Doctoring: The Nature of Primary Care Medicine

Doctoring: The Nature of Primary Care Medicine

ERIC J. CASSELL, M.D.

OXFORD UNIVERSITY PRESS

DOCTORING

This page intentionally left blank













DOCTORING The Nature of Primary Care Medicine











ERIC J. CASSELL, M.D.

1

3 Oxford New York Auckland Bangkok Buenos Aires Cape Town Chennai Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi São Paulo Shanghai Taipei Tokyo Toronto

Copyright © 1997 by Oxford University Press, Inc. First published in 1997 by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Copublished with the Milbank Memorial Fund. www.oup.com. First issued as an Oxford University Press paperback, 2002 Oxford is a registered trademark of Oxford University Press 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, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data Cassell, Eric J., 1928– Doctoring: the nature of primary care medicine/Eric J. Cassell. p. cm. Includes bibliographical references and index. ISBN 0-19-511323-3; ISBN 0-19-515862-8 (pbk.) 1. Physician and patient. 2. Primary care (Medicine) 3. Holistic medicine. 4. Primary care (Medicine)—Study and teaching. I. Title. [DNLM: 1. Primary Health Care. W 84.6 C344d 1997] R727.3.C367 1997 610—dc20 DNLM/DLC for Library of Congress 96-34178 Chapter 3 is adapted from “The Sorcerer’s Broom: Medicine’s Rampant Technology” by Eric J. Cassell, which appeared in The Hastings Center Report, 23:32–39, 1993 copyright The Hastings Center, by permission of the publisher.

2 4 6 8 9 7 5 3 1 Printed in the United States of America on acid-free paper

This book is dedicated to my beloved wife

Patsy

This page intentionally left blank

Contents

Foreword,

IX

Acknowledgments,

XI

Introduction, 3 1 What Is Primary Care? 18 2 The Heavy Hand of the Past: Thinking About Diseases Versus Thinking About Persons, 43 3 The Special Problem of Technology, 62 4 The Clinical Method, 81 5 The Clinical Method and the Patient, 105 6 Where Should Primary Care Be Taught—and by Whom? 126 7 What Should Be Taught? 144 Epilogue, 177 Notes, 189 Index, 199

This page intentionally left blank

Foreword

The Milbank Memorial Fund is an endowed national foundation that supports nonpartisan analysis, study, and research on signi¤cant issues in health care and public health. Since its founding in 1905, the Fund has encouraged research and discussion that may lead to enhanced consideration of policy alternatives that a signi¤cant number of decision makers regard as achievable. Eric Cassell contributes to improved health care policy in this book because he is an active clinician and a gifted writer rather than a policy maker or analyst. The book began during a conversation in which Cassell complained to one of us that many advocates of policy to promote “primary care” and “managed care” misunderstood the problems that patients brought to physicians. Furthermore, he said, policy advocates often oversimpli¤ed what good physicians needed to know about medical science, their patients, and themselves. Cassell has responded magni¤cently to the Fund’s challenge to inform policy rather than to complain about it. He revised the manuscript repeatedly to take into account the views of the physicians and policy makers recruited by the Fund to advise him. In an unconventional conclusion to an editorial process, Cassell spent an exhausting day reviewing the manuscript line by line with a group of

[x]

Foreword

physicians and public of¤cials as well as staff of the Oxford University Press and the Fund. The Fund is most grateful to everyone who helped in the making of this book and, most of all, to Eric Cassell. Daniel M. Fox President Samuel L. Milbank Chairman

Acknowledgments

[xi]

Acknowledgments

This book came into being because of the urging and continued encouragement of Daniel M. Fox, the president of the Milbank Memorial Fund. In the fall of 1993, he asked me to write an essay about primary care for submission to the Milbank Quarterly which should have been about 7500 words. When I ¤nally ¤nished, it was 18,ooo words! I realized that, after more than three decades as a general internist, I had a lot more to say about primary care than I originally believed. A greatly pared-down version of the essay was published.1 The commentary of the reviewers of the ¤rst and second versions of the essay, known and unknown, were very helpful and I extend my thanks to them. I must single out Dr. Barbara Star¤eld, one of the reviewers, for her help. Her excellent book, Primary Care: Concept, Evaluation, and Policy, was a fortunate place to start my review of the literature. The original manuscript of the essay was expanded into a book. It was, I must confess, dreadful. I returned to the literature to understand better the current status and development of primary care medicine. It was necessary to systematically explore problems, interests, and concerns about the care of patients that had been percolating in my own thoughts for years. It was also essential to read more widely to learn about the history and philosophical background of the

[xii] Foreword Acknowledgments present state of medicine as well as how to solve some of the problems posed by the clinical method discussed in Chapters 4 and 5. Many physicians and others concerned about medicine and its future, including Leon Kass, Edmund Pellegrino, Balfour Mount, David Barnard, S. Kay Toombs, and Daniel Callahan, supported the importance of writing about the training of physicians not only for these bleak times, but for what will emerge in coming decades. I am indebted to members of the Master’s Class of 1993, Drs. Eugene C. Corbett, Jr., Joe Gieni, Michael S. Greenberg, Sandra Harewood, Thomas S. Kaluzynski, Allan V. Prochazka, Jill A. Rhymes, and Harvey Schildkraut, for helping me develop this method of teaching medicine. I owe a special debt to one of the master clinicians of our time, Joseph D. Sapira, who helped teach the Master’s Class (and me). Dr. Fox arranged for an extraordinary group of reviewers to read the manuscript and prepare lengthy commentaries. This is, as we all know, a time-consuming task, but the comments from the following—not always kind or gentle—were universally helpful: Drs. Paul D. Gerber, Larry A. Green, Michael S. Greenberg, John Moyer, Fitzhugh Mullan, Richard G. Roberts, John D. Stoeckle, and Neil A. Vanselow. I was taken to task, as had been the case with the reviews of the original essay for my lack of understanding of Family Medicine. To help educate me, Dr. Roberts generously gave of his prodigious energy and his own time, as did his colleagues, to intensively brief me about their Family Medicine program at the University of Wisconsin Medical School. It was an exhilarating day. The Milbank Memorial Fund arranged for those reviewers who were able to attend a day-long session of commentary about the manuscript. It was a day of great conversation and useful criticism of the book. It is an unusual procedure, but so helpful that I wonder why it is not followed more frequently. I am especially indebted to Dr. Michael S. Greenberg. He reads more widely, cares more, and thinks more deeply about medicine than almost any physician I know. He was a willing sounding board for the ideas in this book and a generous and extended commentator on the manuscript in parts and whole. Joan Bossert, at Oxford University Press, has been an encouraging and helpful editor from our ¤rst meeting. The staff at the Milbank Memorial Fund, especially Kathleen Andersen, are due my appreciation for their help. I am grateful to my friends and colleagues at medical schools and hospitals all over the United States for the

