation and poor relationship skills. Individuals may become increasingly isolated at a time when social identity is developing and such isolation encourages the further development of idiosyncratic world views and schemas in the absence of prosocial influences. Sexual coping is another central concept. Marshall and Barbaree suggest that men cope with emotional and interpersonal difficulties through appropriate sexual and interpersonal contact which, if not available, will result in inappropriate sexual strategies or sexual offending. The actual occurrence of an offense is likely to be further promoted by disinhibitors, such as alcohol or stress. This theory resulted in new treatment methods incorporating the promotion of self-esteem, the promotion of appropriate social and sexual strategies, a focus on victim empathy, and an analysis of cognitive distortions that might support an individual’s sexual offending. Following a formulation based on these factors, such treatments may be indicated as appropriate.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
240
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
In his precondition model of child sexual abuse, Finkelhor (1984) considered four preconditions relating to sexual abuse, all of which could be considered in a formulation. Men had to have the motivation driven by sexual arousal and sexual preference for children and this could be caused by the experience of personal sexual abuse or an emotional congruence with children. The perpetrator is also likely to be aware of the social and cultural taboos placed on adult/child sexuality and must overcome these internal inhibitions. He suggested that this might be done by the employment of cognitive distortions supporting a sexual offense or disinhibition by substance abuse. External inhibitors also have to be overcome and so sex offenders must organize opportunities and situations that might allow the commission of an offense. Finally, once these other three preconditions are met, the offender must now overcome the resistance presented by the child. Finkelhor suggested that offenders might employ a range of different methods including the pretence of playing with the child, use of pornography, use of threats or violence and emotional blackmail. Once again, the theory suggests a number of variables to be considered in a formulation, some of which are similar to those suggested by Marshall and Barbaree (1990). Hall and Hirschman (1991, 1992) proposed the quadripartite model that employed some similar factors to those mentioned previously. There were four primary factors accounting for the diversity of types of sexual offending. The first was sexual arousal to an inappropriate stimulus and this provided physiological motivation for sexual offending. The second was that men employed a range of cognitive distortions condoning or justifying the perpetration of sexual abuse. The third primary factor represented an acknowledgment that sex offenders will experience a degree of anxiety or apprehension over the prospect of committing an illegal act. The normal inhibitors, which exert some behavioral and emotional regulation over the individual, must be weakened in order that the offense can take place. The use of alcohol, or emotional dysregulation through anxiety or depression, might serve to weaken personal restraint. The final primary factor is related to personality factors, such as narcissism or antisociality. This includes consideration of maladaptive developmental experiences, such as poor attachment, physical or sexual abuse, or other adverse variables that might disrupt personality development and lead to antisocial means of responding to interpersonal situations. These factors are similar to those proposed by previous theorists, but importantly, they also postulated that each individual has a “threat threshold” that moderates and controls appropriate sexual behavior. If the motivation generated by the primary factors exceeds the threshold, then the likelihood of an offense becomes much greater. If the threat threshold is lowered in any particular situation, again the offense is more likely. For example, the threat threshold may be lowered by the perception (cognitive distortion) that the victim is complicit in a sexual act. Alternatively, if the offender experiences increased isolation and develops antagonistic views toward the world, together with a sense of entitlement through dysfunctional personality traits, then the strength of the primary factors may rise above the threat threshold. In these examples, the perpetration of a sexual offense becomes more likely. These various theories of mainstream sexual offending provide a context and a number of specific considerations when therapists are developing a formulation. There is clear research evidence supporting their importance in offenders with ID
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
241
in relation to attachment (Novaco and Taylor, 2008; Steptoe et al., 2006), deviant sexuality (Blanchard et al., 1999, 2008), sexual abuse in childhood (Lindsay et al., 2001), and cognitive distortions (Talbot and Langdon, 2006).
A THEORETICAL CONTEXT FOR TREATMENT IN SEX OFFENDERS WITH INTELLECTUAL DISABILITIES As suggested earlier, there is evidence supporting the relevance of various motivational factors that have emerged from theory on mainstream offending. A further Counterfeit Deviance Hypothesis (Hingsburger, Griffiths and Quinsey, 1991) was developed to account for inappropriate sexual behavior particularly in men with ID. These authors noted that people with ID often live in circumstances in which sexual development, the acquisition of sexual knowledge, and appropriate sexual relationships are not supported. Many individuals will have had little experience of learning about a range of issues related to sexuality and a number of factors that are related to developmental and environmental issues, rather than sexual deviance, will contribute to inappropriate sexual behavior. They stressed the importance of differentiating counterfeit deviance from true sexual deviance, writing that individuals should not be classified as deviant when the problem may lie in a system that has failed to provide appropriate opportunities and learning for sexual expression. Initial tests of the Counterfeit Deviance Hypothesis did not support the contention that poorer sexual knowledge and poorer knowledge of the laws of society were at the root of inappropriate sexual behavior (Michie et al., 2006; Talbot and Langdon, 2006). With regard to sexual knowledge, however, Lunsky et al. (2007) conducted a sensitive study, splitting sexual offenders into deviant/ persistent offenders and na¨ıve/inappropriate offenders. They found that only the deviant/persistent offenders had greater levels of sexual knowledge than controlled participants and they concluded that the Counterfeit Deviance Hypothesis may hold only with inappropriate offenders. They suggested that it may be more relevant to inappropriate offenders than deviant offenders. Therefore, this hypothesis may continue to be important in relation to a subsection of sexual offenders with ID. Further support for the Counterfeit Deviance Hypothesis came from the Ward and Hudson (1998) self-regulation pathways model. These authors proposed four pathways – two with approach goals and two with avoidant goals. The nature of the pathway within each goal (approach or avoidant) is determined by whether the self-regulation style is active or passive. The first is the approach/explicit pathway in which the individual has a clear wish to offend sexually and uses explicit plans and procedures to carry out the act. The approach/automatic pathway involves the individual engaging in over-learned behavioral scripts (passive regulation) that are consistent with sexual offending. The individual’s behavior may be poorly planned and somewhat impulsive within the context of the behavioral scripts. The third pathway is avoidant/active, where the individual attempts to control the thoughts and behavior that might lead to sexual offending; however, as has been
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
242
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
mentioned in previous theories, the strategies are ineffective and counterproductive leading to an increased risk of offending. Examples already mentioned are an increase in alcohol intake in an effort to control sexual impulses or masturbating to inappropriate sexual fantasies in an effort to “get it out of one’s system”. The fourth pathway is avoidant/passive, where the individual may wish to avoid sexual offending or abusive incidents, but either lacks coping skills to prevent this from happening or attempts to ignore the problem by hoping it will go away. Applications of this model to assessment and treatment have been developed and outlined extensively by Ward and colleagues (Ward et al., 2004; Ward, Yates and Long, 2006). In studies on sex offenders with ID, a number of the Ward and Hudson (1998) predictions emanating from their theoretical model were not supported. In particular, Hudson and Ward (2000) hypothesized that approach/explicit offenders would have a higher rate of reoffending, which has been borne out in research studies on mainstream offenders (Webster, 2005; Yates and Kingston, 2006); however, in a study on sex offenders with ID, Lindsay, Steptoe and Beech (2008a) found the opposite, with approach/explicit offenders having a significantly lower rate of reoffending following treatment. They felt that this supported a conclusion that these men may have an insufficient understanding of how inappropriate their behavior is. One function of treatment is to reinforce the extent to which sexual abuse is condemned by society. Once this realization has been achieved, stronger self-regulation strategies can be reinforced. Therefore, while these individuals are not completely na¨ıve about the fact that their behavior is inappropriate, they have not internalized the extent to which it is against the conventions of society. They had not previously understood the critical requirement for self-control in relation to sexual preference and the need to develop alternative relationship and coping skills. Further research has supported the finding that sex offenders have lower reoffending rates following treatment than mainstream offenders (Gray et al., 2007; Rice et al., 2008). This research may seem to undermine the relevance of the Ward and Hudson (1998) model for offenders with ID; however, Lindsay et al. (2008a) did find that the classifications differentiated sex offenders with ID in other ways. They reported that, after treatment, approach explicit offenders had significantly lower rates of reoffending than approach automatic offenders while the latter had significantly lower IQ. This latter finding was also reported by Langdon et al. (2007). Thus, there was empirical support for the categorization suggesting their usefulness in assessment and treatment. Lindsay (2009) has integrated these findings with the Counterfeit Deviance Hypothesis, suggesting that these sex offending pathways are indeed helpful in conceptualizing strategies used by sex offenders with ID. The concept of the threat threshold is important in relation to the Counterfeit Deviance Hypothesis. If the external threat is seen as low, then the primary factors such as sexual preference, developmental and emotional variables, coping styles, and personality features may be sufficiently activated for the individual to make an appraisal that a sex offense is worth the risk. Given the research already cited, the concept of the threat threshold seems particularly relevant to offenders with ID. If a man has not fully appreciated the extent to which inappropriate sexual behavior is condemned, then the threat threshold is likely to be lowered. In
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
243
circumstances where developmental disturbance is high and this is combined with a degree of sexual drive, then these factors may easily exceed lower threat thresholds, resulting in inappropriate sexual behavior. The severe criticism and sanctions meted out by carers and family may simply serve to mask an understanding of the wider views of society. Individuals may be quite used to excessive criticism from carers and families and sanctions for aberrant sexual behavior are given the same status as sanctions for other less important maladaptive interactions. In this way, the experiences of individuals with ID, along with their restricted developmental histories, contribute toward a view that inappropriate sexual behavior has the same magnitude of threat as other less serious inappropriate behaviors. This is an extremely important consideration for any formulation in that it may provide a crucial variable on perception of threat to be included in the formulation. These theories also lead to important treatment considerations in that a realistic appraisal of the extent to which incidents are condemned by society becomes a major focus for treatment. The way in which this can be used as a motivational force for treatment and an incentive to develop self-restraint in relation to sexuality has been detailed by Lindsay (2009). From these theoretical considerations, it is also clear that treatments for personal sexual abuse, emotional regulation, pathways into sexual offending, the cycle of offending, and self-regulation of sexual impulses will emerge from a formulation of an individual case.
CASE STUDY Framework for the Formulation The outline of a formulation contains setting conditions which include distal, predisposing factors that have happened in the person’s development during childhood and as a young adult. One then goes on to consider antecedent events that might interact with these predisposing factors and act as more immediate triggers for the offending incidents. For sex offenders, these are clearly outlined by Beech and Ward (2004) in their elucidation of the stages of risk assessment and they are also outlined in the four pathways to sex offending in the Ward and Hudson (1998) sex offender pathways model. A formulation will then include consideration of emotional channels associated with the incident; these are the emotions, the cognitions, the behaviors and the physiological sensations associated with the incident. Finally, the formulation considers that there is a feedback loop from the outcome to the precipitating factors that will affect (strengthen) the likelihood of the incident reoccurring. In the Ward and Hudson (1998) pathways model, the feedback loop from outcome and self-appraisal of the incident is crucial in both allocating offenders to pathways and in maintaining the pathway and behavior. For example, for active explicit offenders, the feedback loop is likely to be positive and self-gratifying, emphasizing the success of the perpetration and strengthening the possibility of recurrence. For the avoidant active offender, the negative evaluation is likely to lead to self-recrimination and a determination to increase the frequency and
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
244
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
intensity of the maladaptive strategies that have resulted in an inability to avoid the offending incident.
Case Study: Robert Background Information Robert is a 27-year-old man who has been diagnosed with mild intellectual disability and has committed a number of sexual offenses against pubescent children. When a cognitive assessment was conducted in 1995, he was found to have a full scale IQ of 74. This placed him in the range of borderline intelligence but since childhood intellectual assessment had placed him in the range of mild intellectual disabilities, he continued to be assigned this classification. He also fulfilled the service acceptance criterion of IQ<70, +/− 2 standard errors of measurement. He was brought up by his mother and father until his father died when he was eight years of age. Case notes from his childhood noted that the family were extremely disruptive and dysfunctional. His father was extremely violent toward all the family and it had been reported that he abused alcohol seriously, dying of alcohol-related causes. Aggression at school was reported frequently in Robert, as was chronic truanting. He had five full siblings and two half siblings, the latter of whom his mother had to another relationship after his father died. All of his full siblings were boys, he was the fourth in the sibship and the two half siblings were both girls. Slowness was noted in his development and at around the age of eight years and he was referred to child guidance services for severe behavioral difficulties both at home and at school. Since the death of his father, his mother had felt unable to control him and he was also known to the local police for fighting with other children and vandalizing property. He was aggressive within the family and fire setting was included as a reason for referral at the time. A year later he was taken into child and adolescent psychiatric services for assessment of hyperactivity. His mother was noted to be both overly concerned and disproportionately protective. When he was admitted he was fine until his mother left, whereupon he started kicking, screaming and biting staff. His behavior remained problematic throughout his stay and he was diagnosed as having hyperactivity disorder. As he moved into his teenage years, he began drinking alcohol and engaging in further disruptive and antisocial behavior both at school and in hospital settings. The case notes recorded that, at 16, considerable disruption and antisocial activity with aggression toward other pupils and teaching staff occurred. Although he was excluded from school on several occasions, he was never expelled and he left school at 16 years of age. At that time, his consultant psychiatrist felt that his prospects for independent living in the community were extremely poor. His behavior continued to be problematic and at 17 years he was arrested for a series of shoplifting incidents. He denied the incidents but was charged with two counts of theft. There is no record in the case notes of him being convicted of the charges. His problematic behavior continued and he was apprehended by the
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
245
police on several occasions in the following year. It was also noted that police had contacted the hospital to recommend treatment and supervision.
Sexual and Relationship History Following the death of his father, his mother raised the five boys on her own until she subsequently entered another relationship. The family were known to be chaotic and Robert’s stepfather was known to be physically abusive toward the children. There were allegations that Robert and one of his brothers were sexually abused by members of a neighbouring family, although this was never substantiated. Robert himself says that he was sexually abused by a member of staff and another patient during one of his hospital admissions at around 13 or 14 years of age. He said that the member of staff is now dead and the he remembers the man as extremely vindictive. At the age of 17, police apprehended him for talking to a 12-year-old girl and then going into her house uninvited. She accused him of kissing her and putting his hands on her body. He eventually admitted kissing the girl and that this was uninvited, but denied touching her in any other inappropriate way. During interviews in adulthood, he said that he had spoken to the girl on several occasions while walking past her house and on this occasion she had asked him in. At aged 19 there were further allegations of inappropriate sexual behavior involving gross indecency in relation to a boy of nine years. The case was dropped because of insufficient evidence. Robert was also said to be loitering outside a local primary school and local police demanded his mother supervise him at all times. When he was 21, his mother and stepfather evicted him from the house and he moved into his own flat. The tenancy was surprisingly successful and he managed to cope well with all of the requirements of daily and community living. His alcohol abuse seemed to have reduced, and he managed to cook for himself, do his own shopping and was visited occasionally by community nursing staff who remarked on how tidy the house was. A female patient moved in with him for a short time and then six months later he met and married a woman whom he had met at college classes. The marriage seemed to work fairly well and the couple had two children. After three years, however, Robert was charged and convicted of interfering sexually with two 11-year-old girls; there were three charges related to the offenses. He received a three-year probation sentence and was referred by the court for hospital treatment. His wife and children left him and there were a number of subsequent issues concerning access to his children. Eventually, access was attained under supervision and he continued with his probation reasonably successfully until it finished. Following the end of probation he did not want to continue sessions and stopped all contact with services. While on probation he had begun working as a cleaner in office premises for four hours during the evening. He was such a conscientious worker that he was taken on at another set of offices for a period in the morning. He then obtained another casual job selling newspapers. Thus, he had three separate jobs and was considered extremely conscientious and hard working at all three places of employment. Prior to his convictions for sexual offenses, he had been reducing his alcohol intake and alcohol was not involved in the offenses themselves. Early in his probation he
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
246
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
stopped drinking altogether and continued his abstinence throughout the whole period. At the age of 34, he was once again referred to services for inappropriate touching of an eight-year-old girl. He had been working in a general store and it subsequently became clear that he had become familiar with the children who came into the store. Specifically, he had managed to get three or four of the young girls who came into the shop to help him with stacking shelves and moving newspapers. The offense with which he was charged was in fact fairly innocuous. He had tapped a girl on the shoulder and asked her if she wanted to move some items. By that time, however, the girl’s family had been told by a friend that Robert had a history of sex offenses and they called the police out of concern for the patterns and routines that were developing. The police charged him with breach of the peace and he was convicted and given a further three year probation sentence. As part of the probation order, he was once again admitted into treatment. Toward the end of the probation period, he was seen by one of his neighbours taking photographs of children from his bedroom. The police were contacted and on confiscation of the camera, he was charged with breach of probation. This was the current charge and his probation was continued for a further three years. He continued to live in his own accommodation and always managed to take care of both his house and himself. In the last year he had begun drinking heavily again and could drink up to three bottles of vodka over the weekend. His drinking became increasingly problematic and he was referred to an alcohol treatment service. Unfortunately, they were unable to meet his needs because of his intellectual disabilities. He had continued in employment until he began drinking heavily. He held down two or three casual jobs for around 14 years and it is not clear why there was such deterioration in his drinking and subsequent work habits. Alcohol has not been involved in any of his offenses.
Risk Assessment When conducting an assessment, it is good practice to separate offenders who are at high risk of sexual offending but may be at low risk of violence from those who are high risk for both. In order to gauge, broadly, Robert’s level of risk for future sexual and violent offenses, two static/actuarial risk assessments were completed on him. The Violence Risk Appraisal Guide (VRAG: Quinsey et al., 2005) is a wellestablished actuarial assessment of risk for future violence. It has been standardized across a range of countries, cultures and settings and has three studies attesting to its validity for men with intellectual limitations (Gray et al, 2007; Lindsay et al., 2008b; Quinsey, Book and Skilling, 2004). Robert had a number of factors that contributed to his risk score on the VRAG, including the facts that he had behavioral problems at school, there were alcohol problems in his family, that he himself had alcohol difficulties both as a youth and as an adult, and his extensive criminal history that included nonviolent offenses. He did, however, have a number of factors that are neutral or reduced his risk score on the VRAG, including the fact that he lived in a relationship for more than two years, the fact that his mother and father were together until his father’s death, and his offending was against female
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
247
victims. Taken together, these placed him in, the range of medium risk for future violent offenses. The Static-99 (Hanson and Thornton, 1999) has also been validated with sex offenders with ID (Lindsay et al., 2008b). A number of factors contributed to his score, including his previous charges, his previous convictions, and the fact that his victims were strangers and unrelated to him. Taken together, these items placed him in the range of high risk for future sexual offending. Therefore, the risk assessment context for any functional assessment was that he was a high risk for future sexual offenses and a medium risk for future violent offenses.
Formulation Predisposing Factors The disruption in Robert’s family life and the chaotic nature of relationships reported at the time were certainly an important factor in the lack of development of self-regulation behaviors. His father’s emotional dysregulation with all the family and his persistent alcohol abuse probably acted as potent but antisocial modeling for Robert during his development, leading to a lack of emotional regulation skills. His persistent truanting also suggests that behavioral self-regulation was not instilled on a constant, consistent manner. The next important factor emerging from Robert’s history is the abuse that he suffered in his childhood. The reports attest to the fact that his father was violent with all of the children. Several authors have noted research that finds the level of personal childhood abuse in individuals with intellectual disabilities is likely to be somewhat higher than in nonlearning disabled populations (Beail and Warden, 1995; Sequeira and Hollins, 2003). Furthermore, researchers and have found that sex offenders report a higher rate of sexual abuse in their childhood than either non sexual offenders or non offenders (Dhawan and Marshall, 1996; Lindsay et al., 2001). Therefore, sexual abuse in childhood is seen as a risk factor for the commission of sexual offenses as an adult (Boer et al., 1997). That is not to say that all sex offenders have been abused in childhood. Most studies on the topic report that a majority of sex offenders say they have not been abused in their childhood; however, the fact that they report childhood sexual abuse at a higher rate indicates that it is likely to have association with sex offending in adulthood. Reports in Robert’s childhood indicate that there were investigations into abusive relationships between neighbours and Robert and one of his brothers. In addition, he was extremely vindictive toward one particular staff member in the establishment where he stayed as a young teenager. He was quite vehement saying that if this man were still alive, he would wish to harm him. Therefore, there are two separate pieces of information suggesting that Robert has been sexually abused in childhood and as a young teenager. The mechanism for including this in the formulation involves a developmental distortion of appropriate boundaries between adults and children. This distortion leads to, at worst, a complete misunderstanding of the cultural boundaries in sexuality between adults and children and, at best, an undermining of the cultural taboos that the individual knows about regarding adults and children. Therefore, Robert may have a less than
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
248
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
optimal grasp on the firm taboo regarding sexual relationships between adults and children. This results in the adoption of a range of attitudes and cognitions that may support or condone sexual interaction between adults and children. A further distal factor is the alcohol abuse reported for his father. Although there are no reports about severe alcohol abuse in his mother, as a child, Robert became desensitized to drunkenness in a close adult male. He learned through modeling that persistent heavy drinking was acceptable and that it was an adaptive way to deal with stressors in life. As an adolescent he went onto develop a regular alcohol habit to the extent that he would steal alcohol from shops. Even being apprehended by the police on several occasions did not act as a disinhibitor to stealing alcohol. Although it was reported that none of his offenses involved alcohol as an immediate precipitating factor, regular alcohol abuse would be likely to cause degeneration in his general psychological and physical well-being. This in turn would lower his emotional regulation skills which were already in an underdeveloped state. In this way, alcohol abuse acted as a distal rather than proximal factor in the formulation. It is clear that general behavioral disruption was a feature of his case from around six years to at least his late teenage years. This behavioral disruption is a well known factor in relation to future sexual offending. The Static-99 (Hanson and Thornton, 1999) includes violent offending as a factor in the risk assessment. The Sexual Violence Risk protocol (Boer et al., 1997) includes nonviolent, nonsexual offending as part of the risk assessment. The Sex Offender Risk Appraisal Guide (SORAG: Quinsey et al. (2006) is a well-established risk assessment for sexual offending and also includes factors related to general criminality in addition to sex offending items. Therefore, general behavioral disruption and especially teenage criminal behavior is related to future sexual offending. It is a reflection of behavioral dysregulation that might be manifest in a number of areas. There are records of theft offenses, aggression toward others and there is one reference to fire setting through Robert’s childhood and adolescent years. Therefore, this general behavioral dysregulation should be considered as a significant part of the formulation. These predisposing factors represent a considerable vulnerability in Robert to the commission of sexual offenses. He has factors that would promote emotional dysregulation that might result in a perception that it is not unusual to establish relationships with inappropriate individuals such as children. Furthermore, his behavioral dysregulation might lead to a propensity toward general criminality including sexual offending. The sexual abuse perpetrated upon him by adult men in his childhood and young teenage years would weaken knowledge of taboos concerning adult–child sexuality, while a propensity toward alcohol abuse might further lower his emotional regulation. In addition, it was noted that through his teenage years he was apprehended on several occasions by the police. He was charged on some of these occasions but there are no records of him being placed in the cells or in prison. This is important because he would learn that despite repeated contact with the criminal justice services, they did not follow through on their procedures. Therefore, his perception of threat from society and the police in particular would be lowered. Taken together, these predisposing factors present a series of significant setting events for the commission of an offense and in particular a sexual offense.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
249
Precipitating/Antecedent Factors From the case notes, two factors emerge fairly clearly. The first is that after the birth of their second child, Robert and his wife began to grow apart. This would result in a lower frequency of sexual contact. On interview, Robert himself did not link events with his wife and his sexual offending against girls; however, it is fairly commonly documented that a frequent cognitive distortion in sex offenders is that since they are not gaining sexual contact with their partner, they are entitled to sexual contact in other ways (Bumby, 1996; Ward and Keenan, 1999). This cognitive distortion results in coercive and inappropriate sexuality. It may have been that Robert felt some entitlement to sexuality resulting from less frequent contact with his wife. The second main precipitating factor emerging from the case note is that he began to make contact with 12-year-old girls and one girl in particular. He met her on a few occasions and says that he was invited by her into the house. This represents grooming in the context of sex offender pathways. By walking by her house at a certain time when he knew she was likely to be there, he made contact with the girl and began to gain her confidence. He would either suggest that he go into the house or she may have indeed asked him but, either way, he engineered the opportunity for intimate contact with this pubescent girl. When it was explored during therapy he said he could not remember whether he walked past her house on many occasions and managed to meet her on only a few, but it is likely that he passed on more occasions than he actually met the girl. The incident several years later when he was working in the shop was another example of such planning. At this time, he began enlisting the help of children while stacking shelves and doing other odd jobs. In this way he was gaining both trust and physical proximity to the girls. If he had continued with these behaviors, he might have engineered opportunities to take them into the storeroom, walk them home or develop some other more personal, private relationship. In the Ward and Hudson pathways model this constitutes an approach goal of intimate interaction with the victim. It may have been an automatic process in which his goals for contact were somewhat vague and diffuse; however his self-regulation strategy was certainly active in that he engaged with the victim on several occasions. On the other hand, he may have had explicit goals of sexual contact and each meeting constituted a step toward the implementation of this explicit strategy. Either way, his active self-regulation strategy reflects the predisposing factors of emotional dysregulation and disruption of the taboos surrounding sexual boundaries. One predisposing factor may have also been brought into focus as a precipitating factor. The fact that he had already been apprehended by the police on several occasions with little consequence would have lowered his perception of threat from the criminal justice system regarding any possible future or imminent incident. The threat threshold has been invoked as an important aspect in the perpetration of sexual crimes for mainstream offenders (Hall and Hirschman, 1992) and in particular for sexual offenders with intellectual limitations (Lindsay, 2009). Lindsay (2009) has noted that men with intellectual disabilities may have a poorer understanding of the extent to which inappropriate sexuality is condemned by society
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
250
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
and this in itself may contribute toward a lowering of the threat threshold. As has been mentioned, Robert’s previous experience of sexual abuse in his own childhood and his behavioral dysregulation may have further lowered his perception that society in the form of the criminal justice system was likely to follow through with sanctions for criminal behavior including sexual offenses.