Acknowledgments

[xiii]

opportunity my visits to them, over the years, have given me to hear about what is going on in medicine—new ideas, developing programs, and dedication to lasting values. First, last, and always I owe an immeasurable debt to my patients for being so patient with me in all my travels, but most of all for providing my life in medicine.

This page intentionally left blank

DOCTORING

This page intentionally left blank










Introduction










T

his is a book about primary care medicine and its increasing importance in solving the health care problems of individuals and communities now and in the future. Because primary care is the capstone of twentieth-century medicine, it is also the foundation for twenty-¤rst-century doctoring. The restructuring of the American health care system going on now at a thoughtlessly rapid pace is pushing primary care into prominence in managed care organizations throughout the country, with apparently little understanding of what primary care medicine should be. In the unsettled contemporary medical world, it is little noticed that during its rise to prominence the idea of primary care has been changing from largely an organizational concept—related to the hierarchy of services in medical care—to a sophisticated generalism. The pressure for its new form comes from rising costs and current dif¤culties in health care delivery and subspecialty medicine. The emergence of the new generalism is also propelled by the mismatch between the high-technology medicine at which we excel and the health care needs of large groups of the population—for example, the poor, chronically ill, aged, and disabled. A shortage of physicians in rural and other underserved areas is also an impetus for ¤nding ways of introducing more primary care physicians into these settings.1 [3]

[4]

doctoring

The rapid change in the health care system that is bringing primary care medicine to the fore, mostly to reduce costs, also poses threats to its development. At a time when money speaks so loudly, primary care has come to be seen by many as a kind of medicine with ¤nancial, social, and organizational advantages, with little regard to it as a kind of medicine in its own right. It is inexpensive compared to high-technology specialist care; it can be provided in a physically accessible fashion and can ¤t into the social structure of the patient populations it serves; and it is administratively uncomplicated since it can be delivered in community settings (although less so in some managed care environments). Some other common de¤ning characteristics that have been discussed are that primary care physicians are ¤rst-contact doctors, that they may act as gatekeepers—aiding the more rational use of resources—and that they are not specialists or are not functioning as specialists. It is a common and destructive error, because of these obvious organizational advantages, to act as though the medicine itself is simple. Primary care medicine is based on the centrality of the patient rather than on an organ system or a disease, as is the case with specialism. It is addressed to both the sick and the well. It understands functional impairment and disease to be processes that enter into the patient’s life story, so that its interventions are chosen with the development of that story in mind. Because of this, it is as well suited to prevention as to treatment, to children as to adults, and especially to the care of the chronically ill, who make up the largest number of the sick in our society. Primary care medicine can best be provided by generalists who are speci¤cally trained to meet the broad, as well as the intellectually and technically exacting, demands implied in the de¤nition of the term. These are doctors who are able to come to know the sick or well person and join this information with their knowledge of medical science, disease, and technology in the diagnostic, therapeutic, and preventive processes. So, generalism and generalist are terms that have come into use, in part, to counter the simplistic ideas often associated with the term primary care. Thus, the ideas that underlie current understandings of primary care medicine have been evolving since the 1920s, slowly in the beginning but much more rapidly in the last thirty years. In this period two other movements in medicine, as well as widespread social change, have further de¤ned what is asked of primary care physicians. The family physician movement, gaining force in the

Introduction

[5]

196os, decrying specialty medicine’s concentration on the disease rather than on the patient, sought to focus the doctor on the patient in a special way. In G. Gayle Stephen’s words in 1975: Family physicians know their patients, know their patients’ families, know their practices, and know themselves. Their role in the health care process permits them to know these things in a special way denied to all those who do not ful¤ll this role. The true foundation of family medicine lies in the formalization and transmission of this knowledge.2

A second movement, increasingly visible in the 1970s,, became another force that led to the care of a patient as a sick person within a family and community matrix; it became known as the hospice movement. Indeed, the term palliative care, which was adopted, implied the failure of disease-oriented medicine to cure the patient or meet the needs of the patient and family. Palliative care is often associated with symptom control, but hospice physicians know that symptom control is inadequate in the absence of a much broader understanding of dying patients. Suffering is an af¡iction of persons, not bodies, and can occur in relation to any aspect of a person: physical, psychological, social, or spiritual.3 The family physician and palliative care movements were born during a period in the United States marked by a great expansion in our understanding of the concept of person. The civil rights and women’s movements, the rise of bioethics, the embrace of difference and diversity, and the consequent disappearance of the melting pot metaphor all celebrated the emergence of an enriched concept of person. For medicine, this meant not merely an individual or a bearer of rights but: “Me, doctor, treat me, not just my lungs or liver.” The family physician movement grew rapidly in the early years after its of¤cial designation as a specialty in 1970 but then faltered, its growth slowing until its recent marked resurgence. Palliative care continues to struggle to gain acceptance within mainstream medicine, although the number of hospices in the United States continues to rise. During these years, despite the problem of acceptance and well before the contemporary managed care explosion, the medical literature re¡ected a progressively increasing interest in generalism. In light of the attention given to the subject and its importance, however, what has been written about primary care medicine is disappointing because it is incomplete. The literature makes it clear that primary care physicians—generalists—will no longer focus on a

[6]

doctoring

patient’s physical disorder but will also be aware of psychosocial factors in health and disease, and of their patient’s and their own place in the community. They will be responsive not only to the varied needs of individual patients but also to the other demands of the health care scene, from economic to environmental. They will understand the importance of preventive medicine and their role in helping their patients and communities lead healthy lives. But several questions remain: How will this new generation of primary care physicians accomplish these things? What new kinds of knowledge will have to be gained? Who will teach these doctors-in-training? And what will actual, day-to-day doctoring look like? One might object that the family medicine literature now, and for years, has, for example, emphasized the doctor-patient relationship, the whole patient, communication skills, the context of the patient’s family and community, and a biopsychosocial model of illness.4 No one questions the soundness of these ideas; the problem is that after a full generation of prominence they simply have not thrived within a disease-oriented, technology-driven medical establishment. For two generations we have asked doctors to focus on the patient as a person, yet, more often than not, we still see the patient’s human concerns swept away by the technological imperative. If primary care is a better medical practice, why hasn’t it won the ¤eld? The failed medical programs of the 1960s must be kept in mind for the lesson they teach. Virtually all the descriptive ideas and terms currently used to envision the advantages of primary care— for example, continuing, coordinating, comprehensive, treating the whole person—characterized the medical programs funded as part of President Lyndon Johnson’s War on Poverty. Naturally enough, as the money dried up, the programs and the medical care institutions that were part of them disappeared (with the exception of today’s community health centers, the grandchildren of the neighborhood health center). The ideas, unfortunately, also withered, suffering from malnourishment. They did not catch on or become institutionalized because, without medical knowledge and skills to match the rhetoric, they were blown away by the fresh winds of specialty medicine and burgeoning technology. During the same period, many medical schools had social and behavioral science programs that also generally failed to translate their teaching into medical practice, and they too faded. If medical generalists are illtrained to meet the expectations imposed on both physicians and

Introduction

[7]

the public regarding primary care medicine, primary care medicine will fail. There is little evidence that the lessons of this history have been learned. Each new generation of educators attacking the problem and trying to broaden the approach of physicians to include sick and well persons, in all the dimensions of the word person, confuses the fact that because they know the need to solve the problem, they know what is needed to get things done—as if both are the same. Caring is not enough. In each new era, some persons tend to act as if no one else knows what has to be done, or as if others are not true enough to the ideals, or lack desire or will—as though what has to be taught are the principles of primary care. This book is based on a very different supposition. Let us start by acknowledging that the central idea of primary care—that the person is the subject and object of medical care—is already widely known and accepted. Suppose, further, that it is known and accepted that the social and psychological elements that characterize the lives of persons have an impact on their illnesses. That students and physicians-in-training know these things but are unable to put them in practice to the extent that they win out over disease-oriented medicine and our ubiquitous technology. Granting the truth of these statements changes the educational problem. The issues are clari¤ed by realizing that primary care medicine stresses not only the central place occupied by the individual patient, but equally the position of the individual doctor.* The title of this book, Doctoring, re¡ects the fact that in primary care medicine it is the being of the physician, not just doing, that counts. Physicians are not merely bearers of knowledge and skills, vitally important as those are, but are themselves the instruments of care. This represents a shift away from the idea that has occupied twentieth-century medicine: that it is impersonal objective medical science that knows the disease and effects the treatment. It also represents a change in the requirements of training. Much of what is required to care for sick persons we have asked the art of medicine to do in generations * Primary health care can be, and is being, delivered by nurses and physician assistants as well as by physicians. The need and the function create a framework in which to put roles. There is no need to get tied up in con¡icts of power and hierarchy. What needs to be stressed is that the same principles, the same knowledge base, and the same skills are required of all. It is the kind of medicine that counts, not who does it.

[8]

doctoring

past. Education in regard to medical science and technology has far outpaced that of the art. The training of primary care physicians must recognize a distinction between doctoring itself and the medical science on which it is based. If primary care physicians are to ful¤ll their anticipated role, teaching the techniques and knowledge base of doctoring as well as how to be a doctor should be as explicit as teaching medical science. A true and sustained shift toward the training of primary care physicians, therefore, will rely on distinct changes at all levels of medical education. The principles are clear enough; there have been enough classes that provide examples or even role models; what is needed now are knowledge-based skills. For example, properly taught communication skills based on knowledge of how the spoken language works and an understanding of the nature of relationships will endure within physicians who practice long after the knowledge of medical science learned in medical school has become obsolete. One of the problems I address in this book is the need to train physicians so that their subjectivity can meet objectivity on level ground. The newer focus of primary care physicians is the enhancement, preservation, or restoration of physical, psychological, and social functioning within the context of community. The relief of suffering stands alongside the preservation of life. This focus cannot be adopted merely by reorienting the training of doctors or making them aware of patients’ needs. The patient, as a sick or well person, is in many ways a new object of interest. For this reason, doctors require the methods of the naturalist—understanding, observation, thought, and judgment—that will allow them to really see patients as persons, apart from the mechanisms of disease. We are not speaking of disease and also of the patient, the dominant understanding of this century, but of the patient ¤rst and the disease and pathophysiology through the patient. The kind of information that doctors require in order to know patients in this manner is often subjective—arising within the subject who is the patient or the subject who is the doctor. The subjective becomes objective by being thought about—it becomes an object of thought. The information is often a result of observations and interpretation by the patient that must, in turn, be disentangled by doctors. The patient, after all, is the singular source of facts; only patients themselves can know what they experience. The information that physicians can optimally obtain directly from their senses results from a medical

Introduction

[9]

empiricism whose method reaches back to Thomas Sydenham and John Locke. The problem of designing educational systems to teach these methods is complicated by the fact that the kind of knowledge by which physicians know disease and the output of technology is different from and often in con¡ict with the kind of knowledge by which persons are known. Knowing the history of this con¡ict, as well as how it is expressed in day-to-day medical practice (discussed in Chapter 2), is important to educators if students and physiciansin-training are not to be constantly subverted by the lure of “hard data.” The sweet song of technology itself requires understanding so that physicians are actively trained to make it a tool rather than a master. Considering the unparalleled growth of technology in the last decade and the manner in which it has come to dominate the medical scene, it is clear that this is no simple matter. Chapter 3 addresses this problem. I may seem to be more attentive to graduate medical education than to the teaching of medical students. This is not the case. What is proposed as necessary for the modern generalist should be part of the training of all medical students and part of the skills of all physicians. It is common knowledge that curricular change in American medical education is a slow process subject to powerful internal and external political and economic forces.5 This seems particularly true of the traditional medical schools of the Northeast, which may be the last to get the idea. Since this is the milieu in which I live and work, I had developed a certain cynicism about the possibilities for change. Cynicism is generally an ineducable state, but in writing this book I have discovered that many schools in the United States are challenged by the need to teach the fundamentals of primary care and are moving in that direction. A recent study of ¤ve schools dedicated to the ideals of medicine that underlie this book found that “the most striking institutional characteristic—present in them all—is the strong presence of an explicitly stated mission, philosophy, or theoretical model that embraces and advances a more integrated approach to care and forms the foundation for the curriculum.”6 Forging a new institutional philosophy that is widely embraced is dif¤cult, but it is happening. The managed care explosion may provide further impetus for change. Graduate primary care programs, on the other hand, control their own teaching, so that they do not have to wait for a