The Four Channels of Response Involved in the Incident In relation to sexual incidents, the physiological arousal is likely to be twofold. First, there is likely to be a sexual arousal driving the man to engineer the situation whereby sexual contact is possible and second, the arousal will maintain the motivation to complete the act. In addition, excitement and anxiety might increase general arousal promoting feelings of well-being and fulfilment that are difficult to achieve in other, socialized ways. Offenders often describe the “highs” they get from the offense itself, be it theft offenses, violent offenses, sexual offenses or drug offenses, as some of the best feelings they have ever had. One aspect of treatment is to discuss the importance of avoiding these significant illegal “highs” and viewing them as problematic rather than goals. Without cognitions supporting or justifying the sexual offense, it is unlikely that the incident would occur. Common cognitions for sexual offenses involving children are that the children are acting in a sexualized manner, that they have started the sexual offense and that it does not do them any harm. Subsequent treatment interviews with Robert suggested that the latter two were certainly part of his cognitive repertoire. He was adamant that the victim had asked him into the house and by doing so had started the incident. This is a clear distortion in that even if she did ask him into the house, he had engineered the situation in the first place. In addition, if she had asked him into the house, she would not be envisaging sexual interaction. The sexual abuse that Robert experienced in his own childhood might have combined with his physiological arousal to generate thoughts that children would not be harmed by sexual contact. Although he was very resentful toward the member of staff whom he asserted had sexually abused him, he did not consider that he had suffered long-term damage. Generally, he thought that he got on with his life, had been married, could hold down a job and so on. Therefore, he would de-emphasize the extent of harm done to the victim. In this way, his cognitions were consistent with his physiological arousal. The final channel is behavioral and it is clear that his behavior was consistent with his emotion, cognition and arousal. He engineered the sexual situation over a number of days or weeks and once he achieved this goal, acted in a sexualized manner with the girls. Therefore, his arousal, cognitions, emotions and perception of the lowered threat threshold allowed him to commit sexual offenses with his victims.
Consequences There were two major consequences from these first sexual offenses. The first one was that he had positive appraisal of his behavior and the responses of his victims. This is consistent with the active, explicit offender who is likely to reinforce himself
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
251
on the success of his strategy, seeing it through to a successful conclusion. This in turn strengthens the likelihood of him repeating the act. The feedback loop to the precipitating factors would indicate that, in this case, the success of the previous incident acts as a precipitating factor for the perpetration of a further incident. The second outcome is that there appeared not to be any negative consequences of his behavior. The victims did not tell anyone about the first sexual offense and, as a result, he was not caught. Neither did the parents pursue him personally. This reinforces his perception of a low threat threshold and, indeed, may lower the perception threat even further. A lower threat threshold will act as a factor precipitating more frequent and more serious incidents. He may consider it safe to take more risk since the threat of being caught or reported is perceived as lower. Once again this increases the likelihood of further sexual offenses.
Treatment Implications The treatment implications will develop naturally from the formulation. All of the aspects highlighted in the formulation were dealt with during treatment. The first thing to focus upon is motivation to attend treatment. Previously, threat thresholds have been lowered and the successful outcome (the sexual offense) indicated by the formulation suggests that behaving in this manner has fulfilled an important function in Robert’s life. Given that, he has little motivation to disrupt the successful cycles he has developed for these incidents. Procedures for motivation focus firstly on the importance of the condemnation of society leading to the corresponding importance that he changes his routines and wish to offend. If he wishes to avoid future major disruption to his life with its corresponding episodes of depression and dismay through contact with the police and courts, then he must change. Treatment will then focus on the importance of regulating his lifestyle through adaptive routines, socialized integration with the community and appropriate interaction with adults. This might require some skills training but would certainly require exploration of attitudes and thoughts about interaction with work and other adults. Treatment should also focus on the importance of appropriate use of alcohol. Lindsay (2009) has demonstrated the way in which personal sexual abuse can be incorporated into treatment with an emphasis on coming to terms with previous abuse rather than using it as an excuse or mitigating factor for the abuse perpetrated by the individual. Issues of personal abuse can be explored in detail with support for coming to terms with its effects while, at the same time, noting that it cannot be used as a defense against culpability. Robert was also made aware of the importance of his routines. Previously he had maintained that he was only out walking or was only using his camera and that the sexual offenses had arisen almost by chance. Treatment allowed him to become much clearer that these were quite deliberate acts in a sequence leading deliberately to the incidents themselves. He himself did not invoke the breakup of his marriage as a source of sexual frustration; however, he did refer to the breakup as a source of emotional turmoil and stress that contributed to a wish to be alone and isolated resulting in his walks that eventually led to contact with girls. Again, the treatment process was one of highlighting this variable and making it quite explicit in the formulation.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
252
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Figure 11.1 A graphic, simplified version of the formulation used with the offender.
At several places in the formulation, cognitive distortions are mentioned. They lower personal disinhibition toward behaving in this way, support the commission of a sexual offense, maintain self esteem in the face of antisocial activities and justify the offenses through mitigation. A great deal of time was spent during treatment challenging these cognitions through a variety of exercises. All of these exercises can be seen in Lindsay (2009).
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
253
The variables were all incorporated into a risk plan. Robert himself constructed the relapse prevention plan that incorporated measures for dealing with all of these risks. Future life plans are based on relapse prevention principles and the Good Lives Model (GLM; Ward and Stewart, 2003). Relapse prevention addresses risks and builds future measures for avoiding or counteracting risks in the person’s life while GLM builds a future life on strengths that the individuals has shown in the past. The way in which these can be incorporated into treatment are described in Lindsay (2009).
CONCLUSIONS The way in which the formulation has been described is complex. It involves a number of predisposing factors, precipitating/proximal factors, four response channels (emotion, cognition, behavior and arousal) and a feedback loop into the precipitating factors. This complex elucidation of the cycle of offending would not be presented to the offender during treatment. Rather, a much simpler formulation is presented in Figure 11.1 which includes all of the factors mentioned but simplified into clear, accessible headings. A formulation such as this can be discussed during treatment and given to the offender so that they can work with it between sessions bringing back any new ideas and contributions. In this way, the general principles of formulation for cognitive behavior therapy can be applied to work on sex offenders and can be employed profitably in collaborative work with the men themselves.
REFERENCES Beail, N. and Warden, S. (1995) Sexual abuse of adults with learning disabilities. Journal of Intellectual Disability Research, 39, 382–7. Beech A.R. and Ward, T. (2004) The integration of etiology and risk in sexual offenders: A theoretical framework, Aggression and Violent Behaviour, 10, 31–63. Blanchard, R., Kella, N.J., Cantor, J.M. et al. (2008) IQ, handedness and paedophilia in adult male patients stratified by referral source. Sexual Abuse: A Journal of Research and Treatment, 19, 285–309. Blanchard, R., Watson, M., Choy, A. et al. (1999) Paedophiles: Mental retardation, mental age and sexual orientation. Archives of Sexual Behaviour, 28, 11–127. Boer, D.P., Hart, S.D., Kropp, P.R. and Webster, C.D. (1997) Manual for the Sexual Violence Risk – 20: Professional Guidelines for Assessing Risk of Sexual Violence. Vancouver, British Columbia: British Columbia Institute on Family Violence and Mental Health, Law and Policy Institute, Simon Fraser University. Bumby, K.M. (1996) Assessing the cognitive distortions of child molesters and rapists: Development and validation of the MOLEST and RAPE scales. Sexual Abuse: A Journal of Research and Treatment, 8, 37–54. Christiansen, K.O. (1977) A preliminary study of criminality among twins. In S.A. Mednick and K. Christiansen (eds), Biological Basis of Criminal Behaviour (pp. 177–92). New York: Gardiner Press. Cohen, A.K. (1955) Delinquent Boys: The Culture of the Gang. Glencoe, IL. Free Press. DeLisi, M. (2005) Career Criminals in Society. Thousand Oaks, CA: Sage.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
254
May 28, 2011
8:27
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
DeLisi, M. and Vaughn, M.G. (2008) The Gottfredson Hirschi critiques revisited: Theory, criminal careers, and career criminals reconciling self-control. International Journal of Offender Therapy and Comparative Criminology, 52, 520–37. Dhawan, S. and Marshall, W.L. (1996) Sexual abuse histories of sexual offenders. Sexual Abuse: Journal of Research and Treatment, 8, 7–15. Finkelhor, D. (1984) Child Sexual Abuse: New Theory and Research. New York: Free Press. Gottfredson, M.R. and Hirschi, T. (1990) A General Theory of Crime. Stanford, CA: Stanford University Press. Gray, N.S., Fitzgerald, S., Taylor, J. et al. (2007) Predicting future reconviction in offenders with intellectual disabilities: The predictive efficacy of VRAG, PCL-SV and the HCR-20. Psychological Assessment, 19, 474–9. Hall, G.C.N. and Hirschman, R. (1991) Towards a theory of sexual aggression: A quadripartite model. Journal of Consulting and Clinical Psychology, 59, 662–9. Hall, G.C.N. and Hirschman, R. (1992) Sexual aggression against children: A conceptual perspective of etiology. Criminal Justice and Behavior, 19, 8–23 Hanson, R.K. and Thornton, D. (1999) Static-99: Improving Actuarial Risk Assessments for Sex Offenders. (User report 1999-02). Ottawa: Department of the Solicitor General of Canada. Hingsburger, D., Griffiths, D. and Quinsey, V. (1991) Detecting counterfeit deviance: Differentiating sexual deviance from sexual inappropriateness. Habilitation Mental Health Care Newsletter, 10, 51–4. Hirschi, T. (1969) Causes of Delinquency. Barclay: University of California Press. Hudson, S.M. and Ward, T. (2000) Relapse prevention: Assessment and treatment implications. In D.R. Laws, S.M. Hudson and T. Ward (eds), Remaking Relapse Prevention with Sex Offenders: A Source Book (pp. 102–22). Thousand Oaks, AC: Sage. Kandel, E., Mednick, S.A., Kirkegaard-Sorensen, L. et al. (1988) IQ as a protective factor for subjects at high risk for ante social behaviour. Journal of Consulting and Clinical Psychology, 56, 224–6. Langdon, P.E., Maxted, H. and Murphy, G.H. (2007) An exploratory evaluation of the Ward and Hudson Offending Pathways Model with sex offenders who have intellectual disabilities. Journal of Intellectual and Developmental Disabilities, 32, 94–105. Lindsay, W.R. (2009) The Treatment of Sex Offenders with Developmental Disabilities. A Practice Workbook. Chichester: Wiley-Blackwell. Lindsay, W.R., Hogue, T., Taylor, J.L. et al. (2008b) Risk assessment in offenders with intellectual disabilities: A comparison across three levels of security. International Journal of Offender Therapy and Comparative Criminology, 52, 90–111. Lindsay, W.R., Law, J., Quinn, K. et al. (2001) A comparison of physical and sexual abuse histories: Sexual and non-sexual offenders with intellectual disability. Child Abuse and Neglect, 25, 989–95. Lindsay, W.R., Steptoe, L. and Beech, A.T. (2008a) The Ward and Hudson Pathways Model in sex offenders with intellectual disability. Sexual Abuse: A Journal of Research and Treatment, 20, 379–92. Lunsky, Y., Frijters, J., Griffiths, D.M. et al. (2007) Sexual knowledge and attitudes of men with intellectual disabilities who sexually offend. Journal of Intellectual and Developmental Disability, 32, 74–81. Marshall, W.L. and Barbaree, H.E. (1990) An integrated theory of sexual offending. In W.L. Marshall, D.R. Laws and H.E. Barbaree (eds), Handbook of Sexual Assault: Issues, Theories and Treatment of the Offender (pp. 257–75). New York: Plenum Press. Mednick, S.A., Brennan, P. and Kandel, E. (1988) Predisposition to violence. Aggressive Behavior, 14, 25–33. Mednick, S.A., Gabrielli W.F. and Hutchings B. (1984) Genetic influences in criminal convictions: Evidence from an adoption cohort. Science, 224, 891–4. Mednick, S.A., Moffitt, T., Gabrielli, W. and Hutchins, B. (1982) Genetic factors in criminal behaviour: a review. In D. Olweus, J. Block and M. Radke-Yarrow (eds), Development of Antisocial and Pro social Behaviour (pp. 33–50). London: Academic Press.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
OFFENDERS WITH MILD INTELLECTUAL DISABILITIES
255
Michie, A.M., Lindsay, W.R., Martin, V. and Grieve, A. (2006) A test of counterfeit deviance: A comparison of sexual knowledge in groups of sex offenders with intellectual disability and controls. Sexual Abuse: A Journal of Research and Treatment, 18, 271–9. Novaco, R.W. and Taylor, J.L. (2008) Anger and assaultiveness of male forensic patients with developmental disabilities: Links to volatile parents. Aggressive Behaviour, 34, 380–93. Quinsey, V.L., Book, A. and Skilling, T.A. (2004) A follow-up of deinstitutionalised men with intellectual disabilities and histories of antisocial behaviour. Journal of Applied Research in Intellectual Disabilities, 17, 243–54. Quinsey, V.L., Harris, G.T., Rice, M.E. and Cormier, C.A. (2006) Violent Offenders, Appraisal and Managing Risk (2nd edn). Washington DC: American Psychological Association. Rice, M.E., Harris, G.T., Lang, C. and Chaplin., T.C. (2008) Sexual preferences and recidivism of sex offenders with mental retardation. Sexual Abuse: A Journal of Research and Treatment, 20, 409–25. Sequeira, H. and Hollins, S.A. (2003) Clinical effects of sexual abuse on people with learning disability. British Journal of Psychiatry, 182, 13–19. Steptoe, L., Lindsay, W.R., Forrest, D. and Power, M. (2006) Quality of life and relationships in sex offenders with intellectual disability. Journal of Intellectual and Developmental Disabilities, 31, 13–19. Talbot, T.J. and Langdon, P.E. (2006) A revised sexual knowledge assessment tool for people with intellectual disabilities. Journal of Intellectual Disability Research, 50, 523–31. Ward, T., Bickley, J., Webster, S.D. et al. (2004) The Self-Regulation Model of the Offence and Relapse Process. A Manual: Volume I: Assessment. Victoria, BC: Pacific Psychological Assessment Corporation. Ward, T. and Hudson, S.M. (1998) A model of the relapse process in sexual offenders. Journal of Interpersonal Violence, 13, 700–25. Ward, T. and Keenan, T. (1999) Child molesters’ implicit theories. Journal of Interpersonal Violence, 14, 821–8. Ward, T. and Stewart, C.A. (2003) The treatment of sex offenders: Risk management and good lives. Professional Psychology, Research and Practice, 34, 353–60. Ward, T., Yates, P.M. and Long, C.A. (2006) The Self-Regulation Model of the Offence and Re-offence Process: Volume 2, Treatment. Victoria BC: Pacific Psychological Assessment Corporation. Webster, S.D. (2005) Pathways to sexual offence recidivism following treatment: An examination of the Ward and Hudson Self-Regulation Model of Relapse. Journal of Interpersonal Violence, 20, 1175–96. Yates, P.M. and Kingston, D.A. (2006) The self regulation model of sex offending. The relationship between offence pathways and static and dynamic sexual offence risk. Sexual Abuse: A Journal of Research and Treatment, 18. 259–70.
P1: OTA/XYZ P2: ABC JWST073-11 JWST073-McMurran
May 28, 2011
8:27
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
Chapter 12
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER LAWRENCE JONES Nottinghamshire Healthcare NHS Trust, UK
INTRODUCTION There is little written about case formulations for individuals who have personality disorders and who have offended. The aim of formulation is to link problems, processes of change, and outcomes, with a view to optimizing the clinician’s capacity to impact client outcomes (Koerner et al., 1996). To achieve this, practitioners need to apply theory to the individual case, and evidence from the literature of what is effective to the choice of treatment targets. Persons (2008) identified two types of theory used for this: generic psychological theories, such as behaviorism, and more specific theories, for example theories of sexual offending, such as Ward and Beech’s (2006) Integrated Theory of Sexual Offending (ITSO). Treatments can be applied using ‘off the shelf’ manualized programs; however, Drake and Ward (2003) suggest the need for case formulation when there is case complexity, lack of an evidence base, previous accepted treatment having been unsuccessful, and when clinicians need to understand why a therapeutic relationship has broken down. Personality disordered offenders generally meet most of these criteria. The distinguishing feature of personality-related causal factors for behavior is that they are pervasive and enduring, hence there is relative consistency in behavior across time and place. The task for the forensic clinician, therefore, is to identify themes running through a series of offenses. An individual might, for example, offend only when they are experiencing stress. Analysis of a series of offenses for one offender would identify this theme as an important factor (Jones, 2010a). This is a different process from applying a theoretical model to a particular case which involves matching features of the individual case to a standard model in a procrustean manner. Identification of patterns within an individual case involves exploration of the data to allow individualized themes to emerge. Scientific principles,
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
258
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
such as hypothesis testing through making predictions and collection of relevant data, can then be used to investigate those patterns (Davies et al., 2007) which may or may not then match onto models identified in the more generic literature. There are a number of dangers inherent in using the construct of personality in an overly rigid way to develop formulations. In order for the concept to be useful, personality needs to be conceptualized as something which is changeable, in spite of having been present for a long time. An alternative conceptualization may be one where there are relatively unchanging underlying processes whose end product is changeable. Enduring personality characteristics manifest themselves as behaviors via what Costa and McCrae (1992a) termed ‘characteristic adaptations’, which are changeable. Both of these conceptualizations avoid the danger of implicitly condemning the individual as unchangeable by using a construct of personality as something unchangeable (see Maruna (2001) on condemnation narratives). The literature on the relative importance of situation and personality points toward another challenge to the simplistic use of personality in explaining behavior. This has led theorists to take a more interactionist perspective, which recognizes the impact of context on behavior (Mischel and Shoda, 1995, 1999). Over-reliance on personality constructs in explaining behavior runs the risk of banal over-simplification; formulations involving personality constructs need to recognize the relevance of situational contribution. Kinderman and Tai (2007) distinguished single factor model and multi-factorial Psychological Model of Mental Disorder (PMMD) formulations. A single factor model formulation involves a particular psychological mechanism, such as schema activation, which is used to account for a range of problems. In contrast, PMMD assumes that psychological mechanisms are the final common pathway mediating between causal factors of all kinds and behavioral outcomes, and that a range of different psychological mechanisms underpin a particular problem. The transdiagnostic approach suggests a number of psychological mechanisms which have been subjected to rigorous evaluation and agreed as causal factors across a range of disorders (Harvey et al., 2004). These mechanisms include avoidance and disruptions in attention, memory, and reasoning. The implication is that clinicians must be up to date with the literature on the causes of the kinds of problem that they meet in their practice and not apply one causal model to all cases. Eells (2009) proposes using a common language for integrative case formulations using a variety of theoretical constructs. Jones (2010) proposed using the Morton (2004) causal modeling framework for precisely this purpose (see Figure 12.1). This was devised to be theory neutral and to facilitate formulations that involve a variety of different types of causal factors (e.g., frontal lobe deficit and deficits in behavioral repertoire). Morton developed this model specifically to enable researchers to compare a range of competing theories in accounting for specific problems. It achieves this by dividing putative causal factors into four types: environmental, brain, cognitive (which includes affective) and behavioral. The model implies the notion proposed by Kinderman (2005) and McGuire (2008) that psychological processes are the final common pathway for any causal process impacting on behavior. The Morton framework (Morton, 2004) is similar to Haynes’ (1997) Func¨ tional Analytic Clinical Case Model and Petermann and Muller’s (2001) treatment
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
Environment
High trait
Cognition Poor relationships
Low
Brain extraversion Agreeableness
Dominance and control reinforcing
Threat of abandonment from partner and mother
Poor compassion and offence supportive beliefs
Angry, distressed and Intoxicated states
259
Inhibition of High neuroticism dominant response deficit
Impulsive problem solving
Attachment and relationship schema: rejection abandonment Idealised women ‘falling off pedestal’
Adoption history
Behaviour Reinforcement history linked with coercive sex
Drinking
Attempted rape and violent assault
Figure 12.1 Morton diagram identifying interactions between contributory factors.
justification diagrams, as they all use a diagrammatic strategy for describing causal processes and encourage the clinician to think about the range of factors impacting on the target behavior. It also, usefully, separates causal factors into distinct domains in a manner similar to Padesky and Mooney’s (1990) five-part model. Morton combines emotion and cognition into one domain, otherwise the two frameworks are identical. The clinical imperative of integrating personality models (e.g., Livesley 2003, 2007) and offending behaviour models (e.g., Andrews & Bonta, 2003) makes the need for this kind of integrative framework all the more critical.