[10]

doctoring

change in their schools’ curriculum. They are also smaller and require fewer faculty. Successful graduate programs can demonstrate the effectiveness of the ideas and serve as a training ground for faculty, which will make it easier to change the local climate of opinion and move the program into the medical school. The fundamental knowledge base for primary care remains the traditional preclinical science of medicine. It is the foundation from which modern Western medicine derives its legitimacy. It is the basic source of knowledge about nature as it is expressed in the body in health and disease. All surgical and medical interventions in the pathophysiology and pathoanatomy of disease are founded on it. It is about what Carl Rudebeck calls the body-as-nature.7 We must hope that when we get our recent graduates, they know it well. There is no substitute for clinical skill. It would be nice if they also knew social science, but this is too often not the case. It would make easier our task of teaching about the body-as-self, again after Rudebeck. What primary care doctors require as part of their mandatory training are additional kinds of knowledge and skills. They must be taught the behavior of sick and well persons, advanced communication skills, the lessons about the evaluation of data from clinical epidemiology, how to acquire information from disparate sources and use it in making judgment and decisions, and a greater understanding of human function and disability. They must also learn to master technology through explicit training, as well as learning modern therapeutics. A thorough grounding in preventive medicine is also necessary. These educational developments will provide opportunities for exciting curricular innovation. Chapter 7 is devoted to a detailed discussion of these kinds of knowledge. Chapter 6 addresses who will teach the new knowledge on which modern generalism is based and what to do in the absence of a suf¤cient cadre of teachers. There is always the danger that new programs will speak about, for example, communication skills, the importance of psychosocial elements, and the place of the community but will not teach them adequately because we lack adequate ideas or teachrs. It will be sad if the inadequacies of the primary care physicians trained in these programs are taken as evidence of the failure of the underlying concept. It is a hazard for which we must be alert. When examining any method of teaching, it is reasonable to ask what are the fundamental beliefs on which it is based and then see if the teaching method and content follow logically from the beliefs.

Introduction

[11]

Thus, if effective communication is considered important, does the teaching method merely emphasize that point, as in those exercises in which a resident’s interaction with a patient or family is videotaped and then critiqued? Or, alternatively, are the fundamentals of effective language use taught in recognition of the fact that in medicine we are most effective when our skills are based on underlying basic knowledge, as is the case with physical diagnosis, which is based on anatomy and pathophysiology? How and where the knowledge base of pimary care should be taught is examined in Chapter 6. Much of the suggested change in education has focused on the place of training. It seems clear that the traditional method of training physicians primarily on the wards of teaching hospitals is inadequate. Many consider it essential to provide primary care training in an ambulatory setting such as an outpatient clinic or an of¤ce practice, or in the community.8 Unquestionably, the problems presented by patients outside of the hospital are different from those of inpatients, and different skills are necessary for their care. Furthermore, many patients who previously required a hospital for their care or surgery are now commonly treated outside the hospital. But there can be no change in the direction of medicine without a concurrent change in the training of doctors, so that their education prepares them for their actual tasks in the care of patients. This goes far beyond merely changing their place of training.9 Changing the place of training also changes the kinds of problems physicians face. In an outpatient setting they will gain experience in the everyday issues that face primary care doctors. A number of writers about primary care point out that knowing the frequency of diseases in the populations that family physicians actually see teaches us that patients often come to doctors with symptoms but no disease. This is a reason to change the emphasis from recognizing disease to understanding and ferreting out the biopsychosocial process that leads to the symptom. All symptoms have causes, pathophysiology, and meaning. There is no reason to avoid training about serious diseases even if they are rare. It would, for example, be an egregious and probably fatal error for a doctor to not recognize and treat early meningococcemia because it is uncommon. On the other hand, musculoskeletal disorders are very common and, except for the osteopathic schools, their recognition and treatment are generally inadequately taught to undergraduates or residents. The underlying problem correctly addressed by the stress on knowing the illness patterns of patient populations is the

[12]

doctoring

still too common belief that diseases are more real than the patients who have them.10 The goals of postgraduate training of generalists cannot be adequately met by clinical training alone, no matter what the setting. Systematic instruction in classrooms or seminars is necessary to solve the problem of the con¡ict and lack of balance between trained objectivity and trained subjectivity already discussed. I am well aware that this idea is both unfashionable and repugnant to most medical educators, but I believe we must reexamine it carefully. The present method of training was developed in Sir William Osler’s era, when Osler’s objective (see his textbook) was to teach about the actual presentation of disease and its variability among patients, as well as the impact of this on the diagnosis. He and his colleagues were wedded to the new ideal of science in medicine that was just coming on the scene. The newly developed clinical laboratories were just off the wards of The Johns Hopkins Hospital, and they demonstrated the direct applicability of science to clinical medicine. Most doctors ultimately did either medicine or surgery (or both), and their teachers did the thing they taught—they were practicing physicians. The lesson of Johns Hopkins was then introduced into practice. Osler’s basic message is now taken for granted, but it is forgotten that his teaching method was in the service of an idea. The necessary skills of doctoring are now much more advanced, and new methods of teaching must be developed to meet the idea. Postgraduate instruction must teach doctors to be their own instrument with such con¤dence in the discipline of their subjectivity that it can compete with possibly con¡icting images seen on ¤lms or ideas embodies in the numbers on a printout. Hands-on postgraduate training is no longer adequate to this task. Primary care is not a unitary ¤eld. Family physicians, general internists, and general pediatricians have different perspectives. Family physicians have a wider range of clinical skills and are more concerned with well persons. General internists are more concerned with sick persons, and their knowledge has greater depth but less breadth. General pediatricians are, by de¤nition, interested in children and adolescents, and focus on growth and development. For them, as for physicians in the other ¤elds of primary care medicine, the patient has center stage. Despite their differences, these disciplines share a fundamental concern with persons, sick or well. This book discusses what is common to these different approaches to

Introduction

[13]

primary care and the kind of training required by all. Specialists undoubtedly should also care more about their patients than about their diseases, but appropriate changes in their training are not under the control of generalists and will more likely come about after, rather than before, primary care has demonstrated success in training and clinical performance. This book is written by an internist in the urban Northeast of the United States. It bears the stamp of the context of its author’s life experience. In the current intellectual climate, it is no longer possible to claim that a set of ideas or the problems from which they arise have timeless or universal relevance; even the belief that concepts of molecular biology meet such standards will not hold water. Instead, one wants to see how an author thinks about the problems he or she raises. One also wants to see whether the ideas and the methods of thought are applicable or can be adapted to a different place and time, and under different circumstances, because the organizing thesis has wide applicability. For me, the central issue of this book is that the care of patients, sick or well, has not been adequately served by the high-technology specialty medicine from which we are evolving or by the ideas about patient care inherited from the past. Clinicians have always had to ¤nd methods to overcome the inadequacies of the medicine they were taught. Some solutions are private or even unspoken, while others are shared or published. Few clinicians, however, are fully or adequately trained to take advantage of all that we have come to understand about the state of patienthood and doctoring. Some training programs are ahead of others, and some parts of the nation are (thus far) less af¡icted by the current profound changes in the organization of medical services than others, but all of us have a long way to go. This book is in the service of the journey to educated doctoring. The perspective from which I write may disturb some because of my apparently single-minded concentration on the individual patient and doctor and on their relationship. I believe one cannot know any particular patient except through the relationship with that patient—not any relationship, but the doctor-patient relationship.11 The impact of culture, society, community, or family on the individual patient and illness is profound and omnipresent. These constitute the social fabric of the patient, but their in¡uence comes about because they are instantiated within the person—within the concepts and language, knowledge and beliefs, habits and social

[14]

doctoring

rules that direct behavior. Physicians and other caregivers acknowledge the importance of the social makeup of the patient for health or illness by facilitating the ¡ow of information from the patient to them based on respect for persons and un¤ltered through preconceptions or prejudice. When physicians are in the presence of the patient, connected through the relationship so that they can know the patient, they bring to the experience their knowledge of the social and personal dimensions of the human condition that helps direct and interpret the interaction. In order to understand the individual, doctors must know about the wider cultural and social milieu in which their patients live. If doctors do not know, for example, that corporations are currently downsizing, they will have dif¤culty understanding the concerns of an apparently successful middle manager. Similarly, not to know about the Hasidic family structure is not to understand the dynamics of the Hasidic couple in the consulting room. The saying that anything a doctor does or reads teaches medicine is a truism because it is true. Caution is required so that physicians do not use this knowledge to create abstractions that would interfere with their direct knowledge of the patient. Acting only on knowledge of families in general can produce errors as easily as acting only on knowledge of pneumonias in general. How knowledge of all the aspects of personhood is employed will vary, of course, with the clinical problem—for example, the care of a dying patient or the encouragement of a healthier lifestyle in a well one. The concept of disease is omnipresent in this book, as it is throughout medicine, so I should be explicit about what it means to me, especially since there are critics who believe I am still af¡icted by a nostalgic desire to hunt diseases down and trap them in their lairs— which is partially true. When I see an acutely ill patient and I want to know what’s wrong, I think in disease terms—pneumonia, idiopathic pleuropericarditis, diverticulitis with abscess formation, appendicitis, and so on. They are helpful concepts that contain a large amount of knowledge about etiology, course, prognosis, and treatment. They are always in my mind anytime I hear a patient relating symptoms, in person or on the telephone; they inform my questions and my thoughts. I know that the most common reason for not making the diagnosis of serious disease is the failure to think about it (and not take an adequate history). In the chronically ill, knowledge of disease is the organizing system for understanding

Introduction

[15]

some of what is happening to the patient, as well as, for example, reading, listening to consultants, and keeping the experience of years in convenient categories. When I get the feeling that something is wrong with a patient—and the feeling is not more explicit—ideas about disease underlie my diagnostic actions, conversations with the patient and family, scheduling of visits, and so on. I always know— by now in the marrow of my bones—that a disease is not an object, a thing to be found; it is a process inextricably bound up with the unfolding story of this particular patient. I know that seeing diseases as processes allows me to tap into all the knowledge of pathophysiology that I contain or can discover in the literature or elsewhere. I also know that this process is taking place alongside the processes of the person and (as a shortcut) the context (including me). They exist in relation to one another—person, pathophysiology, and context—and I am trying to discover the relationship so that I can in¡uence its course. These ideas of process and relationship constantly come back to serve me as I try to understand all the states of the human condition to which I am privy. This way of seeing diseases is opposed to understanding them as distinct and exclusive categories, in which the object of diagnosis is to ¤nd the category into which a patient ¤ts (the way rheumatologists used to carry on about which kind of in¡ammatory arthritis a patient had) so that treatment can be chosen on the basis of the category. Still, by the fact of my training, process and relationship start with disease. In clinical medicine, I believe we know as much about the interaction of pathophysiology and persons, or persons themselves, families, their relations to each other (and to doctors), and to social groups as we knew about the body a 1oo years ago. Yet this knowledge is vital to clinical medicine. The following case is illustrative. Many years ago, before modern effective treatments for Hodgkin’s disease existed, I took care of a young woman from age fourteen to her death at age twenty. By the standards of that time, her care was successful. She remained in school, maintained relationships, and spent very little time in the hospital or bedbound at home, although at autopsy she was virtually taken over by lymphoma. The effect of her illness on the family, on the other hand, was devastating. Her father subsequently divorced her mother, survived his own Hodgkin’s disease, and then died in circumstances suggesting that his daughter’s death was a precipitating cause of his death. There also appeared to be profound, long-lasting, destructive effects on the