PERSONALITY AND OFFENDING Hanson and Morton-Bourgon’s (2004) meta-analytic review of research evidence concerning recidivism risk factors for sexual offenders identified two risky traits: antisocial orientation and lifestyle instability. As personality disorders are associated with lifestyle instability, this is likely to play a significant role in offending. Hanson and Morton-Bourgon indicated that these traits are characterized by impulsiveness, reckless behavior and hostile, resentful attitudes. Antisocial orientation, which included antisocial personality, antisocial traits and a history of rule violation, was the major predictor of violent nonsexual recidivism, violent and sexual
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
260
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
recidivism and, indeed, any type of recidivism. This could, however, be seen as tautological. Duggan and Howard (2009) challenged the overly simplistic use of the construct of personality in the explanation of offending, advocating instead causal models where it is seen as one of a range of potential contributory factors. Andrews and Bonta’s (2003) Risk Need responsivity (RNR) model also suggests personality as a responsivity factor, that is a factor that can impact on an individual’s capacity to engage in an intervention. Singer (2005) developed a framework for conceptualizing the construct of personality disorder based on the work of McAdams (1995) and Costa and McCrae (1992a). This framework involves four domains identified as different levels of personality description. These four domains, which will be explained in more detail below, are traits, characteristic adaptations, narrative identity and relational dynamics.
Traits At least three trait theories have relevance to personality disordered offenders. Costa and McCrae’s (1992a) five personality factors – Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism; Blackburn’s two higher order personality dimensions – Coercive and Withdrawn – (Blackburn et al., 2005; Krueger and Piasecki, 2002; Watson, Kotov and Gamez, 2006); and Gray’s (1987) behavioral activation and inhibition model. In his review of personality and offending, Egan (2009) proposed the “FFM [Five Factor Model] as an integrative model and fulcrum by which to organize what can be disparate concepts into a more coherent scheme” (p. 76). He described low Agreeableness, high Neuroticism and low Conscientiousness as predictors of violent offending. In his review, he found that offenders were more likely to be high in Neuroticism, due to the facets of angry hostility, depression and impulsiveness. Egan (2009) and Dennison, Stough and Birgden (2001) identified different personality profiles associated with different kinds of offender. For example, Dennison, Stough and Birgden (2001) (reviewed in Egan, 2009) found incestuous offenders higher in Neuroticism than controls; they also found extrafamilial child sex offenders to be even higher in Neuroticism than either controls or intrafamilial offenders. These differences were strongest for the facets of anxiety, depression, selfconsciousness and vulnerability. Nonoffenders were significantly higher than child sex offenders for Extraversion for the facets of assertiveness and gregariousness. Blackburn et al. (2005) argued that the dimensions of hostility vs. affiliation and dominance vs. submission map onto a number of other key personality constructs in a systematic way. This two factor model can usefully summarize much of the clinically relevant information for any individual case and has the heuristic utility of enabling constructs from different theoretical frameworks to be translated into one simple model. This model also converges well with evolutionary psychological thinking such as Gilbert’s rank status theory (Gilbert, Price and Allan, 1995), which identifies dominance and status as a critical dimension of social behavior. A related two factor theory, Gray’s Reinforcement Sensitivity Theory, proposes a model involving a Behavioral Inhibition System (BIS) that influences sensitivity to punishment and a Behavioral Activating System (BAS) that influences sensitivity
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
261
to reward (Gray and McNaughton, 2000). Persons (2008) illustrated the use of this two factor model for developing case formulations. Hypomania, for example, is conceptualized as an overactive BAS, and depression as an underactive BAS and an overactive BIS. This model is also relevant to personality disordered offenders; many accounts of psychopathy use this framework to describe the characteristic psychopathic presentation as low BIS and high BAS (see Wallace and Newman (2008) for a review).
Transient State Repertoires as Mode of Trait Expression Westen, Gabbard and Blagov (2006) argued that clinicians find both the categorical and the dimensional models of personality disorder unhelpful for building case formulations. What is needed for developing psychological interventions, they argued, is a framework for using dimensional data in a way that makes it meaningful to the clinical problem of identifying “cognitive, affective, motivational and behavioral patterns” (p. 367). To do this, it is necessary to have a model for linking personality and state repertoires elicited in particular ways for individuals in different contexts (Mischel and Shoda, 1999). This notion is akin to Staats’ (1993) behavioral conceptualization of personality as a behavioral repertoire. Staats conceptualized an individual’s state as part of the situation in which the relapse process evolves. In behavioral terms, states are establishing operations. Sturmey (1996), following Michael (1982), defined an establishing operation as ‘the process that establishes a stimulus as a reinforcer, punisher, or changes the value of a reinforcer or punisher’ (p. 15). Sturmey also noted Michael’s (1982) identification of emotional states as establishing operations. These notions allow the clinician to move toward understanding relapse processes in terms of high risk states and situations. Ward, Polaschek and Beech (2006) also advocated exploring how underlying traits can be activated to produce transient states by triggering factors (Eysenck and Eysenck, 1980). This concept of trait activation is at the heart of Ward et al.’s attempt to analyse trait-like risk factors with a view to building causal models of offending. Livesley’s (2003) notion of state repertoires holds that each individual experiences a set of different dissociated psychological states linked with different interpersonal stances, attitudes and beliefs. Interventions aim to enable the individual to understand how they switch between states and the kinds of process associated with the state switches (see also Jones, 2004, 2010a). Baumeister, Catanese and Wallace (2002) similarly suggested a role for narcissistic personality traits and states in sexually coercive behavior. They presented evidence that narcissistic states are experienced by people with narcissistic personalities. These states are, for example, associated with increased susceptibility to reactance (Brehm, 1966), a negative, aversive reaction to the loss of behavioral options as a consequence of the imposition of external constraints. Young (2003) reported a similar decision to move away from a trait model and toward a state model with the schema mode as the primary conceptual construct. Schema modes are moment-to-moment emotional states. Various theorists have developed models for linking personality traits to transient mood states. Clark (2000), for example, argued for a dimensional model of affect that maps onto dimensional models of personality, with negative affectivity linking with Neuroticism and positive affectivity associated with Extraversion.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
262
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Akiskal (e.g., 1981) argued for a spectrum of affective disorders linking bipolar disorder with borderline personality disorder. Bentall (2003) also highlighted the links between Gray’s BIS/BAS model of personality and mood, with the BAS associated with positive appetitive emotions and the BIS with emotions such as anxiety. These models can be useful for hypothesizing, based on knowledge of personality, about the kinds of unique affective processes an individual might be likely to experience in the course of an offense. While emotion regulation is considered to be an important target for offenders (Cortoni and Marshall, 2001), the construct of state is broader and includes a range of intrapersonal processes, such as changes in attention, dissociation, depersonalization, cognitive and metacognitive capacity, and consciousness. A significant proportion of personality disordered offenders report transient psychotic episodes (Jorgensen et al., 1996). It is important to have a way of describing and understanding these, particularly with respect to their role in the offending process. Having a language for describing not just emotion and affect but also different states of consciousness (e.g., dissociative states) is an important task in enabling clinicians to include these in developing an understanding of an offense. It is also important to link these state concepts with states of relationship to others.
Characteristic Adaptations Characteristic adaptations are an individual’s unique adaptation to their particular constellation of traits. Maruna (2001) illustrates this concept when he evidences that when offenders desist from offending they often do not change their underlying personality traits. He argued that even participants who had completely given up crime still thought of themselves as adventurous, rebellious and independent . . .Whilst this cluster of personality traits might make . . .[people] well suited to careers as artists, athletes or even venture capitalists, they also raise the chances that an individual will engage in criminal activities when faced with certain circumstances. (p. 59)
The same trait can be expressed in a number of different ways, some more prosocial than others. Livesley (2003) used the concept of stimulus-seeking as an example, as illustrated in Figure 12.2. An individual can be understood as adapting
Active social life
Stimulus seeking
Cautious risk taking, high-risk sports, high-speed driving Recklessness, fringe lifestyle, drug use
Figure 12.2 Characteristic adaptations: different expressions of the same underlying trait.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
263
to both their internal context (temperament, mood dynamics) and their external environment (ecological niche) (Ward, Polaschek and Beech, 2006). Intervention is aimed at finding an alternative way of meeting their needs. The Good Lives Model (Ward and Brown, 2004; Ward, Mann and Gannon, 2007) is about identifying an individual’s typical reinforcers and their characteristic responses to obtain these typical reinforcers so as to find alternative, prosocial ways of meeting these needs. Ward, Mann and Gannon (2007) offer a framework for strengths-based formulations that focus on meeting unmet needs and avoids the pejorative flavour of deficit-based formulations. They make the point that most deficit-based interventions can be translated into a needs based-model. For example, the Basic Behavioral Repertoire (BBR) model of personality (Staats, 1993) suggests that extending an individual’s repertoire for meeting their needs is going to reduce the possibility of them returning to an old self and employing destructive means of meeting needs.
Narrative Identity Narrative identity is the way in which an individual ascribes meaning to his or her life. Maruna’s (2001) work on desistence emphasized the importance of narrative in shaping an individual’s identity. This work derives its theory of personality from McAdams’ (1995) notions of narrative identity following two basic themes: Agency and communion. Analysing an individual’s narrative identity can be a critical task in assessing and intervening with individuals with a fragmented sense of self or with a significantly foreclosed sense of the future (see also Markus and Nurius’ 1986 construct of ‘possible selves’). Maruna (2001) identifies the importance of “redemption narratives” in the accounts of desistance obtained from ex-offenders who had successfully stopped offending, and “condemnation narratives” obtained from those who persist in offending. Understanding the role of this kind of narrative in people’s change beliefs can be central to the task of establishing and maintaining motivation. Another common theme in offenders’ narratives is that of childhood abuse linking with a range of adult offending behaviors. If this aspect of their sense-making is ignored, then it is likely that there will be significant problems with engagement. Assessment using the Good Lives Model (Ward et al., 2007) can help to identify what kinds of goals and values are important to an individual which can in turn help in making sense of the way values and goals shape the narrative an individual creates to make sense of his or her life. Fragmented or incoherent narratives are considered to be one of the indicators of a problematic attachment style in the Adult Attachment Interview (George, Kaplan and Main, 1996). Various interventions are aimed at developing a coherent sense of self where the individual is able to think about and experience themselves and others in a consistent way.
Relational Dynamics Relational dynamics are the ways an individual interacts with others. Blackburn (1998) highlighted the way in which people pull reciprocating roles from others, dominance pulling submission, submission pulling dominance, hostility pulling
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
264
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
MODEL OF SELF (Dependence)
Positive (Low)
Positive (Low)
Negative (High)
Secure
Preoccupied
Dismissing
Fearful
MODEL OF OTHER (Avoidance) Negative (High)
Figure 12.3 A two-factor model of attachment style.
hostility and friendliness pulling friendliness. Ryle (1997), Livesley (2003), and Gilbert (2000) have independently identified the way that personality style reflects not only interpersonal relating but also how individuals relate to themselves. An individual can have a harsh and controlling relationship with himself or herself, for example. Attachment issues are also part of this domain. Blackburn et al. (2005) wrote: “From an attachment theory framework Bartholomew et al. (2001) have hypothesized that the cognitive affective dimensions of positivity and negativity of others and of self underlie attachment styles, interpersonal styles and personality disorders” (p. 620). Bartholomew et al. (2001) suggested the mapping of attachment styles onto the two dimensions of personality, as illustrated in Figure 12.3. Other theorists (e.g., Blair, 2005) also have questioned the validity of an attachment based model suggesting that attachment style may derive from other underlying factors. Many theorists however do give a significant role to attachment. Insecure attachment is a significant underlying construct behind unstable relationships, which is one of Hanson and Morton-Bourgon’s (2005) strongest predictors of reconviction. Ward, Hudson and Marshall (1996) found that most offenders were insecurely attached. They found rapists and violent offenders to have more avoidant attachment styles. Ward, Hudson and McCormick (1997) suggested three types of insecure attachment amongst sexual offenders: (1) preoccupied, mostly amongst pedophiles; (2) fearful avoidant, mostly amongst rapists; and (3) dismissive avoidant, amongst those more hostile in their offending. Smallbone and Dadds (1998) found that sex offenders generally have less secure attachments. Incest offenders had more anxious and avoidant attachment histories; this they suggest is linked with incest offenders offending in the context of adult relationships. Rapists were found to be more avoidant in their attachment styles. Pedophiles were found to be both anxious and avoidant. Knowing about an individual’s attachment style is not clinically useful, however, until it is conceptualized in terms of a particular individual’s expression of that attachment style. What does this individual do at different stages of a relationship
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
265
– alone, starting a relationship, consolidating a relationship, becoming estranged in a relationship and during or following the end of a relationship? How does their attachment style impact on each of these stages? The concept of state attachment requires the clinician to identify the characteristic way this individual reacts to different eventualities such as avoiding, forming and losing of relationships.
ASSESSMENT Risk Assessment Actuarial and structured clinical judgment measures such as the Historical, Clinical, Risk Management-20 (HCR-20) (Douglas et al., 1999), include personality disorder as a risk factor. Instruments like the Violence Risk Scale (VRS) and the Violence Risk Scale Sex Offender version (VRS-SO) (Wong and Olver, 2009) are useful in that they prompt the clinician to look at a range of criminogenic factors and to assess the extent to which an individual has addressed these. They do not, however, address personality factors in relation to risk. In addition, a careful review of file information and interviews with informants such as family and previous therapists can help to identify factors that might be relevant to developing a risk management plan. Clinical interviews and observation of Offense Paralleling Behavior (OPB: Jones, 2004, 2010a) is useful for risk assessment. They allow the clinician to identify in what ways in the current setting the individual meets the needs that their offending met, and the extent to which they are meeting these needs in an alternative manner. The outcome from an actuarial assessment of risk is an estimate of the probability that an individual will offend again. OPB can provide useful insights into relapse processes in action. Observation of personality-related factors in the offense process is important with personality disordered offenders. Analysis of the range of demographic and psychological factors identified in the literature as being correlated with reconviction is useful (Andrews and Bonta, 2003); however, it is also important to identify the individual’s range of strengths and capacities. It is increasingly being recognized that strengths-based interventions are both ethically more appropriate and that they are more attractive to the client (Kuyken, Padesky and Dudley, 2009). Jones (2009, 2010 a,b) discussed the use of solution focused and narrative approaches to exploring what works for the individual case.
Assessment for Case Formulation Personality measures such as the International Personality Diagnostic Examination (Loranger et al., 1994), the Personality Adjustment Inventory (Morey, 2007), BIS/BAS (Carver and White, 1994), the Chart of Interpersonal Reactions in Closed Living Environments (Blackburn, 1998), and the five personality traits using the NEO-Five Factor Inventory (Costa and McCrae, 1992b) can be useful to understand the individual’s personality structure. Measures of criminogenic thinking (e.g., the
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
266
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Psychological Inventory of Criminal Thinking Styles; Walters, 1995a, 1995b) can also be useful. Daily behavior can be monitored using the Daily Behavior Rating Scale (DBRS; Hogue and Beeley, 2005). In the DBRS, staff rate the observed behavior of the patient on violence/aggression, threatening behavior, anxiety, impulsivity, emotional dysregulation, sexualized behavior, and self-harm twice daily. In addition, the DBRS allows for the recording of individually tailored variables. Measures of cognitive functioning can also be a useful measure of interpersonal change mechanisms, based on the principle that a stable therapeutic reparative attachment experience has the effect of improving the capacity to mentalize (e.g., Fonagy et al., 2002). Kelly (1955/1991) wrote “if you do not know what is wrong with someone, ask them; they may tell you” (1955, p. 323; 1991, p. 241). Kelly’s (1955) Life stories and Self Characterizations can be useful ways of accessing an individual’s narrative identity. Self-report, observation, use of psychometric assessments, and use of collateral information then allow for cross comparison of data sources to establish congruence and disagreement. Thematic analysis of offending (Jones, 2004), which looks for themes in structure and function of behavior chains leading up to offenses, can be useful to establish what learning mechanisms might operate. Observation of OPBs can be useful both as a way of gaining information for the formulation and as a form of baseline and post-treatment measure of outcome (Jones, Shine and Daffern, 2007; Jones, 2010a).
Fictional Case Formulation Christopher Christopher was 32 years old. Diagnostic personality disorder assessment using the International Personality Disorder Examination (Loranger, 1999) indicated that he met the criteria for borderline and antisocial personality disorders. Assessment using the NEO-FFI (Costa and McCrea, 1992b) indicated elevated scores on the Extraversion and Neuroticism scales and a low score on the Agreeableness scale. Interviews about his relationship history suggested that his attachment style was predominantly anxious and dismissive. A neuropsychological assessment using the Wisconsin Card Sorting Test (Berg, 1948) identified him as having problems with inhibiting responses once they had been established and suggested some problem-solving deficits around solution generation and choosing optimal solutions. These problems were exacerbated when he was experiencing stress. Assessment using the BIS/BAS scales (Carver and White; 1994) indicated that he had high BAS scores and a low BIS score. This suggested high drive and reward responsiveness in association with low anxiety Christopher described himself as someone who was suspicious of others and concerned about being dominated by his peers. He often sought to be near the top of the pecking order on the ward. He also reported that he was very easily bored, and when he was outside custodial settings he would often offend simply out of the need for something exciting to do and to offset feelings of boredom and listlessness.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
267
Christopher was adopted at the age of five. He grew up feeling like an outsider in this family and felt that they did not really like him. He described having worried a lot about why his birth mother had put him up for adoption, feeling angry at being abandoned and often fantasizing about telling his real mother what effect this had had on him. He described thinking about this a lot when he had been drinking and when he had had rows with his adoptive mother and his own partner. His relationship was unsettled and he and his partner frequently rowed. His partner was often emotionally labile and would attempt to push him into being more submissive, rowing with him if he fought back. He often thought that his partner was going to leave him and typically he left her first or became assaultive and controlling with her. Either way, he would then return to her the next day and apologize. He had an idealized view of how the relationship had been when it started and was preoccupied with trying to return to this state, with the result that he would be disappointed and begin to become distressed again. Before being apprehended, Christopher was involved in a series of offenses. He had a history of petty crime and theft. Prior to the index offense and the series of offenses leading up to it, he had one previous sexual offense, when he indecently assaulted a girl in his school with whom he had become obsessed and who rejected him. He followed her from school and grabbed her, and placed his hand under clothes and touched her vagina. He said that he could not cope with her rejecting him and he had hoped that if he pushed her she would respond to his advances. The offense sequence leading up to his detention custody involved a series of attacks on lone adult women unknown to him. These are listed in Table 12.1. When caught, he received a life sentence. In prison, he commenced therapy and began to make progress. He formed a particularly strong attachment with his probation officer. Halfway through the group he was given some feedback about a belief that women who teased men sexually deserved to be assaulted. He became angry saying that the facilitators had deliberately done this in order to upset him. Later he told his probation officer that he wanted to have sex with her and took her hostage. It was considered serious enough for him to be charged and he was referred to a high secure hospital. Following his arrival in the high secure hospital, he withdrew. He later engaged in treatment, reporting that he found it hard to take being challenged. At times when he perceived that he had been challenged, he would switch into what he described as his ‘paranoid head’, seeing all staff as ‘having it in for him’ and setting him up to fail. When asked, Christopher indicated that he believed that he could change. He expressed the view that he could make things better and he had hopes for the future. He was interviewed about times when he had managed not to offend when he had felt an urge to.
Formulation and Treatment Plan Each of the hypothesized causal mechanisms contributing to the development of his offending are listed in Table 12.2. See also the Morton diagram in Figure 12.1. It includes a treatment strategy, assessment strategy and suggested OPB to be observed to monitor progress. Care is taken to ensure that the targets are
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
268
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Table 12.1 List of attacks prior to the index offence illustrating development of themes of rejection, loss of temper, victim seeking and assault. Arguing about infidelity/rejection
Leaving in a temper and drinking
Follows/seeks victim
Argued with partner, after seeing her talking to somebody in a pub and thinking that she was flirting. Convinced that she was going to leave him Accused partner of being interested in other men and said that he thought that she was looking for the opportunity to leave him Had a row with his mother which became very heated. Was thrown out of the house by mother
Stormed out of house in a state of agitation and distress. Drank beer
Followed a 26-year-old woman who was alone
Assaulted his partner violently and left the house, Went to a pub and began drinking alone
In the pub, he saw a woman. He followed her home
He felt very angry at being thrown out of the house. Drank beer
Approached his former partner and his mother for help and somewhere to live and was turned down
Homeless and drinking for most of the day. Got into fight. Walked away
Looked for a victim to offend against. Approached a woman who waiting at a bus stop Went to park and waited for victim
Assault Grabbed woman, pulled her off the main road into an alley way and assaulted her sexually, putting his hands under her clothing Assaulted woman and took some of her clothing off, before she screamed and he ran away Grabbed woman and threatened to stab her if she did not do as he said. Attempted to rape her at the bus stop Attacked a young woman. Threatened to kill her, assaulted her, attempted to rape her and left her unconscious
clearly specified and that the treatment, assessment and OPB specification are directly linked with each causal mechanism. In the absence of an evidence base ¨ this tabulation serves as a defensible treatment justification (Petermann and Muller, 2001). Jones (2004) proposed that offending behavior may be seen as a process not as an event. In behavior analytic terms, this means analyzing response chains using functional assessment methodology. OPB analysis (Jones, 2004) requires a thematic analysis of functions in repeating patterns of offending behavior and associated antecedents. The apparently repeating pattern of over involvement with female staff, then him getting his “paranoid head” on, interpreting their behavior as being rejecting and then going on to behave in an inappropriate way was seen both as therapy interfering (Linehan, 1993) and OPB. A prediction was made that if this behavior pattern continued then Christopher could become overly involved again and in this context experience rejection or abandonment. It was further predicted
Information about previous relationships from interview, file data, previous workers Therapist’s observation of his in-session capacity to form and maintain attachments with therapists and others. Focus particularly on perceptions of “being rejected”, humiliated or abandoned by those in his current social world. Observation and self-rating of readiness to recognize own and others’ feelings when angry (rating scale 0–5). Observation of aggressive or coercive behavior such as intimidating peers or staff or making threats of violence, being used to manage rejection, humiliation and abandonment feelings – particularly in the context of perceptions of relationships having deteriorated
Assessment strategy
(Continued)
Work to build emotional regulation skills involving psycho-education and coaching during situations when he became distressed to implement skills and insights developed Relationship, problem solving and mentalization skills to improve capacity to read others’ intentions
Building insight and rehearsing alternative responses following rejection episodes and relationship crises in custody Building a healthy therapeutic relationship with therapist and staff. Modeling skills in repairing the relationship when it goes through difficulties Perspective taking skills encouraged when in angry state
Intervention and model of change
5:56
Using coercive or violent behavior – including sexually coercive behavior – to manage feelings of rejection by taking control of person, thereby preventing abandonment. Monitored using DBRS and case notes Fantasies and urges rated and monitored by patient (rating scale 0–5)
Observation of similar reactions involving beliefs about “not being wanted” and angry feelings, to perceived rejection, humiliation or abandonment on the ward, in reality or in self-reported fantasy Self-report and ratings of rejection and humiliation (rating scale 0–5). Also monitoring of alternative thinking, of being wanted and valued (rating scale 0–5)
Offence paralleling behavior (outcome measure)
May 26, 2011
Offending was used to offset distress around feeling as if he was rejected, humiliated or abandoned
Experiences of rejection, humiliation or abandonment (states linked with borderline personality disorder). These trigger an underlying belief that he is not wanted and also trigger a state of anger associated with thoughts about being picked on and deliberately attacked. When he feels angry and “picked on” there is a deterioration in his ability to be sensitive to/aware of his own and others’ feelings
Hypothesized causal factors/function
Table 12.2 Treatment targets, assessment strategies, offence paralleling behavior and intervention.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran Printer Name: Yet to Come
Measures of “approach explicit” attitudes and urges States of “thrill- seeking”; schema and cognitive distortions associated with hypomanic states Measures of masculine identity (Mosher and Sirkin’s (1984) Hypermasculinity Inventory) and violence supportive beliefs
Evidence of use of extremes on rating scales Interviews with therapists and significant others to get sense of extent to which “black and white thinking” was an issue in these relationships
Offending involved him feeling as if he was in control and powerful and had given him a “buzz”
Deterioration in his relationships prior to his offending had been partly due to his characteristic pattern of “black and white thinking”
Self monitoring for “black and white” thinking Using other skills to deal with shame associated with ‘being wrong’ Experimentation with using “gray area”
Exploration of the meaning of erectile dysfunction Psycho-education to help him recognize that incapacity “perform” sexually is not an indication that he is “not a man”
Encouraging alternative ways of seeking legitimate power, tolerating powerlessness and boredom
Revenge urge management Challenging revenge supportive beliefs
Intervention and model of change
5:56
Ongoing use of violence as a way of managing uncertainty about identity. Expressions of legitimization of violence as a way of establishing a masculine identity, particularly following incidents where his identity may have been challenged Arguments involving rigid use of “black and white” thinking leading to deterioration in relationships with staff and peers. Rated by named nurse in weekly sessions
Evidence of power and thrill seeking orientation in other domains, in reality or in fantasy
Evidence of revenge and revenge being generalized to others (i.e. assaulting other people than the person who feel has slighted you) in reality or in fantasy. Monitored using fantasy and revenge urge diaries, daily case notes
Offence paralleling behavior (outcome measure)
May 26, 2011
Confusion about his sexual identity was also offset by his violent behavior as he believed that violence made him feel like he was a “man”
Evidence of use of revenge as a part of skills repertoire for dealing with slights Fantasy monitoring Beliefs about revenge as legitimate (self report in interview)
Assessment strategy
Offending served to meet his desire for revenge
Hypothesized causal factors/function
Table 12.2 (Continued)
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran Printer Name: Yet to Come
Substance misuse history and interviewing about states entered under influence of alcohol. Urine testing. Interview and psychometrics (e.g. Walters 1995 a,b)
Problem-solving and self regulation skills audit Psychometrics (e.g. BIS-11 Patton et al., 1995) Observation Profiling values and goals Identifying goals that are most important and least important (Ward and Brown, 2004)
Drinking disinhibiting
Holding offence supportive beliefs.