[16]

doctoring

subsequent lives of the patient’s two siblings. The mother continues to be devoted to the patient’s memory, almost to the exclusion of other interests. Why the patient was able to function so well despite the extent of her disease and why these things happened to her family remain matters for conjecture despite their importance in the care of these patients. I cite this case to point out that primary care medicine is not the end of an old medicine but a new direction in the exploration of the human condition and its relationship to sickness—what clinical medicine has always been about. Primary care medicine will inevitably grow and ¡ourish because the roots of its emergence reach deeply into medicine’s history. The pace of change and the quality of newly trained generalists are in question. Changes in the health care system going on in the public and private sectors will have a crucial role in this evolution. If new systems treat physicians as the solution rather than the problem, rising morale will help spread new ideas. By contrast, health care changes based on older, simplistic views of primary care and the absence of educational support will slow the diffusion process. The agenda of the managed care organizations sweeping the country is impressing itself on the form of medical practice for good or for ill. It is too soon to know what will ultimately happen, but when I am asked whether physicians can meet the standards of practice implied in this book in a ¤fteen-minute visit, I think that is the wrong question. I believe that the medicine described here should be more effective in terms of patient well-being and satisfaction, as well as being cost-effective. The ¤fteen-minute visit is hardly the appropriate criterion. At present, the form and content of doctor visits and the promulgation of practice guidelines are often determined by cost control based on motives other than the best interests of the patient. From this perspective, the physician’s act and physicians themselves can be seen as interchangeable commodities in a marketplace. I worry that this will result in a “dumbing-down” of physicians’ education, reversing the major trend in medical education of this century. There are many obstacles to be overcome, so primary care medicine as envisioned in this book will probably not begin to truly ¡ourish for a decade or more, and the numbers of new training programs will be insuf¤cient. But this is an old profession, and ten years is not a very long time. In a nurturing environment of academic, political, and social interests, new teachers and programs will spring up. In a medical world of struggle between the old and the

Introduction

[17]

new, between a basic concern for the patient versus an interest in pro¤t—which is the most likely scenario—fundamental interest in the care of the patient, the success of new ideas that bring better patient outcomes, and the lower cost of this medicine will gradually force change, cultivate teachers, and bring increasing numbers of students and physicians into the ranks of modern primary care medicine.

[18]

doctoring










1 What Is Primary Care?








F

or almost 200 years, the idea of health has been equated with freedom from disease. Developing slowly during the same period have been the explicit medical and social meanings of the word disease; the science necessary to understand diseases; and the entire scienti¤c, institutional, and social structures dedicated to their treatment. The world of medicine, from this perspective, is a world of disease—peopled by those who have an acute disease, are being prevented from having a disease, are being treated for their disease, or are being rehabilitated from the effects of their disease. Organizing medicine according to disease is a theoretical construct that developed, in its present form, at the beginning of the nineteenth century. The knowledge about disease processes gained since then is extremely important; the theory, however, has become passé. Gradually in this century, the idea of health as freedom from disease has been seen as increasingly inadequate, and new de¤nitions have arisen that emphasize the ability of persons to reach social, emotional, and economic goals despite illness, impairment, and functional limitation. This viewpoint, which has been called an ecological theory (illness arising from disturbances in the biological, psychological, and social relationships of individuals with them[18]

What Is Primary Care?

[19]

selves and others), has both fostered and found expression in the rise of primary health care.

Primary Health Care In Great Britain in 1920, not long after national health insurance was instituted, primary health care (the primary health center) was distinguished from the secondary consultative center and the teaching hospital. The idea that primary health care is the most general, entrylevel, medical care, and that it is to be contrasted with referral centers that contain specialist care and with teaching hospitals, has become widespread throughout the world. Primary health care has been a central mode of medical care in many nations for a long time, providing an international body of varied experience. The concept was further developed by the World Health Organization’s search for health care systems that could advance the social goal of member governments for “the attainment by all citizens of the world by the year 2000 of a level of health that will permit them to lead a socially and economically productive life.” The World Health Assembly, in subsequent deliberations in 1979, de¤ned primary health care as follows: Essential health care based on practical, scienti¤cally sound, and socially acceptable methods and technologies made universally accessible to individuals and families in the community by means acceptable to them and at a cost that the community and the country can afford to maintain at every stage of their development in a spirit of self-reliance and self-determination. It forms an integral part of both the country’s health system of which it is the central function and the main focus of the overall social and economic development of the community. It is the ¤rst level of contact of individuals, the family and the community with the national health system, bringing health care as close as possible to where people live and work and constitutes the ¤rst element of a continuing health care process.1

At ¤rst glance this statement seems to be describing a kind of health care that is ideal for dealing with the most common problems in the poorly de¤ned fashion in which they often occur. As Barbara Star¤eld points out in her book Primary Care, “It addresses the most common problems in the community by providing preventive, curative, and rehabilitative services to maximize health and well-

[20]

doctoring

being. It integrates care when more than one health problem exists, and deals with the context in which illness exists and in¡uences people’s responses to their health problems” (p. 4). Similar concepts mark the statement on the generalist physician from the American Boards of Family Practice and Internal Medicine. Three times, in 1978, 1984, and 1994, the Institute of Medicine of the National Academy of Sciences tackled the problem of the de¤nition of primary health care. In 1978, the Institute stated that “primary care should be accessible, comprehensive, coordinated, continual care delivered by an accountable provider of health services.” Primary health care was seen to be founded, in part, on the fact that most illnesses presenting to health care services were minor and could be handled adequately by caregivers other than physicians. Recommendations for training during graduate and postgraduate medical education recognized the need for changes in priorities, including, for example, epidemiology and behavioral science, and a shift away from the belief that primary health care is less sophisticated medicine. The document recognized the dif¤culties such changes would face. The 1984 report saw primary health care as “a strategy for organizing the health care system as a whole,” which emphasized communitybased health care and deemphasized hospital-based, technologyintensive, acute-care medicine. In 1994, the Institute of Medicine’s Committee on the Future of Primary Care, after careful review of the multitude of de¤nitions by then extant and in recognition “of two important trends: the greater complexity of health care delivery and the greater interdependence of health professionals,” de¤ned primary care (provisionally) as follows: Primary care is the provision of integrated, accessible health care services by clinicians who are accountable for addressing a large majority of personal health care needs, developing a sustained partnership with patients, and practicing in the context of family and community.