Impulsive problem solving style
Unsatisfying lifestyle with significant needs not being met
5:56
Education, occupational therapy Encouragement to identify and engage in spiritual or similar meaning generating activity
Problem solving skills training intervention Skills rehearsal in dealing with real life ward problems
Substance misuse interventions using a relapse prevention approach and establishing alternative “approach goals” to meet needs previously met through drinking Identifying function of offence supportive beliefs Finding alternative ways of meeting these needs
May 26, 2011
Evidence of impulsive problem solving on the ward Evidence of failure to implement existing self regulation skills. Monitored by staff and self reported by patient Continuation of limited quality of life and restricted range of activities that were evident at the time of the offence
Expressing and sharing offence supportive beliefs in peer group. Observed by staff in group contexts
Drinking or using drugs on the unit, linked with inappropriate behavior
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
272
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
that at these times he would be prone to either becoming hostile to female staff – in reality or in fantasy. It was hypothesized that if he experienced a secure attachment, in the context of a therapeutic relationship, then his capacity to mentalize could improve (Fonagy et al., 2002). Staff were encouraged to recognize when he was using prosocial coping strategies when he had started to go down the pathway described above. It was predicted, based on accounts of successful previous self-control, that he might use minor offending as a form of self-regulation to get himself into seclusion to prevent himself from doing something more serious. It was also predicted that at times such as these he would be potentially accessible to talking things through with staff to prevent relapse.
Communicating the Formulation with a Brief Summary Document In order to ensure that the formulation has clinical utility a simplified version of it was developed in order to inform Christopher and the staff working with him about the formulation. It was written together with Christopher and involved getting his agreement to address the treatment targets. It was, thus, a statement of intention as well as a document describing the formulation. The document took one side of A4. The headings were problems areas that have caused offending, strengths, OPB, things staff want to help me do more of, and the treatment plan.
Logical Integrity The most obvious test of the validity of a formulation is to look at the logical integrity of the arguments. Exploration of the temporal sequencing of putative causes, for example, can help to test to see if a particular cause is necessary or sufficient. So, if Christopher offends both when he is drinking and when he is sober, it cannot be argued that drink is a cause of his offending, only that it might have a speeding up impact on an existing offending propensity. Other tests of the logic of causal reasoning, e.g., Mills’ methods, can also be applied (the interested reader is referred to Irving and Cohen, 2001).
Validity and Reliability Kuyken (2006) offers four criteria for increasing validity and reliability of a case formulation. 1. Is the theory on which the formulation is founded evidence-based? Although this formulation could be criticized on the grounds that some of the targets do not have an evidence base for addressing offending behavior, there is very little research evidence for what works with personality disordered offenders, so this would be a criticism of most formulations with this population. Several of the predictors, e.g., offense supportive beliefs and impulsive problem solving style, are evidenced in the literature on predictors of sexual reoffending in Hanson and Morton-Bourgon’s (2005) meta-analysis. Others, however, are not, for example, offending as offsetting distress. These could, however, be justified
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
273
on clinical grounds as being psychologically plausible, however their status as causal factors were diminished by not meeting this criterion. 2. Was the formulation process adequate? The formulation was reviewed by peers and they considered whether there had been an adequate use of the case information and whether the clinician was using a systematic approach to offsetting heuristic biases (Kuyken, 2006). There was no clear evidence of using biases in the kinds of information being used for the formulation or the development of the causal model. 3. Can clinicians agree on it? The formulation was subjected to peer review at a psychology team meeting. There was a good level of agreement between members of the team’s formulations and those of the author of the formulation about the core interventions and treatment targets. Where there was disagreement this was explored and an attempt at developing consensus was made. 4. Does the formulation triangulate with the client’s experience and/or any standardized measures, professional/expert/panel consensus, and clinical supervisor’s impressions? This test involved assessing the formulation’s face validity with Christopher. Christopher was involved in the development of the simplified formulation. The final version was written down in report format and he was asked if it matched his own perceptions of his problems. Christopher agreed with much of it but felt that many of his problems were due to not having work. This difference of emphasis was explored with Christopher. A collaborative version of the formulation was written on one sheet of A4 for use by Christopher and staff. The impact on the target variables following one year’s intervention allowed for the exploration of possible changes in OPB linked with the formulation, namely sexualized behavior (rated by staff on a scale of 1–100 based on ward observation of behaviors, such as staring at women and inappropriate sexual comments), blackand-white thinking (rated by staff on a scale of 1–100 based on ward observation), and perceived rejection (rated by Christopher on a scale of 1–100). Data were analysed using SINGWIN software (Bloom et al., 2009). A significant downward trend was identified in sexualized behavior and black and white thinking, but not in ratings of perceived rejection (see Figure 12.4). Both he and staff on the ward indicated that these changes had impacted positively on his relationships. Results suggested the possibility that perceived rejection was not as significant as originally hypothesized in contributing to sexualized behavior and that problem solving style might be a more effective treatment target.
SUMMARY AND CONCLUSIONS Working with personality disordered offenders is a complex task and requires the practitioner to have the capacity to conceptualize not just a range of causal factors but also the ways in which these interact with each other. Knowledge of both the forensic literature on RNR (Andrews and Bonta, 2003) and the literature on personality disorder is necessary. In this chapter the emphasis has been on looking at personality as a contributory factor in offending and desistance processes. The reader is encouraged to access the literature on offending behavior in order to
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
274
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
90 80 70 Sexualized behaviour
60 50
Perceived rejection
40
Black and white thinking
30 20 10 0 Jan
Feb
Mar Apr
May
Jun
Jul
Aug
Sep
Oct Nov
Dec
Figure 12.4 Trends in clinician ratings of three of the target behaviors identified in the formulation: sexualized behavior, perceived rejection and black and white thinking, illustrating use of single-case methodology to test formulation.
get a good sense of the kinds of factors typically used to explain offending in nonpersonality disordered individuals. The formulation framework used in this chapter is based on Morton’s (2004) causal modeling approach and Kinderman and Tai’s (2007) psychological model of mental disorder. This requires the practitioner to employ a range of different psychological mechanisms, not just one limited set applied to all cases, to build an integrated account based on the current evidence base pertaining to the problems experienced by the individual. The risk of developing an overly complicated formulation is always present when the practitioner is incorporating a range of different psychological factors into their formulations. This risk can be offset by attending to how the formulation is communicated and testing for underlying variables that can account for several more superficial variables, for instance, an underlying belief that drives a range of different kinds of negative self-talk. In the end, the strength of a formulation has to be assessed by the extent to which it changes the problems that it addresses beyond what would have been done without the formulation. This is as much an issue relating to how it is communicated and how it is developed in the context of the therapeutic relationship as it is about the content of the formulation. The aim should be to get as close an overlap as possible between the individual’s perspective of the problem and the range of factors identified by the clinician. The exercise is as much about listening to the individual’s formulation and learning from this as it is about teaching and helping the individual access the information about their problems from the literature to which the practitioner has access.
REFERENCES Akiskal H.S. (1981) Subaffective disorders: Dysthymic, cyclothymic and bipolar II disorders in the “borderline” realm. Psychiatric Clinics of North America, 4, 25–46. Andrews, D.A. and Bonta, J. (2003) The Psychology of Criminal Conduct (3rd edn). Cincinnati,OH: .Anderson.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
275
Bartholomew, K., Kwong, M.J. and Hart S.D. (2001) Attachment. In W.J. Livesley (ed.), Handbook of Personality Disorders: Theory, Research and Treatment. New York: Guildford. Baumeister, R.F., Catanese, K.R. and Wallace, H.M. (2002) Conquest by force: A narcissistic reactance theory of rape and sexual coercion. Review of General Psychology, 6, 92–135. Bentall, R. (2003) Madness Explained: Psychosis and Human Nature. London: Penguin. Berg, E.A. (1948) A simple objective technique for measuring flexibility in thinking. Journal of General Psychology, 39, 15–22. Brehm, J.W. (1966) A Theory of Psychological Reactance. New York: Academic Press. Blackburn, R. (1998) Psychopathy and personality disorder: Implications of interpersonal theory. In D.J. Cooke, S.J. Hart and A.E. Forth (eds), Psychopathy: Theory, Research and Implications for Society (pp. 269–301). Amsterdam: Kluwer. Blackburn, R., Logan, C., Renwick, S.J.D. and Donnelly, J.P. (2005) Higher order dimensions of personality disorder: Hierarchical structure and relationship with the five-factor model, the interpersonal circle and psychopathy. Journal of Personality Disorders, 19, 597–623. Blair, J., Mitchell, D. and Blair, K. (2005) Psychopathy, Emotion and the Brain. Oxford: WileyBlackwell. Bloom, M., Fischer, J., and Orme, J. (2009) Evaluating Practice: Guidelines for the Accountable Professional (6th edn). Boston: Allyn & Bacon. Carver, C.S. and White, T. (1994) Behavioural inhibition, behavioural activation, and affective responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and Social Psychology, 67, 319–33. Clark, L.A. (2000) Mood, personality, and personality disorder. In R. Davidson (ed.), Anxiety, Depression, and Emotion (pp. 171–200). Oxford: Oxford University Press. Cortoni, F. and Marshall, W. (2001) Sex as a coping strategy and its relationship to juvenile sexual history and intimacy in sexual offenders. Sexual Abuse: A Journal of Research and Treatment, 13, 27–43. Costa, P.T. and McCrae, R.R. (1992a) The five factor model of personality and its relation to personality disorders. Journal of Personality Disorders, 6, 343–59. Costa, P.T. and McCrae, R.R. (1992b) Revised NEO Personality Inventory (NEO PI-RTM ) and NEO Five Factor Inventory (NEO-FFI). Odessa, FL: Psychological Assessment Resources, Inc. Daffern, M., Jones, L., Howells, K., Shine, J., Mikton, C. and Tunbridge, V. C. (2007) Refining the definition of offence paralleling behaviour. Criminal Behaviour and Mental Health, 17, 265–73. Day, A. and Bryan, J. (2007) Personality change and personality disorder: Some initial thoughts on the application of McAdams’ Triarchic Model to the treatment of personality disorder. Psychology, Crime and Law, 13, 19–27. Davies, J., Howells. K., and Jones, L. (2007) Evaluating innovative treatments in forensic mental health: A role for single case methodology. Journal of Forensic Psychiatry and Psychology, 18, 353–67. Dennison, S.M., Stough, C. and Birgden, A. (2001) The Big 5 dimensional personality approach to understanding sexual offenders. Psychology, Crime and Law, 7, 243–61. Drake, C.R. and Ward, T. (2003) Treatment models for sex offenders: A move toward a formulation-based approach In W. Ward, D.R. Laws, and S.M. Hudson (eds), Sexual Deviance: Issues and Controversies, Thousand Oaks, CA: Sage Publications. Digman, J.M. (1997) Higher-order factors of the Big Five. Journal of Personality and Social Psychology, 73, 1246–56. Douglas, K.S., Ogloff, J.R.P., Nicholls, T.L. and Grant, I. (1999) Assessing risk for violence among psychiatric patients: The HCR-20 risk assessment scheme and the Psychopathy Checklist: Screening Version. Journal of Consulting and Clinical Psychology, 67, 917–30. Duggan, C. and Howard, R. (2009) The ‘functional link’ between personality disorder violence: A critical appraisal. In M. McMurran, and R. Howard (eds), Personality, Personality Disorder and Violence (pp. 19–38). Chichester: Wiley-Blackwell. Eells, T. (2009) Contemporary themes in case formulation. In P. Sturmey (ed.) Clinical Case Formulation: Varieties of Approaches (pp. 293–315). Chichester: John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
276
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Egan, V. (2009) The ‘Big Five’: Neuroticism, extraversion, openness, agreeableness and conscientiousness as an organisational scheme for thinking about aggression and violence In M. McMurran and R. Howard (eds), Personality, Personality Disorder and Violence (pp. 63–83). Chichester: Wiley-Blackwell. Eysenck, M.W. and Eysenck, H.J. (1980) Mischel and the concept of personality. British Journal of Psychology, 71, 191–204. Fonagy, P., Gergely, G., Jurist, E. and Target, M. (2002) Affect Regulation, Mentalization, and the Development of the Self. New York: Other Press. George, C., Kaplan, N. and Main, M. (1996) The Adult Attachment Interview. Unpublished manuscript, Department of Psychology, University of California at Berkeley. Gilbert, P. (2000) Social mentalities: Internal ‘social’ conflicts and the role of inner warmth and compassion in cognitive therapy. In P. Gilbert and K.G. Bailey (eds), Genes on the Couch: Explorations in Evolutionary Psychotherapy (pp. 118–50). Hove: BrunnerRoutledge. Gilbert, P., Price, J. and Allan, S. (1995) Social comparison, social attractiveness and evolution: How might they be related. New Ideas in Psychology, 13, 149–65. Gray, J.A. (1987) The neuropsychology of emotion and personality. In S.M. Stahl, S.D. Iversen and E.C. Goodman (eds), Cognitive Neurochemistry (pp. 171–90). Oxford: Oxford University Press. Gray, J.A. and N. McNaughton (2000) The Neuropsychology of Anxiety: An Enquiry into the Functions of the Septo-Hippocampal System. Oxford: Oxford University Press. Greenberg, L. and Paivio, S. (1997) Working with Emotion in Psychotherapy. New York: Guilford. Gresswell, D.M. and Hollin, C.R. (1992) Towards a new methodology for making sense of case materials: an illustrative case involving attempted multiple murder. Criminal Behaviour and Mental Health, 2, 329–41. Hanson, R.K. and Morton-Bourgon, K.E. (2005) The characteristics of persistent sexual Offenders: A meta-analysis of recidivism studies. Journal of Consulting and Clinical Psychology, 73, 1154–63. Harvey, A., Watkins, E., Mansell, W. and Shafran, R. (2004) Cognitive Behavioural Processes across Psychological Disorders: A Transdiagnostic Approach to Research and Treatment. London: Oxford Press. Haynes, S.N., Leisen, M.B. and Blaine, D.D. (1997) Design of individualized behavioral treatment programs using functional analytic clinical case models. Psychological Assessment, 9, 334–48. Hogue, T.E. and Beeley, C. (2005) The Daily Behaviour Rating Scale (DBRS). Unpublished document, Rampton Hospital. Horowitz, M. (2001) Configurational analysis of the self: a state of mind approach. In J.C. Muran, J.C. (ed.) Self-Relations in the Psychotherapy Process (pp. 67–81). Washington DC: American Psychological Association. Horowitz, M.J. (1994) States, schemas, and control: general theories of psychotherapy integration. Clinical Psychology and Psychotherapy, 1, 143–52. Irving, C. and Cohen, C. (2001) Introduction to Logic (10th edn). Upper Saddle River, New Jersey: Prentice Hall. Jones, L. (2002) An individual case formulation approach to the assessment of motivation. In M. McMurran (ed.), Motivating Offenders to Change: A Guide to Enhancing Engagement in Therapy (pp. 31–54). Chichester: John Wiley & Sons, Ltd. Jones, L. (2004) Offence Paralleling Behaviour (OPB) as a framework for assessment and interventions with offenders. In A. Needs and G. Towl (eds), Applying Psychology to Forensic Practice (pp. 34–63). Oxford: Blackwell. Jones, L.F. (2009) Working with sex offenders with personality disorder diagnoses. In A.T. Beech, L. Craig and K. Browne (eds) Assessment and Treatment of SEX offenders: A Handbook (pp. 409–30). Chichester: John Wiley & Sons, Ltd. Jones, L.F. (2010a) Case formulation with personality disordered offenders. In A. Tennant, A. and K. Howells. (eds), Using Time, Not Doing Time (pp 45–62). Chichester: WileyBlackwell.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
277
Jones, L.F. (2010b) Working with people who have committed sexual offences with personality disorder diagnoses. In A. Tennant and K. Howells (eds), Using Time, Not Doing Time (pp. 125–40). Chichester: Wiley-Blackwell. Jones, L.F., Shine, J. and Daffern, M. (eds) (2007) Offence Paralleling Behaviour: A Case Formulation Approach to Offender Assessment and Intervention. Chichester: Wiley-Blackwell. Jorgensen, P., Bennedsen, B., Christensen, J. and Hyllested, A. (1996) Acute and transient psychotic disorder: Comorbidity with personality disorder. Acta Psychiatrica Scandinavia, 94, 460–4. Kelly, G.A. (1955) The Psychology of Personal Constructs. Vol. I, II. New York: Norton (2nd printing: 1991, London, New York: Routledge). Kinderman, P. (2005) A psychological model of mental disorder. Harvard Review of Psychiatry, 13, 206–17. Kinderman, P. and Tai, S. (2007) Empirically grounded clinical interventions: Clinical implications of a psychological model of mental disorder. Behavioural and Cognitive Psychotherapy, 35, 1–14. Koerner, K., Kohlenberg, R.J. and Parker, C. (1996) Diagnosis of personality disorder: A radical behavioral alternative. Journal of Consulting and Clinical Psychology, 64, 1169–76. Krueger, R.F. and Piasecki, T.M. (2002) Towards a dimensional and psychometrically informed approach to conceptualising psychopathology. Behaviour, Research and Therapy, 40, 485–99. Kuyken, W. (2006) Evidence based case formulation: Is the emperor clothed? In N. Tarrier (ed.), Case Formulation in Cognitive Behaviour Therapy: The Treatment of Challenging and Complex Cases (pp. 12–35). London: Routledge. Kuyken, W., Padesky, C.A. and Dudley, R. (2009) Collaborative Case Conceptualization. New York: Guilford Press. Linehan, M. (1993) Cognitive Behavioural Treatment of Borderline Personality Disorder. New York: Guildford. Livesley, W.J. (2003) Practical Management of Personality Disorder. New York: Guildford. Livesley, W.J. (2007) The relevance of an integrated approach to the treatment of personality disordered offenders. Psychology, Crime and Law, 13, 27–46. Loranger, A.W. (1999) International Personality Disorder Examination (IPDE). Odessa, FL: Psychological Assessment Resources. Loranger, A.W., Sartorius, N., Andreoli, A. et al. (1994) The International Personality Disorder Examination, IPDE. The WHO/ADAMHA International Pilot Study of Personality Disorders. Archives of General Psychiatry, 51, 215–24. McAdams, D.P. (1995) What do we know when we know a person? Journal of Personality, 63, 365–396. McGuire, J. (2006) An invited critique of Cognitive Analytic Therapy for offenders. In P. H. Pollock, M. Stowell-Smith and M. Gopfert (eds), Cognitive Analytic Therapy for Offenders: A New Approach to Forensic Psychotherapy (pp. 323–9). London and New York: Routledge. McGuire, J. (2008) A review of effective interventions for reducing aggression and violence. Philosophical Transactions of the Royal Society B, 363, 2577–97. McMurran, M. and Howard, R. (2009) Personality, Personality Disorder and Violence: An Evidence Based Approach. Chichester: Wiley-Blackwell. McMurran, M. (2009) Social problem solving, personality disorder and violence. In M. McMurran, and R. Howard (eds), Personality, Personality Disorder and Violence: An Evidence Based Approach (pp. 265–80). Chichester: Wiley-Blackwell. Maruna, S. (2001) Making Good: How Ex-convicts Reform and Rebuild Their Lives. Washington DC: American Psychological Association. Markus, H. and Nurius, P. (1986) Possible selves. American Psychologist, 41, 954–69. Michael, J. (1982) Distinguishing between the discriminative and motivational functions of stimuli. Journal of Experimental Analysis of Behaviour, 37, 149–58. Mischel, W. and Shoda, Y. (1995) A cognitive-affective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246–68.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
278
May 26, 2011
5:56
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Mischel, W. and Shoda, Y. (1999) Integrating dispositions and processing dynamics within a unified theory of personality: The Cognitive Affective Personality System (CAPS). In L.A. Pervin and O. John (eds), Handbook of Personality Theory and Research (2nd edn, pp. 197–218). New York: Guilford. Morey, L.C. (2007) The Personality Assessment Inventory professional manual. Lutz, Odessa, Florida: Psychological Assessment Resources. Morton, J. (2004) Understanding Developmental Disorders: A Causal Modelling Approach. Oxford: Blackwell. Mosher, D.L. and Sirkin, M. (1984) Measuring a macho personality constellation. Journal of Research in Personality, 18, 150–63. Padesky, C. A., and Mooney, K. A. (1990) Clinical tip presenting the cognitive model to clients. International Cognitive Therapy Newsletter, 6, 28–9. Panksepp, J. (1998) Affective Neuroscience: The Foundations of Human and Animal Emotions. New York: Oxford University Press. Patton, J.M., Stanford, M.S. and Barratt, E.S. (1995) Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51, 768–74. Persons, J.B. (2008) The Case Formulation Approach to Cognitive Behavior Therapy. New York: Guilford. ¨ Petermann, F. and Muller, J.M. (2001) Clinical Psychology and Evidence-Based Practice: A Practical Approach to Treatment Planning and Evaluation. New York: John Wiley & Sons, Inc. Ryle, A. (1997) Cognitive Analytic Therapy for Borderline Personality Disorder: The Model and the Method. Chichester: John Wiley & Sons, Ltd. Singer, J.A. (2005) Personality and Psychotherapy: Treating the Whole Person. New York: Guildford. Smallbone, S.W. and Dadds, M.R. (1998) Childhood attachment and adult attachment in incarcerated adult male sex offenders. Journal of Interpersonal Violence, 13, 555–73. Staats, A.W. (1993) Personality theory, abnormal psychology, and psychological measurement: A psychological behaviorism. Behavior Modification, 17, 8–42. Sturmey, P. (1996) Functional Analysis in Clinical Psychology. Chichester: John Wiley & Sons, Ltd. Wallace, J.F. and Newman, J.P. (2008) Reinforcement sensitivity theory and psychopathy: Associations between psychopathy and the behavioral activation and inhibition systems. In P.J. Corr (ed.), Reinforcement Sensitivity Theory of Personality, (pp. 398–414). Cambridge: Cambridge University Press. Walters, G.D. (1995a) The Psychological Inventory of Criminal Thinking Styles. Part I. Reliability and preliminary validity. Criminal Justice and Behavior, 22, 307–25. Walters, G.D. (1995b) The Psychological Inventory of Criminal Thinking Styles. Part II. Identifying simulated response sets. Criminal Justice and Behavior, 22, 437–45. Ward, T. and Beech, T. (2006) An integrated theory of sexual offending. Aggression and Violent Behavior, 11, 44–63. Ward, T. and Brown, M. (2004) The Good Lives Model and conceptual issues in offender rehabilitation. Psychology, Crime and Law, 10, 243–57. Ward, T., Hudson, S.M. and McCormack, J. (1997) Attachment style, intimacy deficits, and sexual offending. In B.K. Schwartz and H.R. Cellini (eds), The Sex Offenders: New Insights, Treatment Innovations and Legal Developments (Vol. 2, pp. 2:1–2:14). Kingston, NJ: Civic Research Institute. Ward, T., Hudson, S.M. and Marshall, W.L. (1996) Attachment style in sex offenders: A preliminary study. The Journal of Sex Research, 33, 17–26. Ward, T., Mann, R.E. and Gannon, T.A. (2007) The good lives model of offender rehabilitation: Clinical implications. Aggression and Violent Behavior, 12, 87–107. Ward, T., Polaschek, D. and Beech, A.R. (2006) Theories of Sexual Offending. Chichester: John Wiley & Sons, Ltd. Watson, D., Kotov, R. and Gamez, W. (2006) Basic dimensions of temperament in relation to personality and psychopathology. In R.F. Krueger, J.L. Tackett (eds), Personality and Psychopathology (pp. 7–38). New York: Guildford.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
CASE FORMULATION FOR INDIVIDUALS WITH PERSONALITY DISORDER
279
Westen, D., Gabbard, G.O. and Blagov, P. (2006) Back to the future: Personality structure as a context for psychopathology. In R.F. Krueger, and J.L. Tackett (eds), Personality and Psychopathology (pp. 335–84). New York: Guildford. Wong, S.C.P. and Olver, M.E. (2009) Two treatment and change oriented risk assessment tools: The Violence Risk Scale and the Violence Risk Scale-Sexual Offender version. In R. Otto and K.S. Douglas (eds), Handbook of Violence Risk Assessment (pp. 121–46). New York: Routledge. Young, J., Klosko, J. and Weishaar, M. (2003) Schema Therapy: A Practitioners Guide. New York: Guildford.