They went on to de¤ne in detail each term in the de¤nition and to point out the amount of further work that would be required in their ¤nal report so that, for example, a training program might be based on their conclusions. They based their ¤ndings on several critical assumptions: 1. Primary care will be the logical foundation of an effective health care system because primary care can address a large majority of the health problems present in the population.

What Is Primary Care?

[21]

2. Primary care will be essential to achieving the objectives that together constitute value in health care—quality of care (including achievement of desired health outcomes), patient satisfaction, and the ef¤cient use of resources. 3. Personal interactions, including trust and partnership between patients and clinicians, will remain central to primary care. 4. Primary care will be an important instrument for achieving a stronger emphasis on (a) health promotion and disease prevention and (b) care of the chronically ill, especially the elderly with multiple problems. 5 The trend toward integrated health care delivery systems will continue, and will provide both opportunities and challenges for primary care. In general, it is fair to say that the Institute of Medicine, in its progressive deliberations, moved from seeing primary health care as a way of organizing and delivering health care to an elaboration of the function of primary care medicine (and thus the primary care physician) in a manner that goes beyond the classical understanding of a doctor, a patient, and a disease—but not quite far enough to know what the modern generalist is actually going to do. Over the years and as these de¤nitions have been elaborated, primary health care has taken on a complexity that is a far cry from its original purpose in the British National Health Service. A vital distinction has arisen between primary health care and primary care medicine that should always be kept in mind when considering this subject, much like the distinction between diseases and the individuals, knowledge, and skills for treating them. Primary care medicine and primary care physicians deliver primary health care. Primary health care creates the need for a wide and varied set of functions that can be delivered by practitioners other than physicians—for example by nurses, nurse clinicians, nurse practitioners, and physician assistants. The academic health centers have consistently turned away from this need and the roles it creates until very recently. The imperative is there for medicine to enter, understand, and train. This book is about primary care medicine and the knowledge and skills necessary for its optimal delivery. Because I am a physician, it is about the knowledge and skills that must be taught to primary care physicians. Naturally, I believe that others performing the same functions should have the same knowledge and skills.

[22]

doctoring

Chronic Disease Alongside the changing de¤nition of health and the emerging ecological theory of illness in this century has been the shift to chronic disease as the cause of the major burden of illness on the American (and other Western) population. Medicine in the United States in terms of its institutions, payment mechanisms, goals, practices, technology, training, myths, and symbols is a profession concerned primarily with acute disease. It is a medicine modeled on the treatment of infectious diseases such as pneumonia or meningitis. Primary care medicine also is frequently thought about in terms of criteria developed for an earlier, acute disease era of medicine. This is very strange since even before the 1920s, more people in this country died of chronic than acute disease and injury.2 In our era, perhaps 75 percent of all deaths are attributable to chronic disease. The aged consume the most medical care, and their burden of illness is overwhelmingly attributable to chronic disease. This predominance of chronic illness and disease means that primary care physicians must be trained speci¤cally to deal with chronic illness. So what are the differences between acute and chronic disease, and what are their implications for primary care medicine? The ¤rst and foremost difference is that the basic struggle in chronic disease is not against death; it is against disability. (By disability I mean the inability of a person to perform a social role because of functional loss in contrast to impairment, which is the inability of a part to perform normally because of pathology.3) Since antiquity, the image of the physician has been of a warrior battling against death. The military metaphor has seemed apt for the image of death stealing away the lives of the unsuspecting. Of course, people die from chronic disease, but disability—the loss of function and independence—has always come ¤rst and marked their lives. Keep in mind multiple sclerosis, severe strokes, Tay Sachs disease, many cancers, and Alzheimer’s disease. In considering these diseases personally and professionally, it is not the deaths of these patients that we ¤nd so awful but their lives. In fact, the present era has been marked by struggles over the inappropriateness of keeping many of these persons alive (for which reason living wills and durable powers of attorney for medical care have been developed). Legal battles over the terminal care of the chronically ill have come about because the state has an interest in keeping its citizens alive—long established in

What Is Primary Care?

[23]

law—but only a very recent interest in preventing or treating naturally occurring disability. We have no dif¤culty with the de¤nition of death even if it has to be adjusted to ¤t a technological era. Disability, on the other hand, is often dif¤cult to determine, heterogeneous, and situational. Finally, there has been a natural presumption since antiquity that death is the worst fate. In the modern era we know more awful futures than death, and they are all related to disability. Acute diseases remove people from normal life. When the disease is over, however, they usually return to their previous existence. Pneumonia, severe food poisoning, croup, gallbladder surgery, appendicitis, most fractures, and other trauma are examples. These diseases have a relatively sudden onset, a de¤ned course, and a discrete end. For this reason, the illness can truthfully be considered an event, treated as an event, and paid for as an event. There are exceptions, of course, and acute diseases may have sequelae that have lasting impact on the patient; paralytic poliomyelitis is a striking example. Chronic diseases have an insidious onset, a variable expression throughout their existence, and usually last the lifetime of the patient. They may have acute exacerbations that are often mistakenly considered to be acute disease. Acute myocardial infarction is the classic example of an event often considered and treated like an acute disease by both patients and physicians. It is, however, an episode occurring as part of the life history of arteriosclerotic cardiovascular disease. Chronic disease may exist in the absence of any symptoms, illness, impairment, or disability. Hypertension and adult-onset diabetes are examples. When illness is present in these diseases, it often interdigitates with the person’s life, ¤ltering into every aspect and coloring each decision and act. Although the perspective misses much of their human dimension, it is possible to see acute diseases and their illnesses as purely physical happenings; this is how they have been viewed for much of medicine’s history. Such a perspective is impossible with chronic diseases—to go further than the previous point—because they have an impact on every sphere of life, social and emotional as well as physical. In The Healer’s Art I discussed the effects of sickness on persons.4 I showed how patients become disconnected from others and their environment, lose a sense of their own indestructibility, ¤nd reason and their system of beliefs inadequate to fully comprehend their illness, and feel themselves unable to control events. These characteristics of illness provide much of the basis of the healing