P1: OTA/XYZ P2: ABC JWST073-12 JWST073-McMurran
May 26, 2011
5:56
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
PART V
CONCLUSION
8:33
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
Chapter 13
FORENSIC CASE FORMULATION: EMERGING ISSUES PETER STURMEY Queens College and City University, New York, USA
MARY MCMURRAN University of Nottingham, UK
In this concluding chapter, we will comment upon the previous chapters in this volume, review emerging issues in clinical case formulation generally and forensic case formulation specifically, and suggest directions for future research and practice in forensic case formulation. Many issues are common to all forms of case formulation. The issues of reliability and validity have haunted the literature on case formulation for over 30 years. The literature is full of vivid assessments of individual cases and sleight of hand translations of formulations into startling, innovative and counter-intuitive treatments. For example, Wolpe’s (1986) assessment of an apparent height phobia revealed that, although the person was fearful when in her apartment in a tall building, her fear was not a fear of heights, bur rather a fear of people thinking negatively about her which commonly happened when she was alone in her apartment. Hence, the counter intuitive treatment for this apparent height phobia was not exposure to heights – the clients already had such formulaic treatment and it had failed – rather, her treatment was based on exposure to imagining criticism from other people, a treatment that was highly effective. But such clinical scintillation should not blind us to the presence of the naked emperor in the room. In a series of reviews and empirical studies, Kuyken has repeatedly pointed out that despite the continuing interest in case formulation for many years, despite the recognition of case formulation as a core skill by many different professional organizations, despite over two decades of research into case formulation, the evidence supporting the use of case formulation remains sparse, incomplete and contradictory (Kuyken, 2006). The chapters
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
284
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
by Mumma on reliability and Ghaderi on treatment validity show that this parlous state of affairs continues despite calls to action.
THE CURRENT STATUS OF CASE FORMULATION Mumma has conducted a series of sophisticated empirical studies on case formulation (Mumma, 2001, 2004; Mumma and Mooney, 2007a, 2007b; Mumma and Smith, 2001). These studies form the basis of his chapter which presented a notable conceptual extension of reliability issues in case formulation. Most of the literature on the reliability of case formulation has asked piecemeal questions, such as do clinicians agree on a target behavior for change or the relevant variables that may influence the target behavior? (No studies have attempted to ascertain if clinicians can agree on the target behavior and the variables that influence the target behavior and write a treatment plan that matches the formulation.) By incorporating psychometric concepts of construct and content validity, interformulator reliability, internal consistency reliability, test-retest reliability and stability, predictive validity, construct validity, and discriminant and convergent validity, Mumma parses the problem of reliability and validity of case formulation into several component, focused questions. By asking each of these questions of psychodynamic and behavioral and cognitive-behavioral case formulations separately, Mumma revealed a number of surprising differences between the literatures on case formulation from different theoretical perspectives. Ghaderi, who has conducted his own study of the validity of case formulation (Ghaderi, 2006), was charged with answering the question of whether case formulation improves client outcomes – one of the main rationales for conducting a case formulation – and found mixed evidence to support the validity of case formulation. Some studies found that case formulation did not improve client outcomes compared to standard treatment. This may reflect a number of factors. For example, several trials that found no effect of case formulation on treatment outcome beyond standard treatments were conducted on treatments of anxiety disorders, such as phobias (Schulte et al., 1992) and obsessive compulsive disorders (Emmelkamp, Bouman, and Blaauw, 1994). These results may reflect the power of standardized treatments based on exposure therapy. It is also possible that the results of these studies may reflect the ineffectiveness of either case formulation or translating the formulation into an idiographic treatment by relatively inexperienced therapists, or the exclusion of more complex cases for the purpose of conducting an explanatory trial, such as when clients present with a co-occurring anxiety or mood disorder (Schulte et al., 1992). The failure to detect differences between treatments may also reflect the use of insensitive dependent variables or lack of experimental power. A possible example of the lack of experimental power may come from Jacobson et al.’s (1989) study of case formulation and marital problems. In this study the authors detected no effects of case formulation on outcome immediately post treatment, but subsequently resulted in a large effect on marital functioning six months’ post-treatment due to the standard treatment group deteriorating when the case formulation group did not. Although it is tempting to infer that there were only delayed treatment effects, treatment effects may have been present
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
285
immediately post-treatment but perhaps this study did not have sufficient power to detect them or perhaps did not use the right dependent variables that would have measured the change that may have in fact occurred. By contrast, Ghaderi presented evidence that case formulation seemed to make a difference to treatment out comes for depression (McKnight et al., 1984; Strauman et al., 2006), where the formulation of each person’s own problem seems to make a difference to treatment outcome. Perhaps even more impressive is Ghaderi’s (2006) own study of the benefits of case formulation for the treatment of bulimia, where standardized cognitive behavior therapy is robustly highly effective (Hay et al., 2009) both at the group level and for many, but not all, individuals. Ghaderi (2006) observed that standard treatment was highly effective, and hence there was relatively modest room for improving outcomes beyond the standard treatment. Nevertheless, treatment based on case formulation still improved treatment outcome beyond standard treatment: 80% of nonresponder participants were in the standard treatment group but only 20% of the nonresponder participants were in the case formulation group. In contrast with these mixed outcomes in studies of adult mental health problems, evidence for case formulation based on functional assessment and analysis is robust in the area of developmental disabilities, and there is growing evidence that this approach is effective in determining the most effective treatment outcome in other childhood and adolescent behavioral problems such as school refusal (Kearney and Silverman, 1990) and emotional and behavioral problems (Cipani and Golden, 2007). One enduring issue in case formulation is whether clinicians should make formulations using one theory – and if so which one – or whether they should use eclectic approaches to formulation. Eells and Lambert’s chapter on definitions of case formulation addresses this issue. They write With all this theory, these sources of empirical evidence, and these formulation models available, how should one choose among them? The clinician committed to a single theoretical orientation could answer this question straightforwardly simply by disregarding the approaches that do not fit his or her orientation. We do not recommend this approach since we believe that each theory, source of information and formulation model has something to offer. For this reason, we recommend initial consideration of several models, recognizing that there is overlap among a number of them (p. 24).
Additionally, one of their practical tips for case formulation is to “formulate a case from more than one theoretical approach or structured model” (p. 27). Thus, these authors adopted an eclectic position on this question. Tension remains between integrationist approaches to case formulation and monotheoretical approaches and between different monotheoretical approaches. One of us (Sturmey, 2009a) presented five examples of pairs of case formulations of the same case made by authors of contrasting theoretical perspectives. Each case and pair of formulations was accompanied by a commentary by an independent author. In the final chapter of that book, Sturmey (2009b) argued that integration was not possible beyond superficial comparisons between theoretical orientations. Certainly, it is possible for clinicians of any theoretical orientation to pick and
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
286
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
choose among treatments associated with different theoretical orientations; clinicians might do so for pragmatic reasons, such as a client’s treatment preference or availability of trained therapists for each treatment. For example, one might select Eye Movement Desensitization and Reprocessing (EMDR) as a convenient and acceptable vehicle for exposing a client to fearful images without subscribing to the theory behind EMDR because the client had experienced positive outcomes with that treatment in the past and because the treatment is readily available locally. But beyond technical eclecticism is a more serious challenge of conceptual eclecticism, which is much more difficult because formulations made from different perspectives give priority to or exclude different kinds of variables as relevant or irrelevant to formulation. For example, cognitive therapy adopts a model in which changing core schemas causes change in behavior (Persons, 2008; Kuyken et al., 2009), whereas behavior analytic approaches see private behavior (thinking and feeling) as covert behavior to be explained in the same terms as overt behavior, namely control by the environment (Chiesa, 1994; Skinner, 1953). Although empiricism can inform the debate with new facts, empiricism alone cannot resolve these conceptual and philosophical differences. For example, if a therapy associated with a different school of psychology from one’s own is shown to be effective, this observation is merely something to be explained within the terms of one’s own theory.
Forensic Case Formulations The case studies in this volume also illustrate the differences between approaches to forensic case formulations. Several of the formulations in this volume are explicit in adopting certain theoretical frameworks. For example, the formulation by Lindsay adopted Ward and Hudson’s (1998) four pathways model, which includes both a typology of client history and a feedback loop from the outcome to the precipitating factors that will affect (strengthen) the likelihood of the incident reoccurring . . . for active explicit offenders, the feedback loop is likely to be positive and self-gratifying . . . For the avoidant active offender, the negative evaluation is likely to lead to selfrecrimination and a determination to increase the frequency and intensity of the maladaptive strategies . . . (p. 243)
Thus, this formulation incorporates both cognitions as causes, such as selfgratification and self-recrimination, and (without using the actual words) positive and negative reinforcement of behavior. Other cognitive-behavioral approaches to forensic case formulation, while including cognitions and nonobservable constructs, such as personality, in the formulation, place greater weight on and are more explicit about situational control of violence and aggression. Howells included both personality as an antecedent and, building upon earlier analyses of aggression and violence in psychiatric settings (Daffern and Howells, 2002, 2007), a typology of functions, the Assessment and Classification of Function, both to classify individual incidents of aggression and violence and to develop a profile of an individual person’s behavior. Thus,
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
287
Howells contrasted imprisoned violent men. The first, Jim, showed violence that was mostly classified as due to anger expression and tension reduction, whereas the second, Chris, showed violence that served to force compliance, obtain tangibles, enhance status and was sensation seeking during times of boredom. Most explicitly behavior-analytic, rather than cognitive-behavioral, were Hart, Greswell, and Braham’s use of Gresswell and Hollin’s (1992) Multiple Sequential Functional Analysis (MSFA), which uses a series of interlocking AntecedentBehavior-Consequence analyses of a history to describe the learning that took place, to identify the functions of criminal behavior and any presenting symptoms and problematic behavior, and to develop an idiographic treatment plan. This approach to forensic case formulation is closest to behavior analysis, rather than cognitive-behavior therapy or behavior therapy, in that it explicitly uses behavior analytic concepts and language derived directly from Skinner (1953). Similar to Howells’s formulations of a pair of cases with topographically similar problems but functionally completely different problems, Hart et al. use MSFA to describe the different possible learning histories and functions of presenting problems; in so doing they develop very different treatment plans for each offender based on the MSFA. Thus, both of these chapters shared a common belief that function is more important that topography. Finally, several chapter authors presented case formulations referencing attachment theory and the development of criminal behavior. For example, Rich’s formulation of sexual acting out and violence in a 15-year-old with a history of extremely poor quality and abusive parenting and disrupted relationships from an early age used attachment theory and Erikson’s developmental model. Such a formulation is tied tightly to the person’s development as well as the current problems. Rich writes that Mitchell fails to feel a significant level of empathy for others, given his own early experiences of maltreatment, and a pattern of insecure avoidant attachment . . .. Mitchell’s early experience of the world as unresponsive, unreliable, untrustworthy, and quite possibly a dangerous place, has contributed significantly in Mitchell’s experience of the world fitting that very schema . . . (p. 230)
The treatment implied by this formulation includes “development of critical social skills, including those of social perspective taking, self-reflection and metacognition, and distress tolerance” (p. 231). Thus, although some of the treatment plan shares elements with standard cognitive-behavior therapy, such as social skills training, the conceptual basis of the formulation is quite different. In contrasting these approaches to formulation one cannot help but notice that despite some similarities there are fundamental differences. For example, both MSFA and attachment theory address development and history. MSFA, however, analyzes development using antecedents, behavior and consequences, references operant and respondent learning, behavioral excesses and deficits, and interventions are based on current strengths and the learning of key functionally equivalent adaptive behaviors. In contrast, Rich’s attachment theory based formulation does reference social skills, but also cognitive and emotional processes such as selfreflection and metacognition, to understand his own and others’ needs. In all the
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
288
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
chapters, there were no examples of explicitly eclectic formulations, although several formulations borrowed concepts and treatments from more than one school of psychology. Another substantive issue is the question of whether or not forensic clinicians can develop and effectively use clinical formulations to develop treatment plans that are more effective than standard treatment plans or treatments based on diagnoses. Again, as with reliability and validity, none of the chapters cited data on this issue.
Summary The issues in case formulation generally – its conceptual foundations, reliability and validity – are similar to those specifically in forensic case formulation. The authors of Chapters 4–12 all presented case formulations which were very similar to those to be found in other books on case formulations with different populations. Hence, despite the special challenges to forensic case formulation noted earlier, we can conclude that forensic case formulation is readily possible. It is notable, however, that there is almost no empirical literature on forensic case formulation. Thus, future case formulation researchers with the forensic populations should conduct some of the basic studies on reliability and validity and of training practitioners to make case formulations similar to those found elsewhere in the case formulation literature.
SOMETHING NEW The interaction between case formulation and risk assessment and management is one notable forensic issue which has implications for the safety of others and the disposal of the client him- or herself. Errors in judgment in routine case formulation results in moderate costs, such as delays in symptom reduction, but errors in forensic case formulation may result in additional significant adverse outcomes, such as repeat serious offending, significant injuries and trauma to others, and large costs of incarceration and long-term treatment. Accurate case formulation in forensic settings is crucial, yet there are often major obstacles to information gathering and hypothesis testing. The behavior under assessment – such as extreme violence, murder and sexual assault – cannot be elicited for examination for ethical and practical reasons. Additionally, the scope for testing hypotheses regarding antecedents and behavior is limited, partly because of the nature of the behavior under study, but also because secure environments eliminate many natural antecedents and consequences that are present in the environment where the offending behavior took place. Perhaps this explains why research in forensic case formulation is sparse. Nonetheless, research into forensic case formulation is absolutely vital for the benefit of individual clients and of society at large. It would be a highly satisfactory consequence of producing this book if research in this area were stimulated and promoted.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
289
Risk Assessment and Risk Management Risk assessment procedures have developed from clinical judgment, which was shown to be of poor predictive validity, through actuarial, evidence-based risk assessments, which showed much better predictive validity but consisted largely of static, historical risk factors that cannot be changed by any form of treatment (Bonta and Wormith, 2008). A later generation of risk assessments focused on dynamic factors that were predictive of future offending (Andrews, Bonta and Hoge, 1990). Dynamic risk factors are those that could be changed in treatment, and they include antisocial attitudes, criminal associates, poor family functioning low self-control, poor educational attainment, and unstable employment; however, these risk factors have been identified through group data, and they may not necessarily apply in individual cases (Cooke and Michie, 2010). The key to identifying which known risk factors apply in individual cases is through case formulation. Indeed, Mumma noted in his chapter that, although a dynamic risk factor might be of zero or of negligible magnitude for a group, it might be highly predictive for a specific person. Likewise, a dynamic risk factor might be highly predictive for the group, but may have no predictive value for any one person. Individual information should be used to inform decisions about each specific individual, including entry to mental health or criminal justice systems, detention or community disposal, and the design of individually relevant treatment and management plans. Case formulation clearly has a major contribution to play in offender risk assessment and management because many of the predictor variables in a formulation are dynamic risk factors. To the extent that a formulation-driven treatment plan might be effective, these predictor variables/dynamic risk factors might change. For example, suppose a forensic case formulation identifies periods of lack of stimulation and boredom as a risk factor for thrill-seeking dangerous behavior and a formulation-based treatment plan effectively teaches an offender a new repertoire of behavior that provides her/him with acceptable ways to obtain thrills. It would now be appropriate to both revise the formulation and eliminate periods of boredom as a predictor variable and also to modify the risk assessment for this person, since the initial risk factor is now no longer a dynamic predictor of violence. The interplay of case formulation and risk management poses many questions that need to be addressed by researchers and practitioners. The main issue is to what extent does the production of a case formulation that informs subsequent treatment and management plans lead to improved outcomes over actuarial assessments in terms of risk reduction? In Chapter 4, Stephen Hart and Caroline Logan gave a clear account of recent developments in risk assessment and risk management. The structured clinical judgment approach is founded on decision theory. Based on as comprehensive an understanding as possible of the individual and the factors that promoted a decision to offend, possible future scenarios are constructed. These then inform decisions about treatment and management. The question regarding this approach is whether it is a true advancement on actuarial methods or if, on the contrary, it is a return to the discredited clinical judgment method under a new guise. In a recent meta-review of forensic risk assessment, Singh and Fazel (2010) identified six meta-analyses that compared actuarial measures with
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
290
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
clinically based instruments. Five of the six meta-analyses found that actuarial methods produced higher rates of predictive validity. The sixth meta-analysis found no difference in efficacy between actuarial tools and those that employ structured clinical judgment. However, as Hart and Logan point out, structured clinical judgments guide the professional through an evidenced-based procedure and are not the same as unstructured and idiosyncratic professional judgments. Additionally, Hart and Logan remind us of the utility of structured professional judgments (SPJ) when they say that “of the two approaches to evidence-based violence risk assessment, only the SPJ approach assists the development of risk management plans based on an understanding of the causes of past violence. The actuarial approach is not intended and cannot be used for this purpose” (p. 91).
Mentally Ill Offenders Case formulation for risk assessment and management needs to be considered as it applies to mentally disordered offenders. For example, the structured professional judgment described by Hart and Logan in Chapter 4 is based on a rational decision theory, raising the question of whether this can be applied to mentally disordered offenders. There is a body of research investigating the role of mental disorder in the risk of future violence which can inform risk assessment with this population. For example, regarding major mental illness, a meta-analysis conducted by Bonta, Law and Hanson (1998) found that predictors of violent recidivism were the same for mentally disordered offenders as for nonmentally disordered offender. In fact, the use of an insanity defense and having a diagnosis of psychosis, were small but significant negative predictors of risk (effect sizes (Zr) −.07, −.04, and −.03 respectively). That is, the evidence was that mentally ill offenders were less likely to recidivate violently than were nonmentally ill offenders. Conclusions from this study are that risk assessments for mentally ill offenders should focus on the same issues as for non mentally ill offenders, and mentally ill offenders are likely to need the same types of intervention to reduce offending as are non mentally ill offenders. There is, however, evidence to the contrary. In a longitudinal study of the general population in the USA, Swanson (1994) reported on a group (N = 7,000) who were given a psychiatric examination and whose violent behavior was recorded. Of the entire group, 9% had been seriously violent at some time in their lives, rising to 18% when less serious violence was included. Looking at violence in the previous year, 1.2% of the population with no mental disorder had been seriously violent compared with 3.8% of those with a mental illness (schizophrenia or affective disorder). The odds of having been seriously violent in the past year were, therefore, three times higher for those with schizophrenia or a major affective disorder compared to those without a mental illness, but this should be interpreted in light of the low overall likelihood of violence. The contradiction between the findings of Bonta et al. (1998), who showed that mental illness was negatively related to recidivism, and Swanson’s (1994) research in which mental illness was a positive predictor of risk may be explained by
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
291
the differences between samples. Swanson (1994) studied the general population, some members of which had a mental illness, perhaps untreated, which placed the person at-risk of offending. By contrast, Bonta et al. (1998) studied known mentally ill offenders. In this group, the mental illness had been treated and individuals were likely closely monitored in the community. Thus, perhaps untreated mental illness increases the likelihood of violence, whereas treatment and management lowers the risk of violence. Symptoms of mental illness do, therefore, have relevance in forensic case formulation, but, where mentally ill offenders are concerned, care should be taken not to ignore other risk factors that apply more generally.
Offenders with Personality Disorders Compared with mentally ill offenders, those with personality disorders are more likely to reoffend after discharge from hospital (Bailey and MacCulloch, 1992; Jamieson and Taylor, 2004; Steels et al., 1998). In Jamieson and Taylor’s sample of 197 offender patients discharged from a secure psychiatric hospital, compared with those diagnosed as mentally ill, the odds ratio of serious reoffending for patients with personality disorders was 7.00 (95% confidence limits 2.05 – 23.93). In the UK, forensic services for people with personality disorders have been under considerable development in recent years, particularly services for offenders considered to have a “dangerous and severe personality disorder” (DSPD). Offenders meeting the criteria for DSPD have been diverted to treatments in special units in secure psychiatric hospitals and prisons. In these specialized units, there have been many developments in the assessment and treatment of offenders with personality disorders. Where case formulation is concerned, Lawrence Jones (Chapter 12) demonstrates the integration of knowledge about personality disorder and offending in case formulation. However, the treatment of DSPD offenders in mental health services has always been contentious, largely on the grounds of forcing the medical profession to adopt responsibility for the detention of high-risk offenders who may not be treatable (Tyrer, 2007). Financial cutbacks have led to the closure of the DSPD service. The new joint plan of the UK’s Ministry of Justice and Department of Health is to focus on a more comprehensive screening of high-risk offenders at entry to the criminal justice system to identify those with personality disorder. Those screened will be assessed so that high-quality formulations can be created, and clear treatment and intervention pathways based on the case formulation will be set out (Joseph and Benefield, 2010). Attention is likely to be paid to the link between personality disorder and offending in determining who receives costly mental health services and who does not. One criterion for admission to DSPD services, and perhaps one which was largely ignored, was that there must be a functional link between the personality disorder and the risk of violence. This same criterion is likely to apply in the newly configured Ministry of Justice and Department of Health collaboration. The nature of the functional link has been explored in a review by Duggan and Howard (2009).
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
292
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
They noted that the nature of the functional link has never been defined, and it is unclear whether it is intended to be covariation or causal. Their thesis is that it is entirely reasonable to interpret the functional link as meaning a causal link, since only this interpretation gives the link sufficient clinical importance. They base their judgement of causality on Haynes’s (1992) model, namely that for causality to be assumed the following relationships must be demonstrated: (1) covariation between variables; (2) temporal precedence of the causal variable; (3) exclusion of an alternative explanation of the relationship; and (4) establishing a logical connection between the variables. One problem with elevating personality disorders to causes of behavior is that personality disorders are supposed to be relatively unchanging. If that is true, it is hard to show covariation between personality disorder and risk, because where there is little or no variation in a variable, covariation is not possible. One way to circumvent this issue is to consider that although the unobservable personality disorder is invariant, the manner in which it is expressed may vary depending upon environmental triggers and the persons coping skills. It may also be that a search for deterministic causes is a misguided pursuit. There are other ways to define causality. A probabilistic approach to causality is where an antecedent increases the probability of a certain outcome. A common example is smoking and lung cancer. Smoking increases the risk of lung cancer, but it cannot be said to be a deterministic cause; some smokers do not get lung cancer. In behavioral case formulations, we may be best advised to search for variables which increase the risk of a certain outcome, and not seek variables that are both necessary and sufficient for a certain outcome (see Haynes, 1992; Haynes and O’Brien, 1990). Furthermore, as Duggan and Howard admit, personality disorder is a high-level construct; that is, personality disorder is a diagnosis that consists of several lower-level variables and use of the high-level variable may mask causal relationships at lower levels. This is a major challenge in formulation with personality disordered offenders: At what level of personality does one conduct the analysis – trait facets or super-ordinate domains?
Too Much Information! The most serious and prolific offenders are usually those with a long history of contact with criminal justice services and often also social services and mental health services. This leads to extensive case files containing a wealth of information (although, curiously, never the precise bit of information you are looking for at the time! The authors of at least one chapter in this volume noted that they were unable to complete certain aspects of the risk assessment because of lack of needed information). The challenge is how to organize the large quantity of material, and how to make sense of developmental and forensic histories. In Chapter 6, Aidan Hart, Mark Gresswell and Louise Braham describe multiple sequential functional analysis – MSFA. This scheme conveys an understanding of the development of a person’s behavior over time, showing which variable impacted in what ways across the life span. This innovative approach has been underused and underevaluated in forensic case formulation.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
293
No Criterion Behavior One issue in forensic case formulation is that the behavior that is being addressed in treatment is usually not evident. The reasons for this are obvious in that the behavior is criminal and so it is both unethical and dangerous to elicit the behavior for observation or hypothesis testing. How does the forensic professional deal with this problem? One approach is to focus upon intrapersonal variables that relate to risk. As mentioned by Kevin Howells in Chapter 5, these include cognitive predictors, for example, information processing biases, dysfunctional schemata, maladaptive beliefs, and antisocial values and underlying beliefs and values supporting aggression; affective predictors, for example anger or fear; physiological antecedents; and coping and self-regulatory skills. While these are mainstream targets for assessment and intervention, they do not cure the problem of whether they relate to offending behavior. A second approach is to look at behaviors that are apparently functionally similar to the offense and conduct an analysis and treatment of those behaviors in the hope that this analysis and treatment will also apply to the offending behavior. Two approaches attempt to do this. First, Offense Paralleling Behavior (OPB) is an emerging forensic case formulation methodology described by Lawrence Jones (Chapter 12), Aidan Hart, David Gresswell and Louise Braham (Chapter 6), and elsewhere (Daffern, Jones and Shine, 2010.) OPB attempts to identify current observable and covert behavior that may serve the same function as the index offense behavior. Using OPB, the professional then conducts a case formulation and treatment of the OPB(s). This approach assumes that by assessing and treating the proxy OPB, one may subsequently influence the offending behavior itself and thereby reduce the likelihood of reoffending. The notion of OPB overlaps to some extent with the behavioral notion of response chains (Sturmey, 2010) and is similar to the clinical application of chain analysis found in Linehan’s (1993) treatment of Borderline Personality Disorder. The assumption here is that earlier members of the response chain serve the same function as the terminal response in the chain and that intervention earlier in the chain is more effective than intervention later in the response chain. There is evidence for these assumptions in other research literatures outside of the forensic literature (Sturmey, 2009, pp. 102–12). The concept of OPB, however, is not without controversy. There is, as yet, no empirical evidence for the validity of the construct or for the value of working within this framework. These are serious issues because decisions may be made upon an individual’s supposed OPBs. Choosing a behavior that is not actually related to the offense would likely have significant implications for offenders and potential victims. Daffern, Jones and Shine (2010) address some of these issues.
It’s Not All about Risk Much of the emphasis in forensic case formulation relates to risk assessment and risk management; however, there is an emerging view that a focus on an offender’s
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
294
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
assets and positive life goals is just as valid in reducing risk. ‘Positive psychology’ focuses on the study of people’s strengths rather than their deficits (Seligman and Csikszentmihalyi, 2000), and, in line with this, positive approaches to rehabilitation are emerging. Ward and colleagues (Ward, 2002; Ward and Brown, 2004; Ward and Stewart, 2003) have developed the Good Lives Model (GLM) of offender rehabilitation, which focuses on helping offenders obtain satisfaction in a range of life areas rather than focusing purely on reducing risk. The underlying principle is that offenders seek satisfaction in life in the same domains as us all, but do so in problematic or distorted ways. There is evidence that offenders do aim for prosocial goals, such as wanting a better lifestyle, gaining work experience, having good family relationships, gaining skills, and getting fit and healthy (McMurran et al., 2008). It follows from the GLM that the offender should be assisted to develop the skills and resources, building on existing preferences, strengths, and opportunities, to equip him or her to live a more fulfilling and offense-free life. There is controversy over whether a focus solely on developing positive outcomes would actually reduce risk, and the GLM has been criticized as lacking empirical evaluation (Bonta and Andrews, 2003); however, it is likely that having positive things in life, such as a partner, a job, and a place to live, provide the offender with an incentive to stay crime-free. Also, a strength-focused approach capitalizes on what the offender is already good at, as well as getting away from an emphasis on depressing avoidance goals. The challenge in forensic case formulation is to identify the offender’s strengths and protective factors, as well as risk factors.
FUTURE DIRECTIONS The preceding sections raise many challenges for forensic case formulation. The final part of this chapter identifies three directions for future research: the need for basic data in forensic case formulation; the issue of monotheoretical versus eclectic case formulation; and professional training in case formulation generally and forensic case formulation in particular, including the process of how cases are formulated.
Data and Forensic Case Formulation As noted earlier, the first three chapters of this volume record the limited quantity and quality of case formulation research generally. It is worth noting that in the subsequent chapters on forensic case formulation not one author cited evidence specifically on the reliability and validity of forensic case formulation or on training professionals to conduct forensic case formulations. This failure to address these basic questions is a serious deficiency in the literature. Forensic case formulation is likely to be more challenging than formulating regular clinical cases due to the large quantity of material, problems in obtaining all the relevant history, problems
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
295
in the veracity of much of the information from all parties, and the presence of multiple problems in the individual concerned (Sturmey, 2010). Clearly an important area for research in forensic case formulation is some fundamental research on some relatively simple questions. Researchers need to determine what constitutes adequate forensic formulations, show that clinicians can write formulations that meet these criteria to an adequate extent, and demonstrate the reliability and validity of forensic case formulations. This is an unexplored field and at this time any research into these topics, which could readily take and improve upon existing research on case formulation, will make a contribution to an unexplored field.
Eclectic or Monotheory Formulations As this and other volumes show (Eells, 2007; Sturmey, 2009a) case formulations are made from very different perspectives. Most volumes on case formulation do so from a cognitive behavioral perspective (Kuyken et al., 2009; Persons, 2008) as well as psychodynamic approaches (Horowitz, 1997; McWilliams, 1999); fewer volumes have used behavioral approaches (Sturmey, 1996, 2007, 2008; Turkat, 1985) or eclectic approaches (Weerasekera, 1996) to case formulation. How do we reconcile and integrate formulations on the same person that come from different perspectives? Indeed, should we even try to reconcile different approaches? As Eells and Lombart noted, one answer to this question is not to try to integrate different approaches to psychotherapy. A variant on this approach is to extend the ideas in one theory to try to explain phenomena or treatments from another theory’s approach. Two examples can illustrate this approach. Skinner (1953) suggested that behavioral mechanisms might explain what happens in psychotherapy and how effective psychotherapy might come about. He speculated that the environment of classic psychotherapy perhaps reinforced high rates of clients’ talking since it was nonpunitive, and other therapist behavior, such as minimal verbal contribution, might reinforce clients’ talking. Consequently, clients may eventually talk a lot in therapy and may, indeed, say things to their therapists that they never say to anyone else. In so doing, they are exposed to their own emotionally laden words repeatedly and, if those words are conditioned stimuli, the emotional catharsis that may occur in psychotherapy may be respondent extinction of conditioned responses to those conditioned stimuli. A different example of the same approach comes from behavioral explanations of the effectiveness of cognitive-behavior therapy for depression (Sturmey, 2009b). Cognitive-behavior therapy has two distinct components: cognitive interventions, such as Socratic questioning and cognitive restructuring; and behavioral activation in which the client engages in progressively more activities related to positive mood. Comparisons of behavioral activation alone with the package of cognitive and behavioral interventions have produced similar effect sizes. Further, comparisons of behavioral activation with cognitive therapy alone have found behavioral activation alone to be superior to cognitive therapy (Sturmey, 2009b). Thus, one
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
296
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
may conclude that cognitive-behavior therapy works through behavioral activation, but not cognitive therapy. Some authors present the case for integrated approaches in a number of ways. As mentioned earlier, Eells and Lombart (2004) argued that formulating a case from more than one theoretical perspective was a positive thing since “each theory, source of information and formulation model has something to offer” (p. 24). But, note here that Eells and Lombart are suggesting multiple, independent formulations, rather than integrating formulations. Others have attempted to integrate different approaches. For example, Weerasekera (1996) developed a methodology to formulate the case many times from biological, psychodynamic, psychiatric, cognitive and behavioral approaches and then to integrate these alternative formulations into one single integrated formulation and to select multiple different kinds of interventions during treatment. Although this is an explicit and interesting approach to integrative case formulation, there is currently no research on this approach. An alternative approach to psychotherapy integration is to attempt to identify apparently disparate concepts from different approaches to psychotherapy and to claim that there commonalities exist (Norcross and Goldfried, 2005; Wachtel, 1977). This approach remains controversial.
Professional Training in Case Formulation There is evidence that not all clinicians are competent as conducting case formulations. For example, Kuyken et al. (2005) evaluated the case formulation skills of 115 clinicians who participated in a continuing education workshop on case formulation. They found that, although there was generally good agreement in identifying the target behavior accurately, there was much less agreement about identifying aspects of the case that required theory-driven inference. Indeed, the quality of case formulations ranged from “very poor” to “good”, but the authors rated only 44% of case formulations as “at least good enough.” Dudley et al. (2010) evaluated the accuracy of formulations in a group of 85 mental health professionals, including psychiatric nurses (46%), doctoral clinical psychology students (22%). They asked the participants to make a formulation of a case of psychosis based on a 30 minute video vignette and additional clinical material. They compared the participants’ formulations to a benchmark formulation of the case. Like Kuyken et al. (2005), Dudley et al. found widely varying levels of accuracy in case formulation. For example, there was more than 80% accuracy in identifying problem behaviors such as physical attacks, avoidance, and drug use, but the accuracy of identifying problematic thoughts was only 32% for the entire sample. Although there was high agreement in identifying anxious and paranoid feelings (84%), accuracy in identifying shame as a significant problem was poor (9%). Likewise, identifying early experiences, core beliefs, schemas and stressors was generally poor (55%, 50% and 64%), although these data were very variable with some items being identified accurately by most participants but many being identified by few. This study was interesting because it only required participants to write their formulation using a prepared template that prompted them to respond to cues as to the elements of a formulation, but did not require them to
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
297
write a treatment plan based on that formulation; hence, one key professional performance – writing a formulation-based treatment plan – was omitted from the study. Despite the need for empirically supported approaches to teaching case formulation skills, there are few studies on this topic perhaps. Kendjelic and Eells (2007) evaluated a two-hour training program in case formulation delivered to a group of 20 clinicians. They found that, compared to a control group, after training, the clinicians’ formulations were more likely to be of better quality, be more elaborated, comprehensive and concise, and more likely to identify the mechanisms behind the presenting problems. Clearly, more research is needed to develop effective training in case formulation skills, not only for forensic cases but for all clinical cases. Such a program of research would be effortful and require a number of explicit decisions which might be uncomfortable for researchers. For example, such a research program would require a reliable and valid measure of the quality of a formulation. As the previous studies show, it is possible to measure the quality of case formulations reliably using trained raters on Likert scales of formulation quality (Kendjelic and Eells, 2007). Further, Dudley et al. (2010) reported kappas of “greater than .85” for the total score of correct elements in a formulation, but did not report kappas for individual items. To date, however, we are unaware of any measures of the appropriateness of treatment plans based on a formulation. Although measures such as that designed by Kendjelic and Eells (2007) can be applied to formulations using different theoretical frameworks, research into case formulation skills might also require theory-specific measures. Although there is some evidence of agreement about what constitutes an adequate formulation, contradictory evidence also exists. For example, Flitcroft et al. (2007) conducted a Q-sort methodology study of the features of cognitive-behavioral case formulations of depression that clinicians found to be most important. They found that there were important differences between the raters as to what was rated as important in a case formulation. For example, some raters found the content features of a formulation, such as how the problem is maintained, matching thoughts and emotions, to be most important. In contrast, others found features related to using the case formulation for treatment, such as how to intervene and the acceptability of the formulation to the client, to be most important. A legitimate couple of questions to ask are whether we need specific training for all forensic staff in clinical case formulation and do we need additional specific extensive training for psychologists, psychiatrists and other professionals in this area? Of the two papers on training cited above, Kendjelic and Eells (2007) provided training only for advanced professional, whereas Dudley et al. (2010) included a wide range of staff. The training needs of staff are likely to vary considerably. Most staff working directly with offenders may need some basic knowledge of the concepts behind formulation, and may benefit from training on data collection to assist functional assessment and treatment evaluation. They may also require skills to implement treatment plans, decide when to call for additional assistance from professionals, and decide when to draw professional’s attention to some significant event. Professional staff may have rather different training needs. Professional staff who are not directly involved in case formulation and treatment planning may also
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
298
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
benefit from some general training on case formulation so that they can understand the rationale for this approach and give relevant input into the assessment and treatment process. On the other hand, for example, psychology and other staff who conduct assessments to complete a case formulation and to write, implement and evaluate an intervention plan may need more extensive training in the clinical skills related to these functions. One observation on the limited training literature on case formulation is that it has not directly addressed the issue of generalization of clinicians’ formulation skills. Typically, these training papers have only evaluated participants’ skills in writing a case formulation for only one vignette. Such an approach may include direct coaching during the training on that specific vignette; however, the clinician has a much more challenging task than writing one formulation, perhaps with some coaching. Rather, clinicians must respond to novel clinical material, including novel diagnoses, combinations of diagnoses, problematic behaviors, functions and interventions. One approach to conceptualizing and designing training is to use general case training (Sprague and Horner, 1984) which identified all the dimensions of the training stimuli and responses, such as the range of common diagnoses, presence or absence of drug or alcohol abuse, and main presenting topography. A generalization matrix is then constructed and training exemplars are strategically selected so that they sample all the relevant features of training. Once training has been completed on these strategically selected examples which sample all the relevant dimensions, untrained exemplars can be presented so that generalization of clinicians’ skills to novel untrained exemplars can be assessed. Table 13.1 presents a partial analysis of such training which samples diagnoses, presence or absence of substance/alcohol abuse and presenting topography. Note, however, that this analysis is incomplete since it does not sample the presenting problem functions of the full range of common treatment strategies. Determining all the relevant dimensions and constructing a complete generalization matrix that sampled all those dimensions would require extensive analysis beyond the scope of this chapter.
The Process of Case Formulation and Professional Training Often when case formulations are presented they are presented as apparently finished products and so the process by which the therapist arrived at the final formulation is obscure. Clinicians begin case formulations before they see their client. Their knowledge base, theoretical orientation, and their own learning history all influence their decision about what is a legitimate target for change, the variables that might influence the target that they will assess, and the relative weight given to different variables. Sometimes, these preconceptions about case formulation may be explicit and at other times they may be implicit. Clinicians may begin to make a formulation upon first referral or upon first meeting their client. For example, Wolpe described how he used his first impression of a client – her physical appearance, dress and her answers to the first couple of questions – to make an initial formulation about her problem and how he used these initial hypotheses about her problem to guide subsequent assessment
Pedophilia Test for generalization
Test for generalization
Train
Test for generalization
Test for generalization
Personality Disorder
Test for generalization
Train
Test for generalization
Train
Test for generalization
Test for generalization
No substance/ Substance/ alcohol abuse alcohol abuse
Personality Disorder
Psychosis
Test for generalization
Test for generalization
Train
Test for generalization
Train
Test for generalization
No substance/ Substance/ alcohol abuse alcohol abuse
Psychosis
8:33
Test for generalization
Train
Test for generalization
Male to female rape
Train
Test for generalization
Test for generalization
Train
Anxiety
Violence
Anxiety No substance/ Substance/ alcohol abuse alcohol abuse
Depression
May 28, 2011
No substance/ Substance/ alcohol abuse alcohol abuse
Depression
Table 13.1 An example of how general case training can be used to design and evaluate training in case formulation.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran Printer Name: Yet to Come
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
300
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
(Wolpe and Turkat, 1985). But a formulation may also develop over time during the course of an initial and subsequent assessment interviews. Another example is Kinderman and Lobban’s (2000) approach to working with a client to develop a progressively more differentiated case formulation. Initially they presented a very simple template of a clinical problem to their client: “Vulnerability + stress = distress disorder”. They used this template to generate their clients’ vulnerabilities and stressors and developed two progressively more differentiated formulations of their clients’ auditory hallucinations and other problems. Thus, in this approach the development of the formulation is both explicit and done collaboratively with the client. Finally, there are examples of reformulating cases after new information comes to light or when something unexpected occurs, such as a relapse or set back in therapeutic progress. Persons (2008, pp. 225–9), for example, noted that treatment failure may reflect an inaccurate or incomplete case formulation, among other things. She described a case in which the initial assessment did not precisely identify the client’s central fear. The initial formulation suggested that the client, who had Post Traumatic Stress Disorder following an assault by a homeless person, was uncomfortable being in a public place in which she might encounter a homeless person. Treatment based on this formulation was ineffective. Only when Persons carefully interviewed her again did the client eventually verbalize that her core fear was of being knocked down, rather than being around a homeless person. Then Persons could reformulate the case and begin treatment based on exposure to being bumped into by other people. Persons noted at least three kinds of problems with an initial formulation that might result in treatment failure. The first problem was selecting a target behavior that is not functionally relevant. An example was an initial treatment failure with a student having problem writing her dissertation. Treatment was initially ineffective when the target behavior was increasing time studying, but treatment was effective when the target was changed to increasing the number of pages written. A second example of having to reformulate was inadvertently omitting important variables in the initial formulation. An example of this was not knowing that a client with a sleep problem incorrectly took his sleep medications when he woke up at 4AM, rather than when he went to bed and that this lead to day time drowsiness. Thus, the initial case formulation omitted an important variable and this led to initial treatment failure. A third example of failure in initial formulation that led to a reformulation was incorrectly identifying the function of the main presenting problem. In this example, a client with poor coping with daily stressors often reported that when a challenge occurred she thought “I can’t do it”. Cognitive therapy failed to address this problem effectively. Persons reformulated the problem, not as a deficit in attribution, but as a target behavior negatively reinforced by avoidance of challenges. Thus, Persons changed the intervention to teach the client to break down daunting tasks into component parts and complete each part one at a time and this strategy based on the reformulation was an effective one. These examples show that case formulation does not involve the delivery of a finished product. Rather, case formulations grow and change over time. In future, researchers should give more attention to the process of case formulation rather than the final formulation itself and provide clinicians realistic models of how to develop a case formulation over time.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
301
CONCLUSION This text has provided a baseline of forensic case formulation practice and research, along with ideas and suggestions for developing the field empirically. Much can be learned from non forensic work in clinical case formulation, but there are still particular aspects in forensic work that need attention. It is our hope that the ideas presented in the excellent contributions to this book by both non forensic and forensic practitioners will stimulate readers to work to improve professional practice in this important area.
REFERENCES Andrews, D.A., Bonta, J. and Hoge, R.D. (1990) Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, 19–52. Bailey, J. and MacCulloch, M. (1992) Characteristics of 112 cases discharged directly to the community from a new special hospital and some comparisons of performance. Journal of Forensic Psychiatry, 3, 91–112. Bonta, J. and Andrews, D.A. (2003) A commentary on Ward and Stewart’s model of human needs. Psychology, Crime, and Law, 9, 215–18. Bonta, J., Law, M. and Hanson, K. (1998) The prediction of criminal and violent recidivism among mentally disordered offenders: A meta-analysis. Psychological Bulletin, 123, 123–42. Bonta, J. and Wormith, S. (2008) Risk and need assessment. In G. McIvor and P. Raynor (eds), Developments in Social Work with Offenders (pp. 131–52) London: Jessica Kingsley. Chiesa, M. (1994) Radical Behaviorism: The Philosophy and Science. Boston: The Authors Cooperative. Cipani, E. and Golden, J.A. (2007) Differentiating behavioral and traditional case formulations for children with severe behavioral and emotional problems. International Journal of Behavioral Consultation and Therapy, 3, 537–45. Cooke, D.J. and Michie, C. (2010) Limitations of diagnostic precision and predictive utility in the individual case: A challenge for forensic practice. Law and Human Behavior, 34, 259–74. Daffern, M. and Howells, K. (2002) Psychiatric inpatient aggression: A review of structural and functional assessment approaches. Aggression and Violent Behavior, 7, 477–97. Daffern, M. and Howells, K. (2007) Antecedents for aggression and the function analytic approach to the assessment of aggression and violence in personality disordered patients within secure settings. Personality and Mental Health, 1, 126–37. Daffern, M. and Howells, K. (2009) What’s the point? Towards a methodology for assessing the function of psychiatric inpatient aggression. Behaviour, Research and Therapy, 45, 101–11. Daffern, M., Jones, L. and Shine, J. (2010) Offence Paralleling Behaviour: A Case Formulation Approach to Offender Assessment and Intervention. Chichester: Wiley-Blackwell. Dudley, R., Park, I., James, I. and Dodgson, G. (2010) Rate of agreement between clinicians on the content of a cognitive formulation of delusional beliefs: The effect of quantifications and experience. Behavioural and Cognitive Therapy, 38, 185–200. Duggan, C. and Howard, R. (2009) The functional link between personality disorder and violence: A critical appraisal. In M. McMurran and R. Howard (eds), Personality, Personality Disorder, and Violence (pp. 19–37). Chichester: Wiley-Blackwell. Eells, T.D. (ed.) (2007) Handbook of Psychotherapy Case Formulation ( 2nd edn). New York: Guilford.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
302
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Eells, T.D. and Lombart, K.G. (2004) Case formulation: Determining the focus in brief dynamic psychotherapy. In D.P. Charman (ed.), Core Processes in Brief Psychodynamic Psychotherapy: Advancing Effective Practice (pp. 119–43). Mahwah, NJ: Lawrence Erlbaum. Emmelkamp, P.M.G., Bouman, E.K. and Blaauw, E. (1994) Individualized versus standardized therapy: A comparative evaluation with obsessive-compulsive patients. Clinical Psychology and Psychotherapy, 1, 95–100. Flitcroft, A., James, I. A., Freeston, M. and Wood-Mitchell, A. (2007) Determining what is important in a good formulation. Behavioural and Cognitive Therapy, 35, 325–34. Ghaderi, A. (2006) Does individualization matter? A randomized trial of standardized (focused) versus individualized (broad) cognitive behavior therapy for bulimia nervosa. Behaviour Research and Therapy, 44, 273–88. Gresswell, D.M. and Hollin, C.R. (1992) Toward a new methodology of making sense of case material. Clinical Behaviour and Mental Health, 2, 329–41. Hay, P.P., Bacaltchuk, J., Stefano, S. and, Kashyap, P. (2009) Psychological treatments for bulimia nervosa and binging. Cochrane Database Systematic Review, Oct, 7(4):CD000562. Haynes, S.N. (1992) Models of Causality in Psychopathology. New York: Macmillan. Haynes, S.N. and O’ Brien, W.H. (1990) Principles and Practice of Behavioral Assessment. New York: Kluwer. Horowitz, M.J. (1997) Formulation as a Basis for Planning Psychotherapy Treatment. Washington, DC: American Psychiatric Press. Jacobson, N.S., Follette, V.M., Follette, W.C. et al. (1989) A component analysis of behavioral marital therapy: 1-year follow-up. Behaviour, Research and Therapy, 23, 549–55. Joseph, N. and Benefield, N. (2010) The development of an offender personality disorder strategy. Mental Health Review Journal, 15, 10–15. Kearney, C.A. and Silverman, W.K. (1990) A preliminary analysis of a functional model of assessment and treatment for school refusal behavior. Behavior Modification, 14, 340–66. Jamieson, L. and Taylor, P.J. (2004) A reconviction study of special (high security) hospital patients. British Journal of Criminology, 44, 783–802. Kendjelic, E.M. and Eells, T.D. (2007) Generic psychotherapy case formulation training improves formulation quality. Psychotherapy: Theory/Research/Practice/Training, 44, 66–77. Kinderman, P. and Lobban, F. (2000) Evolving formulations: Sharing complex information with clients. Behavioural and Cognitive Therapy, 28, 307–10. Kuyken, W. (2006) Evidence-based case formulation: Is the emperor clothed? In: N. Tarrier (ed.), Case Formulation in Cognitive Behaviour Therapy, (pp. 12–35.) Hove: BrunnerRoutledge. Kuyken, W. Fothergill, C.D., Musa, M. and Chadwick, P. (2005) The reliability and quality of cognitive case formulation. Behaviour, Research and Therapy, 43, 1187–1201. Kuyken, W., Padesky, C.A. and Dudley, R. (2009) Collaborative Case Conceptualization. Working Effectively with Clients in Cognitive-Behavioral Therapy. New York: Guilford. Linehan, M. (1993) Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York: Guilford. McMurran, M., Theodosi, E., Sweeney, A., and Sellen, J. (2008) What do prisoners want? Current concerns of adult male prisoners. Psychology, Crime and Law, 14, 267–74. McNight, D.L., Nelson, R.O., Haynes, S.C. and Jarrett, R.B. (1984) The importance of treating individually asessment response classes in the amelioration of depression. Behavior Therapy, 15, 315–35. McWilliams, N. (1999) Psychoanalytic Case Formulation. New York: Guilford. Mumma, G.H. (2001) Increasing accuracy in clinical decision making: Towards an integration of nomothetic-aggregate and intraindividual-idiographic approaches. The Behavior Therapist, 24, 77–94. Mumma, G.H. (2004) Validation of idiosyncratic cognitive schema in cognitive case formulations: An intra-individual idiographic approach. Psychological Assessment, 16, 211–30. Mumma, G.H. and Mooney, S.R. (2007a) Comparing the validity of alternative cognitive case formulations: A latent variable, multivariate time series approach. Cognitive Therapy and Research, 31, 451–81.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION: EMERGING ISSUES
303
Mumma, G.H. and Mooney, S.R. (2007b) Incremental validity of cognitions in a clinical case formulation: An Intraindividual test in a case example. Journal of Psychopathology and Behavioral Assessment, 29, 17–28. Mumma, G.H. and Smith, J.L. (2001) Cognitive-behavioral interpersonal scenarios: Interformulator reliability and convergent validity. Journal of Psychopathology and Behavioral Assessment, 23, 203–21. Norcross, J.C. and Goldfried, M.R. (eds) (2005) Handbook of Psychotherapy Integration ( 2nd edn). New York: Oxford. Persons, J.B. (2008) The Case Formulation Approach to Cognitive-Behavior Therapy. New York: Guilford. ¨ Schulte, D., Kunzel, R., Pepping, G. and Schulte-Bahrenberg, T. (1992) Tailor-made versus standardized therapy of phobic patients. Advances in Behaviour Research and Therapy, 14, 67–92. Seligman, M.E.P. and Csikszentmihalyi, M. (2000) Positive psychology: An introduction. American Psychologist, 55, 1–5. Singh, J.P. and Fazel, S. (2010) Forensic risk assessment: A meta-review. Criminal Justice and Behavior, 37, 965–88. Skinner, B.F. (1953) Science and Human Behavior. New York: Macmillan. Sprague, J.R. and Horner, R.H. (1984) The effects of single instance, multiple instance, and general case training on generalized vending machine use by moderately and severely handicapped students. Journal of Applied Behavior Analysis, 17, 273–8. Steels, M., Roney, G., Larkin, E. et al. (1998) Discharged from special hospital under restrictions: A comparison of the fates of psychopaths and the mentally ill. Criminal Behaviour and Mental Health, 8, 39–55. Strauman, T.J., Vieth, A.Z., Merrill, K.A. et al. (2006) Self-system therapy as an intervention for self-regulatory dysfunction in depression: a randomized comparison with cognitive therapy. Journal of Consulting and Clinical Psychology, 74, 367–76. Sturmey, P. (1996) Functional Analysis and Clinical Psychology. Chichester: John Wiley & Sons, Ltd. Sturmey, P. (ed.) (2007) Functional Analysis in Clinical Treatment. New York: Academic Press. Sturmey, P. (2008) Behavioral Case Formulation and Intervention. A Functional Analytic Approach. Chichester: Wiley-Blackwell. Sturmey, P. (ed.) (2009a) Varieties of Case Formulation. Chichester: John Wiley & Sons, Ltd. Sturmey, P. (2009b) Behavioral activation is an evidence-based treatment of depression. Behavior Modification, 33, 818–29. Sturmey, P. (2009c) Case formulation: A review and overview of this volume. In: P. Sturmey (ed.), Varieties of Case Formulation (pp. 3–30). Chichester: John Wiley & Sons, Ltd. Sturmey, P. (2010) Case formulation in forensic psychology. In M. Daffern, L. Jones and J. Shine (eds), Offence Paralleling Behaviour: An Individualised Approach to Offender Assessment and Treatment (pp. 25–52). Chichester: John Wiley & Sons, Ltd. Swanson, J.W. (1994) Mental disorder, substance abuse, and community violence: An epidemiological approach. In J. Monahan and H.J. Steadman (eds), Violence and Mental Disorder. Chicago: University of Chicago Press. Turkat, I.D. (ed.) (1985) Behavioral Case Formulation. New York, Plenum. Tyrer, P. (2007) An agitation of contrary opinions. British Journal of Psychiatry, 190, S49, S1–S2. Wachtel, P.L. (1977) Psychoanalysis, Behavior Therapy, and the Relational World. Washington: American Psychological Association. Ward, T. (2002) Good lives and the rehabilitation of offenders: Promises and problems. Aggression and Violent Behaviour, 7, 513–28. Ward, T. and Brown, M. (2004) The Good Lives Model and conceptual issues in offender rehabilitation. Psychology, Crime and Law, 10, 243–57. Ward, T. and Hudson, S.M. (1998) A model of the relapse process in sexual offenders. Journal of Interpersonal Violence, 13, 700–25. Ward, T. and Stewart, C.A. (2003) Criminogenic needs and human needs: A theoretical model. Psychology, Crime, and Law, 9, 125–43.
P1: OTA/XYZ P2: ABC JWST073-13 JWST073-McMurran
304
May 28, 2011
8:33
Printer Name: Yet to Come
FORENSIC CASE FORMULATION
Weerasekera, P. (1996) Multiperspective Case Formulation: A Step toward Treatment Integration. Malabar, Florida: Krieger. Wolpe, J. (1986) Individualization: The categorical imperative of behavior therapy practice. Journal of Behavior Therapy and Experimental Psychiatry, 17, 145–53. Wolpe, J. and Turkat, I.D. (1985) Behavioral formulation of clinical cases. In I.D. Turkat, Behavioral Case Formulation (pp. 5–36). New York: Plenum.
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX Note: Page numbers with italicized n’s refer to footnotes. abandonment, 228, 230, 259, 268, 269 threat of, 186, 259 ABC, see Antecedent-BehaviorConsequence analyses aberrant behavior, 74 ACF, see Assessment and Classification of Function actuarial measures, 176–9, 289 ACUTE-2007, 177, 187, 202 acute dynamic risk factors, 177 adaptive extinction, 13 adaptive shaping, 12 Addiction Severity Index (ASI), 156 adolescent development, 136–8 further experiences, 139–42 adolescents, forensic case formulation with, 217–33 adoption, 259, 267 Adult Attachment Interview, 263 adult victims, female perpetrators with, 199, 200 adulthood and development of offense pathway, 142–5 adults, sexual offences, 195–210 affective neuroscience, 23 age of onset, 18 agency and communion, 263 aggravated sexual assault, 203, 204, 206, 207 aggression, 33, 35, 36, 46, 52, 73, 135, 136, 138, 142, 148, 154, 167–9, 186, 197, 205, 209, 239, 244, 245, 286, 293 angry affect in formulation of, 112, 113 cognitive behavioral (CB) approaches to, 107–23
mental disorder and, 115, see also violence and aggression Aggressive Sexual Behavior Inventory (ASA), 203 agoraphobia, 69 agreeableness, 260 low, 259 alcohol abuse/dependence, 12, 18, 100–102, 117, 122, 130, 139, 140, 142–7, 149, 150, 154–9, 163–9, 180, 181, 183, 186, 187, 204, 205, 209, 239, 240, 242, 244–8, 251, 271, 298, 299 balance sheet, 160 functional analysis, 164 Alcohol Use Disorder Identification Test (AUDIT), 156 adapted to include drug use (AUDIT-ID), 156 alienation, 229, 231 American Psychological Association (APA), 5 anal stage, 8 anamnestic risk assessment, 86 anger, 35, 36, 40, 46, 48, 50, 52, 108, 109, 111–15, 117–20, 122, 131, 136, 145, 146, 150, 187, 197, 198, 205, 207, 208, 267, 287, 293 handling, 167 and substance abuse disorders, 153–69 animal research, 16 Antecedent-Behavior-Consequence (ABC) analyses, 133, 134, 287 antecedent/precipitating factors, 249 antecedent stimuli classes, 132–3 antisocial disorders, 107
Forensic Case Formulation, First Edition. Edited by Peter Sturmey and Mary McMurran. C 2011 John Wiley & Sons, Ltd. Published 2011 by John Wiley & Sons, Ltd.
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
306
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
antisocial personality disorder, 107–8 anxiety, 8, 13, 14, 16, 18, 34, 40, 46, 49, 50, 65, 69, 70, 95, 108, 109, 135, 137, 139, 159, 163, 185, 186, 224, 240, 250, 260, 262, 266, 290 disorders, 14, 67, 68, 73, 284 APA, see American Psychological Association APA Task Force on Evidence-Based Practice in Psychology, 7, 16 approach automatic offenders, 242 approach explicit offenders, 242 approach- or theory-specific aspects of validity, 42–3 assault, violent, 259 assertiveness, 150, 168, 260 assessment, 225, 265–73 tools, 202–3 Assessment and Classification of Function (ACF), 119, 121, 122, 286 attachment insecure, 264 and relationship schema, 259 social connection, and moral behavior, 224, 225 style, two-factor model of, 264 theory, 264 attention-seeking behavior, 12 AUDIT, see Alcohol Use Disorder Identification Test AUDIT-ID, see Alcohol Use Disorder Identification Test: adapted to include drug use avoidance activity, 13 avoiding cues versus learning to cope differently, 166 Axis I and Axis II disorders, 107, 169 assessment of, 157, 158 BAS, see Behavioral Activating System Basic Behavioral Repertoire (BBR) model of personality, 263 BBR, see Basic Behavioral Repertoire BDI, see Beck Depression Inventory Beck, A.T., 10, 11 Beck, J.S., 21 Beck Depression Inventory (BDI), 66 behavior, no criterion, 293 behavioral case formulation and cognitive-behavioral case formulations, 53, 54 predictive validity in, 46 Behavioral Activating System (BAS), 260 Behavioral Inhibition System (BIS), 260
behavioral problems, 244 behavioral theories, 11–14 operant conditioning, 12, 13 respondent conditioning, 13, 14 behaviorism, 10, 11 binge-drinking, 18 binge-eating, 18 biological research and sociological theories, 237, 238 BIS, see Behavioral Inhibition System Blackburn’s higher order personality dimensions (coercive and withdrawn), 250 borderline personality disorder, 18, 50, 74, 155, 262, 269, 293 British Psychological Society, 5 bulimia nervosa, case formulation for treatment of, 71, 72 bullying, 135–7, 148, 150 burglary, 137, 139, 196, 197, 205, 206 cannabis, 154, 158, 159, 162, 183, 184, 186, 187 case conceptualization diagram, 38, 54 case formulation, see forensic case formulation causal modeling framework (Morton), 258, 259 CBCFs, see cognitive-behavioral case formulations CBT, see cognitive behavioral therapy CCRT, see Core Conflictual Relationship Theme CF, see forensic case formulation; clinical case formulation chaining, 13 change, increasing motivation for, 159–61 characteristic adaptations, 258 Chart of Interpersonal Reactions in Closed Living Environments, 265 child abuse, 136 child guidance services, 244 child sexual abuse, 240, see also children and adolescents: sexual offenses against childhood and adolescence behavioral disorders, case formulation in treatment of, 73, 74 experiences, 20, 134–6 operation of multiple risk domains during, 222, 223 children and adolescents forensic case formulation with, 217–33 assessment, 225–7 case study, 227–31
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX literature review, 217–25 treatment planning, 231, 232 sexual offenses against, 175–191 clinical case formulation (CF), 33 validity, 41–52 clinical hypothesis testing, 4 clinical pathogenesis map, 43 clinically relevant institutional behaviors, 147 cocaine, 156, 158–66, 168, 205, 206, 208, 209 balance sheet, 161, 162 functional analysis, 165 coercive (higher order personality dimension), 250 cognition, 11, 17, 46, 63–5, 76, 85, 114, 137, 138, 166, 168, 169, 187, 233, 243, 248, 250, 252, 253, 259, 286 cognitive-behavioral case formulations (CBCFs), 36, 37 and behavioral case formulation, 53, 54 comments and issues, 36, 37 predictive validity in, 46 cognitive-behavioral therapy (CBT), 4, 6 problem-solving perspective, 23 cognitive case formulation method, 21 cognitive distortions, 11, 62, 180, 184, 185, 205, 207, 209, 239–41, 249, 252, 270 cognitive-dynamic model (Ryle), 6 cognitive interventions, 295 cognitive theories, 10–11 cognitive therapy (CT), 21, 63–6, 71, 73, 286, 295, 296, 300 cognitive triad, 10, 11 collaborative cognitive case conceptualization, 22 communion and agency, 263 community risk domain, 222 comorbid Axis I and Axis II disorders, assessment of, 157–8 compassion, poor, 259 concurrent validity, 43 conditional response (CR), 13, 14 conditional stimulus (CS), 13, 14, 132 Configurational Analysis, 36, 47 confirmatory dynamic factor analysis, 47 confusion, 206, 270 conscientiousness, 260 construct explication and content validity in case formulation, 34 construct validity discriminant validity, 49–52 general issues and convergent validity, 47–9 in case formulation, 51, 52
307
content validity, 34–7 in cognitive-behavioral and behavioral case formulations, 36 and construct explication in case formulation, 34 in psychodynamic case formulation, 35, 36 control mastery theory and plan formulation method of case formulation, 20, 21 convergent validity, 47–9 across different types of formulations, 49 interformulator, 49 on item level, 48 on measure level, 48 coping avoiding cues versus learning to cope differently, 166 measures to assess, 157 skills, 23 handling anger, 167 handling craving, 166, 167 core beliefs, 7, 11, 19, 21, 38, 179, 296 Core Conflictual Relationship Theme (CCRT), 19, 20 Costa and McCrae’s five personality factors, 260 see also Five Factor Model counterconditioning, 14 counterfeit deviance hypothesis, 241 CR, see conditional response craving, handling, 166, 167 crime, 26, 139, 154, 161, 197, 218, 222, 224, 238, 262, 267, 294 juvenile, 221 sexual, 89, 198, 199, 202, 249 violent, 107, 108, 115, 198 criminal behavior, 93–5, 138, 143, 144, 150, 155, 196, 208, 218–21, 223, 224, 232, 233, 237, 238, 248, 250, 287 and substance abuse disorders, 153, 154 criminal justice system, 291 criterion validity, 43 using experts’ case formulations to evaluate training in case formulation skills, 54 CS, see conditional stimulus culture, 5, 246 Curtis, J.T., 20 cyclical maladaptive pattern, 47 Daily Behavior Rating Scale (DBRS), 266 dangerous and severe personality disorder (DSPD), 291 DAST, see Drug Abuse Screening Test (DAST)
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
308
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
data and forensic case formulation, 294, 295 DBRS, see Daily Behavior Rating Scale defense supportive beliefs, 259 delinquency, 238 of siblings and peers, 222 dependence and impaired control, measuring, 157 depressed mood, 5, 71, 159, 163, 168, 209 depression, 5, 9, 13, 17, 18, 23, 24, 34–6, 41, 42, 46, 47, 49, 50, 70, 108, 109, 113, 117, 154, 182, 209, 228, 240, 251, 260, 261, 285, 295, 297, 299 case formulation for treatment of, 63–8 depressive disorder, 5, 9 deprivation states, 12 desensitization, 14, 65 developmental history, 183 developmentally informed formulation, 229–31 deviant sexuality, 182, 184, 185, 187, 191, 200, 201, 204, 208, 241 Diagnostic and Statistical Manual of Mental Disorders, 4th edn (DSM-IV), 5, 66, 155 discretionary procedures evidence-based violence risk assessment, 86–8 discriminant validity, 49–52 aggregate level: disorder-related, 50 changes in case formulation associated with treatment response, 50, 51 item level, 49, 50 within-patient, generalizability across facets, 51 discriminative stimulus (SD), 132 disease, 18, 19 distress, 4, 5, 17, 24, 35, 40, 43, 46, 50, 53, 55, 69, 95, 100, 114, 122, 140, 163, 182, 185, 224, 230, 231, 259, 267–9, 272, 287, 300 tolerance, 231 domestic violence, 158, 228 dominance and control reinforcing, 259 dream accounts, 7 Drinking Motives Measure, 157 drive/structural theory, 8 drug abuse/misuse, 100, 137, 154–9, 162, 166–9, 180, 181, 183, 191, 204, 227, 229, 230, 250, 262, 271, 296, 298 Drug Abuse Screening Test (DAST), 156 Drug-Taking Confidence Questionnaire (DTCQ), 157, 169 Drug Use Screening Inventory (DUSI-R), 156
DSM-IV, see Diagnostic and Statistical Manual of Mental Disorders, 4th edn DSPD, see dangerous and severe personality disorder DTCQ, see Drug-Taking Confidence Questionnaire Dudley, R., 22 DUSI-R, see Drug Use Screening Inventory dynamic risk factors, 176–9 dynamic time series regression, 47 dysphoria, 229 early secondary school developmental experiences, 134–6 EBTs, see evidence-based treatments eclectic approaches, 16–19 EFT, see emotion-focused therapy ego, 8 Ellis, A., 11 EMDR, see Eye Movement Desensitization and Reprocessing emotion-focused therapy (EFT) case formulation, 23, 24 emotional problems, 247 empirical research on case formulation, 63–75 conclusions, 74, 75 empirically supported treatments (EST), 5 epidemiology, 18, 19 escape activity, 13 EST, see empirically supported treatments establishing operation, 132 ethnicity, 18, 238 evidence as guide for formulation, 16 evidence-based formulation, 7 evidence-based practice, 5, 85, 103 evidence-based treatments (EBTs), 17 evidence-based violence risk assessment approaches to, 85, 86 discretionary procedures, 86, 87 what is it? 84, 85 explanatory hypothesis, 26 extraversion, 260 high trait, 259 Eye Movement Desensitization and Reprocessing (EMDR), 286 family, 5, 25, 47, 91, 99, 108, 112, 122, 136, 141, 148, 149, 180, 181, 196, 206, 223, 225, 228, 232, 238, 243–7, 265, 267, 289, 294 conflict, 222 history, 43, 156, 159, 183 risk domain, 222 therapy, 68
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX fear, 8, 13, 14, 16, 20, 111, 136, 137, 141, 145, 149, 150, 161, 185, 199, 200, 207, 210, 230, 264, 283, 293, 300 female perpetrators with adult victims, 199, 200 FFM, see Five Factor Model fictional case formulation, 266, 267 financial problems, 26 Five Factor Model (FFM), 260 flooding, 14 follow-up, 58, 72, 74, 90, 156, 169, 203 forensic case formulation of aggression, cognitive behavioral approaches, 107–23 application of ideas in, 225, 226 application of structured instruments in, 226, 227 with children and adults, 217–33 current issues, 33–56 current status, 284–8 data and, 294, 295 definition, 3, 4 developmentally informed, 229–31 difficulties in research on, 63 emerging issues, 283–301 evidence as guide for, 16 future directions, 294–300 future research, 76 general framework, 24–26 goals, 6, 7 individual, 110, 111 for individuals with personality disorder, 257–74 instrument-based, 228, 229 for offending behavior with people with learning disabilities, 237–53 practical tips, 27 process, 187, 188 professional training in, 296–300 research on: empirical findings, 63–74 scientific bases, 34–54 serious violent offending, Multiple Sequential Functional Analysis approach, 129–151 sexual offences against adults, 195–210 sexual offences against children, 175–91 not a single approach, 62, 63 substance abuse disorders and anger, 153–69 structured systematic models, 19–24 theoretical and evidence-based approaches, 3–27 theory as guide to, 7–14 and treatment, 52–4 of bulimia nervosa, 71, 72
309
of childhood behavioral disorders, 73, 74 of depression, 63–8 in management of mental retardation, 74 of marital problems, 68, 69 of obsessive-compulsive disorder, 70, 71 of phobic anxiety, 69–79 of psychosis, 72, 73 and treatment outcome, 61–76 and treatment planning, 231, 232 validity issues in, 44, 45 violence risk, 92–6 illustrative, 96–103 why formulate?, 5, 6 forensic risk, 218, 221, 289 forensic work with juveniles, 219, 220 formulation is not formulaic, 217, 218, see also forensic case formulation formulation-guided therapy, 6 Freud, Sigmund, 8 functional analysis, 12, 163–5 of alcohol use, 164 of cocaine use, 165 Functional Analytic Clinical Case Models (FACCMs), 22 future directions, 294–300 future research, 76 GAD, see generalized anxiety disorder gender, 18, 200, 238 General Personality and Cognitive Social Learning (GPCSL), 93 general systems theory, 10 generalized anxiety disorder (GAD), 14 genital stage, 8 GLM, see Good Lives Model goal setting, 23 Good Lives Model (GLM), 92, 93, 253 GPCSL, see general personality and cognitive social learning Gray’s behavioral activation and inhibition model, 250 Gray’s Reinforcement Sensitivity Theory, 260 Greenberg, L., 23 gregariousness, 260 grief, 18 guided clinical judgment, 87 Guidelines for Stalking Assessment and Management (SAM), 94 guilt, 8, 155, 167 Hayes, S.C., 11 Haynes, S.N., 22
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
310
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
HCR-20, see Historical, Clinical, Risk Management-20 heroin, 158, 159 high-risk situations, motives and coping, measures to assess, 157 high trait extraversion, 259 Historical, Clinical, Risk Management-20 (HCR-20), 130, 202, 265 Horowitz, M.J., 29 hospitalization, 122, 209 psychiatric, 43, 227 humanistic theory, 14, 15 hyperactivity disorder, 244 hypersexuality, 206 hypomania, 260 hypothesis testing, 4, 288 ICS, see Impaired Control Scale ID, see intellectual disabilities id, 8 ideas, application of in case formulation, 225, 226 idiographic assessment, 42 idiographic conflict formulation, 49 IDS, see Inventory of Drinking Situations IDTS, see Inventory of Drug-Taking Situations illness, 156 mental, 115, 142, 200, 201, 203–6, 290, 291 impaired control and dependence, measuring, 157 Impaired Control Scale (ICS), 157 impulsive problem solving, 259 impulsivity, 108, 111, 117, 118, 180–82, 204, 208, 266 incest, 176, 260, 264 incremental validity, 41, 42 index offenses, 97, 98, 145–147 individual and aggregate levels, 42 individual risk domain, 222 information gathering, 24 at intake, 159 information theory, 10 inhibition of dominant response deficit, 259 instrument-based formulation, 228, 229 Integrated Theory of Sexual Offending (ITSO), 257 intellectual disabilities (ID), 74, 239, 244, 246, 247, 249 theoretical context for treatment in sex offenders with, 241–3 interest, loss of, 5 interformulator reliability, 37 internal consistency reliability, 40 International Personality Diagnostic Examination, 265
International Personality Disorder Examination, 266 interrater reliability, 37–40 Inventory of Drinking Situations (IDS), 157 Inventory of Drug-Taking Situations (IDTS), 157 isolation, social, 13, 135–7, 139, 142, 147, 186, 187, 231, 238–40 ITSO, see Integrated Theory of Sexual Offending Kuyken, W., 22 learning disabilities formulating offending behavior with people with, 237–53 biological research and sociological theories, 237, 238 case study, 243–53 consequences, 258, 259 formulation, 247–53 four channels of response involved in incident, 250 framework for formulation, 243, 244 predisposing factors, 247, 248 risk assessment, 246, 247 sexual and relationship history, 245, 246 treatment implications, 251–3 precipitating/antecedent factors, 249 theoretical context for treatment in sex offenders with, 241–3 Level of Service-Case Management Inventory (LS-CMI), 93, 226 Life Stories and Self Characterizations (Kelly), 266 logical integrity, 272 LS-CMI, see Level of Service-Case Management Inventory Lubin Depression Adjective Checklist, 64 major depressive disorder, 5, 66 maladaptive extinction, 12, 13 maladaptive shaping, 12, 13 management plans, 101, 102 mania, 66, 204, 205, 209 marital problems, CF for treatment of, 68, 69 MAST, see Michigan Alcoholism Screening Test MCMI-III, see Millon Clinical Multiaxial Inventory measuring outcomes, 189, 190 mental, 115, 142, 200, 201, 203–6 mental health services, 84, 107, 110, 142, 291, 292
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX mental retardation, case formulation in management of, 74 see also intellectual disabilities mentally ill offenders, 290, 291 metacognitive skill, 224, 229 Michigan Alcoholism Screening Test (MAST), 156 Millon Clinical Multiaxial Inventory (MCMI-III), 185 Minnesota Multiphasic Personality Inventory (MMPI), 17 Minnesota Multiphasic Personality Inventory (MMPI-2), 202 Minnesota Sex Offender Screening Tool-Revised (MnSOST-R), 202 MMPI, see Minnesota Multiphasic Personality Inventory MMPI-2, see Minnesota Multiphasic Personality Inventory-2 modeling, 13, 47, 93, 108, 196, 237, 247, 248, 258, 269, 274 monitoring problem behavior, 162 mood, depressed, 5, 71, 159, 163, 168, 209 moral behavior, attachment, and social connection, 224, 225 Morton, J., 258 Morton causal modeling framework, 258, 259 MSFA, see Multiple Sequential Functional Analysis MSI II, see Multiphasic Sex Inventory – Adult Male Form Multiphasic Sex Inventory – Adult Male Form (MSI II), 202 multiple regression analysis, 47 Multiple Sequential Functional Analysis (MSFA), 131–4 in case formulation of serious violent offending, 129–51 case studies, 134–48 treatment implications, 149–51 murder, 151, 198, 288 narcissism, 117, 188, 203, 240, 261 narrative identity, 263 narratives, patients’, 7, 16, 19, 20, 51, 263 National Comorbidity Survey (NCS) community study, 155 National Incident-Based Reporting System (NIBRS), 199 NCS, see National Comorbidity Survey negative reinforcement, 133 negative thinking, 18 NEO-Five Factor Inventory (NEO-FFI), 265, 266 neurobiology, 5
311
neuroticism, 260 high, 259 neutral stimulus (NS), 132 Nezu, A.M., 23 NIBRS, see National Incident-Based Reporting System no criterion behavior, 293 nomothetic formulation, 21 nondiscretionary risk assessment procedures: evidence-based violence risk assessment, 86, 88, 89 nonpsychotic mood disorder, 67 NS, see neutral stimulus nursing, 84, 245 obesity, 19 object relations theory, 8 occupational therapy, 84 obsessive-compulsive disorder (OCD), 14, 71 case formulation for treatment of, 70, 71 occupational impairment, 5 offense history, 184 offense paralleling behavior (OPB), 92, 131, 147, 148 offense-topographic approach, 110, 111 OPB, see offense paralleling behavior openness, 260 operant conditioning, 12, 13, 16 operant learning, 12, 287 oral stage, 8 outcome research, 17 overdeterminism, 8 Padesky, C.A., 22 pedophiles, 264 PAI, see Personality Assessment Inventory panic attack, 14, 69, 159 paranoid delusions, 73, 205, 206 parent–child separation, 222 parent criminality, 222 pathogenic beliefs, 10 patient(s), 4, 6–10, 14, 16–19, 21, 33, 35–7, 39–41, 43, 45, 47, 49–51, 54, 61, 63–72, 75, 84, 90, 91, 94, 111, 115, 120, 121, 148, 151, 153–63, 166, 167, 169, 189, 190, 201, 229, 231, 232, 245, 266, 269, 271, 291 effect of sharing case formulation with, 53 as guide, 16 narratives, 7, 16 Pavlov, N., 13 PCC, see psychology of criminal conduct PCL-R, see Psychopathy Checklist-Revised PD, see psychodynamic
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
312
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
peer group risk domain, 222 personality, tests, 17 and offending, 259–65 Personality Adjustment Inventory, 265 personality as antecedent in cognitive behavioral formulation of aggression, 117 Personality Assessment Inventory (PAI), 17, 202 personality disorder(s) as antecedent in cognitive behavioral formulation of aggression, 115–17 case formulation for individuals with, 257–74 assessment, 265–73 for case formulation, 265, 266 characteristic adaptations, 262, 263 communicating formulation with brief summary document, 272 fictional case formulation, 266, 267 formulation and treatment plan, 267–73 narrative identity, 263 personality and offending, 259–65 relational dynamics, 263–5 risk assessment, 265 transient state repertoires as mode of trait expression, 261, 262 traits, 260, 261 offenders with, 291, 292 and substance abuse disorders, 155, 156 Persons, J.B., 21 phallic stage, 8 phobias, 14, 16, 23, 69, 70, 283, 284 phobic anxiety, case formulation for treatment of, 69, 70 plan formulation method, 39, 49 pleasure principle, 8 PMMD, see Psychological Model of Mental Disorder positive psychology, 294 positive reinforcement, 133 possible selves, construct of, 263 post-traumatic stress disorder, 14, 18, 34, 300 poverty, 222 precipitating/antecedent factors, 249 prediction of individual response to specific intervention or therapy event, 52, 53 predictive validity, 43–7 in behavioral case formulation, 46 in cognitive-behavioral case formulation, 46, 47 in psychodynamic case formulations, 47
predisposing factors, 247, 248 prior offenses, 98–100 prison, 109, 138, 142–5, 147, 148, 158, 159, 180, 181, 183, 184, 200, 203, 204, 206, 238, 248, 267, 287, 291 problem behavior, monitoring, 162 problem-solving, 111, 114, 232, 239, 266, 271 impulsive, 259 perspective for cognitive behavioral therapy case formulation, 23 process of case formulation and professional training, 298–300 process diagnosis, 23 professional training in case formulation, 296–300 psychiatric hospitalization, 43, 227 psychodynamic (PD) case formulations, 39, 40 comments and issues, 39, 40 predictive validity in, 47 psychodynamic (PD) theory, 8–10 psychological testing, 7, 25, 201 Psychological Inventory of Criminal Thinking Styles, 266 Psychological Model of Mental Disorder (PMMD), 258 psychology of criminal conduct (PCC), 93 psychometric applications, 16, 17 psychopathology, 5, 7, 16, 20, 41, 55, 61, 62, 71, 202 research, 17, 18, 35 psychopathy, 88, 111, 115, 118, 155, 180, 182, 187, 196, 202–4, 209, 231, 261 Psychopathy Checklist-Revised (PCL-R), 182, 202 psychopharmacological treatment, 26 psychosis, 66, 70, 72, 73, 108, 130, 141, 150, 154, 205, 207, 208, 250, 296 case formulation for treatment of, 72, 73 psychosocial development, 26 psychotherapy, 3, 5, 7, 18, 19, 39, 50, 61–3, 68, 74, 75, 156, 295, 296 process and outcome research, 17 psychotic symptoms, 18, 209 race, 238 rage, 8, 9, 149, 197, 198 randomized controlled trial (RCT), 45, 66, 67 rape, 97, 100, 180, 184, 195–200, 204, 208, 209, 299 attempted, 259, 268 Rapid Risk Assessment for Sex Offender Recidivism (RRASOR), 178, 202 rating scales, 7, 17, 73, 270 RCT, see randomized controlled trial
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX Reasons for Drinking Scale, 157 relapse prevention, 167, 168 relational dynamics, 263–5 relationships, poor, 135, 149, 223, 259 Relative Operating Characteristic (ROC) analysis, 178 relaxation, 13, 14, 167 reliability, 37–41 internal consistency, 40 interrater, 37–40 test-retest, 40, 41 Research Diagnostic Criteria, 63 respondent conditioning, 13 respondent learning, 12 risk assessment and risk management, 289, 290 with sex offenders, 176–80, 246, 247 assessment for case formulation, 182–90 case formulation, 179, 180, 186, 187 clinical example, 180–90 clinical and personality features, 185, 186 cognitive factors, 185 conclusions regarding risk, 182 developmental history, 183 dynamic factors, 176–9, 181, 182 formulation process, 187, 188 measuring outcomes, 189, 190 offense history, 184 psychopathy, 182 for recidivism, 200–203 reliability and accuracy of, 178, 179 static factors, 180, 181 treatment for sexual offending, 184, 185 risk factors, 18, 19, 84, 85, 87–96, 100–103, 147, 150, 183, 187, 190, 191, 201–4, 218, 220, 223, 224, 226, 228, 229–32, 259, 261, 289, 291, 294 dynamic, 176–9 static, 221, 222 Risk-Needs-Responsivity (RNR), model, 93–6, 226 Risk for Sexual Violence Protocol (RSVP), 87 analysis using, 100–102 RNR, see Risk-Needs-Responsivity ROC, see Relative Operating Characteristic Rogers, C.R., 15 role play, 73, 168 Role Relationship Model’s Configuration (RRMC), 20 RRASOR, see Rapid Risk Assessment for Sex Offender Recidivism RRMC, see Role Relationship Model’s Configuration
313
RSVP, see Risk for Sexual Violence Protocol rumination, 17, 18, 117 sadistic rape, 197, 208 SADQ, see Severity of Alcohol Dependence Questionnaire SAM, see Guidelines for Stalking Assessment and Management satiation, 12 SAVRY, see Structured Assessment for Violence Risk in Youth scenarios, 100, 101 Schedule for Affective Disorders and Schizophrenia Interview, 63 Schizoaffective Disorder, Bipolar Type, 206 schizophrenia, 12 school, 34, 99, 114, 135–7, 139, 158, 183, 196, 205, 223, 226, 227, 238, 244, 245, 246, 267 refusal, 53, 285 risk domain, 222 truancy, 222 SCID, see Structured Clinical Interview for DSM Disorders scientific bases of case formulation case formulation and treatment, 52–4 components and content validity, 34–7 reliability, 37–41 validity of case formulation and its components, 41–52 screening instruments, 156 SD, see discriminative stimulus S, see stimulus delta secure attachments as protective, 223, 224 self-actualization, 15 self-esteem, 149, 163, 230, 239 self-harm, 18, 38, 107, 143, 266 self-injurious behavior, 53, 74, 227–9 self-monitoring diary, 162 selfobject, 9 self-psychology, 8, 9 self-report, 47, 49, 67, 181, 190, 201, 202, 266, 269 self-system therapy (SST), 66 serial rape typologies, 197–9 Severity of Alcohol Dependence Questionnaire (SADQ), 157 Sex Offender Need Assessment Rating (SONAR), 177, 202 Sex Offender Risk Appraisal Guide (SORAG), 201, 248 sex offenders, risk assessment with, 176–80 sexual abuse, 135, 136, 139, 141, 143, 145, 181, 183, 187, 196, 205, 207, 239–43, 247, 248, 250, 251
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
314
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
sexual assault, 97, 99, 180, 200, 203, 204, 206–9, 288 of adult females, 195–9 of adult males, 199 sexual coping, 239 Sexual Fantasy Questionnaire (SFQ), 202 sexual offences against adults, 195–210 assessments, 200–203 case formulation, 203–9 literature review, 195–200 against children, 175–91 risk assessment with sex offenders, 176–80 treatment for, 184, 185 sexual and relationship history, 245, 246 sexual violence, 87–9, 96, 97, 100–103, 154, 195, 197, 201–5, 207, 208, 248 Sexual Violence Risk-20 (SVR-20), 202 sexuality, deviant, 182, 184, 185, 187, 191, 200, 201, 204, 208, 241 SFQ, see Sexual Fantasy Questionnaire shoplifting, 184, 227, 244 Silberschatz, G., 20 Situational Confidence Questionnaire, 157 Skinner, B.F., 12 social anxiety, 18, 65, 71, 135 social connection, attachment, and moral behavior, 224, 225 social development, 189, 220, 229 social disability, 232 social disconnection, 231 social impairment, 5 social isolation, 13, 135–7, 139, 142, 147, 186, 187, 231, 238–40 social skills, 26, 63–5, 73, 74, 101, 143, 149, 150, 188, 207, 208, 219, 231, 232, 287 social work, 84, 159 sociological theories and biological research, 237, 238 SONAR, see Sex Offender Need Assessment Rating SORAG, see Sex Offender Risk Appraisal Guide SPJ, see structured professional judgment (SPJ), 83–104 SST, see self-system therapy, 66 stability, test-retest reliability and, 40, 41 STABLE-2000, 178 STABLE-2007, 177, 178, 202 stable dynamic risk factors, 177 STATIC99, 180, 181, 202 static risk factors, 221, 222
stimulus delta (S), 132 stress, 122, 181, 186, 187, 221, 239, 251, 257, 266, 300 Strosahl, K.D., 11 structural diagnostic approach, 110, 111 Structured Assessment for Violence Risk in Youth (SAVRY), 226 Structured Clinical Interview for Diagnostic and Statistical Manual Disorders (SCID), 17, 169 structured interviews, 156 structured professional judgment (SPJ), 83–104 guidelines, 83, 84 substance abuse disorders, 43 and anger, 153–69 assessments in, 156 case formulation, 158–69 avoiding cues versus learning to cope differently, 166 building therapeutic relationship, 161, 162 coping skills handling anger, 167 handling craving, 166, 167 follow-up, 169 functional analysis, 163–5 increasing motivation for change, 159–61 information gathered at intake, 159 monitoring problem behavior, 162 relapse prevention, 167, 168 role play, 168 setting goals for treatment, 162, 163 treatment plan, 166 treatment progress, 168, 169 and criminal behavior, 153, 154 and personality disorders, 155, 156 and violence, 154, 155 suicidal behavior, 154, 227 suicide, 154, 155 attempts, 19, 42, 43, 46, 143, 158 risk, 19, 71, 154 superego, 8 SVR-20, see Sexual Violence Risk-20 symptom production, 8 symptom rating scales, 17 TDCRP, see Treatment of Depression Collaborative Research Program test-retest reliability and stability, 40, 41 therapeutic relationship, building, 161, 162 therapy interventions congruent with CF, benefits of, 53, 54 thought record, 7
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX threat threshold, 240 Time-line follow-back interview (TLFB), 156 TLFB, see time-line follow-back interview training and competency issues, 54, 55 criterion validity: using experts’ case formulations to evaluate training in case formulation skills, 54 level of training and expert/novice differences in case formulation, 54, 55 professional, 296–300 traits, 260, 261 transdiagnostic approach, 258 transient state repertoires as mode of trait expression, 261, 262 trauma, 9, 20, 21, 26, 112, 161, 200, 209, 225, 288, see also post-traumatic stress disorder treatment implications, 251–3 plan, 166 planning, 232 progress, 168, 169 setting goals for, 162, 163 for sexual offending, 184, 185 Treatment of Depression Collaborative Research Program (TDCRP), 67 treatment justification diagrams, 258 trichotillomania, 12 United Kingdom, 89, 107, 120, 141n, 144n, 291 Department of Health, 291 Ministry of Justice, 291 United States, 154, 198, 199 Department of Justice, 196 US, see unconditioned stimulus UR, see unconditioned response unaided clinical judgment, 86 unconditioned response (UR), 13 unconditioned stimulus (US), 13, 14, 132 unconscious motivation, 8 unemployment, 99, 115, 156 unstructured professional judgment, 86 validity of case formulation, 41–52 complexity, 43 general issues, 41–3 incremental, 41, 42 predictive validity, 43–7 and reliability, 272, 273 victim empathy, 239
315
victimization, 196, 197, 200, 208, 230 Violence Against Women Survey, 196 violence and aggression cognitive behavioral approaches, 107–123 angry affect in formulation of aggression, 112, 113 definitional problems, 109 function assessments, 119, 120 further assessment issues, 118, 119 heterogeneity, 113–15 implementation of clinical formulation in practice, 120, 121 individual and population functional assessments, 117, 118 mental disorder and aggression, 115 personality as antecedent, 115–17 personality disorder as antecedent, 115–17 problem for whom?, 109, 110 two contrasting case formulations, 121–3 types of antecedent, 111, 112 why formulate individual case?, 110, 111 multiple sequential functional analysis approach (MSFA), 129–51 SPJ approach, 83–104 illustrative case, 96–103 analysis using Risk for Sexual Violence Protocol, 100–102 analysis using STATIC-99, 102, 103 comment, 103 conclusory opinions, 102 formulation, 100 index offenses, 97, 98 management plans, 101, 102 prior offenses, 98–100 risk factors, 100 scenarios, 100, 101 social history, 99, 100 summary of findings, 97–100 practice of risk assessment, 84–91 and substance abuse disorders, 154, 155 Violence Proneness Scale (VPS), 156 Violence Risk Appraisal Guide (VRAG), 246 Violence Risk Scale (VRS), 265 Violence Risk Scale Sex Offender version (VRS-SO), 265 violent crime, 107, 108 VRAG, see Violence Risk Appraisal Guide VRS, see Violence Risk Scale VRS-SO, see Violence Risk Scale Sex Offender version
P1: OTA/XYZ JWST073-IND
P2: ABC JWST073-McMurran
316
May 28, 2011
13:35
Printer Name: Yet to Come
INDEX
war veterans, 14 Ward and Hudson pathways model, 249 Weiss, J., 20 What Works movement, 110 Wisconsin Card Sorting Test, 266 withdrawn (higher order personality dimension), 250 Wolpe, J., 14
YLS-CMI, see Youth Level of Service-Case Management Inventory YOI, see young offenders institution Young, J.E., 11 young offenders institution (YOI), 137, 138 Youth Level of Service-Case Management Inventory (YLS-CMI), 226