[24]

doctoring

functions of physicians. The same characteristics are found in chronic illness to a variable degree. The success of persons with chronic disease in resolving these problems has much to do with the degree of illness they manifest. Put another way, persons with the same pathology and the same impairment manifest different degrees of illness, depending not only on their situation but also on how they deal with the effects of disease. Illness is distinct from disability. Persons may be ill but not disabled because despite their pathology, impairment, and dysfunction they may ful¤ll their social functions.* The personal effects of illness—for example, the feeling of having no control over events—are distinct from the disease, while the speci¤c disability is always related to the disease. Severe disability may be present with no concurrent illness. For example, people who have had polio in childhood that left them unable to perform many social functions may have no illness—they have adapted to their impairment and its effects—and feel healthy. As they age and experience increasing dysfunction (the post-polio syndrome), illness may begin as they ¤nd themselves unable to explain what is happening, control their lives, or avoid feeling increasingly vulnerable. The effects of the dysfunctions of chronic disease are different for each person. For example, the same degree of motion disorder resulting from Parkinson’s disease may have little impact on a reclusive mathematician but completely disable a politician. Further, the dysfunctions and disabilities of chronic disease are only partly explained by the disease process, just as the extent of pathology may be different in two persons with an identical disease. Unlike the situation in acute disease, this means that basic research into the molecular biology of the mechanisms of disease may produce only limited understanding of a particular person’s illness or disability. This does not mean that the nature of the disease is not important. It is crucial but not central. Since this appears to contradict a cardinal tenet of modern medicine, it is necessary to make it absolutely clear. The situa* The variable expression of the same disease in different persons and the degree to which individual patients have an in¡uence on the amount of disability they experience from the same functional impairment is a major obstacle to research on the outcomes of treatment. On the other hand, these differences suggest that research on treatment must focus not only on the usual modalities such as medication, surgery, and physical and occupational therapy, but on interventions into the person. Psychotherapy, support groups, vocational rehabilitation, and the therapeutic behavior of doctors and other caregivers are examples of such intercession.

What Is Primary Care?

[25]

tion is similar to what happens to people who live in different places. Whether someone lives in Bosnia, Greenland, Charlotte (North Carolina), or New York City is crucial in determining the kind of life they will lead. Their environment (like the kind of chronic disease someone has) sets limits, determines opportunities, and so on. It is the matrix, the framework in which the life will be lived. But the nature of the person is central. John Simon will be John Simon and not Eric Cassell wherever they may live and whatever diseases they have. The next difference between acute and chronic disease will sharpen the point. Acute diseases are generally treated by doctors, nurses, or other caregivers (including family members) in the hospital or outside. Chronic diseases, on the other hand, are usually treated by patients themselves or their parents (generally under the direction of their caregivers). Think of diabetes, ¤brocystic lung disease, coronary heart disease, or chronic obstructive pulmonary disease. Patients either keep their doctor’s appointments, follow their diet, do their exercises, take their medications, and stop smoking or they don’t. They try hard to keep on working or going to school or they retire. All of these things have a profound effect on the course and outcome of the disease, dysfunction, and disability. This means that central to medical consideration are the patients’ motivations, values, purposes, concerns, and relationship to self, others, body, and (maybe above all) the doctor. All these things that in acute disease seemed peripheral have now become central.* We see the dif¤culty for medicine. If chronic disease is overwhelmingly personal, then the person is central. Just as in classical acute disease the disease is central. This means that the body of knowledge of medical science that has served medicine so well in acute disease is only part, albeit a crucial part, of the story in chronic disease. We can also see that if the logical focus of medicine is disease, as developed in the era of the acute diseases, then doctors will miss the boat in treating chronic diseases because they will focus on the wrong thing, as discussed in Chapter 2. This tells us again why the person has become the logical (conceptual) center of medicine. The implication is that the kinds of knowledge that primary care physicians—clinicians in general—require is not only about disease * What makes something central or peripheral? If you can disregard something and still carry out your major function (even if less well), the thing is peripheral. If the disregard of the thing makes it impossible for you to succeed in your major function or achieve your important goals, the thing is central.

[26]

doctoring

and molecular biology but also about persons. As I write this, I can hear all my professors of medicine for the past forty-¤ve years rising in outrage at the thought of abandoning medical science. The issue is not molecular biology or le deluge. The question is simply one of different kinds of knowledge for different aspects of the problem. I hope the examples I give throughout this chapter will make this point. Walsh McDermott understood; he knew thirty years ago, as he examined the patterns of illness and mortality among the Navajo at Many Farms, in Arizona, that medical science and technology were only partly the answer. He called the phenomenon the technological mis¤t.5 Perhaps that is the best way to see the problem of chronic disease as well. There is no medicine without science; it is literally unthinkable. But equally, there is no medicine without patients, and in chronic disease they come to center stage as the particular individuals they are. As we go on to examine what we mean by primary care and the training that primary care physicians require, it is chronic disease we must keep in mind. If we do a good job, the care of all patients will bene¤t. Myrna Paolita is a forty-nine-year-old physical therapist who is changing managed care plans. She considers herself aggressively healthy. She watches her (and her husband’s) diet very carefully, exercises regularly, and believes in preventive medicine. She had repeated episodes of bronchitis in childhood but no asthma. For several years she has used a bronchodilator inhaler once or twice almost every day because of a cough and tightness in her chest. She sometimes wheezes at night. The allergist she went to once tried to get her to use her inhaler regularly and wanted to add a steroid inhaler. She did not do it because she “doesn’t like drugs”; besides, she couldn’t see the point. She has two cats but knows she isn’t allergic to them. Her physical examination was normal as were her screening laboratory tests. Pulmonary function tests revealed moderate obstructive airway changes greater than expected from her history.

This seems a simple enough case. All the woman wants is a prescription for her bronchodilator. She has a chronic disease, however, which she is treating in a manner that will perpetuate it. She seems not to have allowed herself to realize the extent or duration of her dysfunction (commonly called denial). It will take knowledge of the disease and its treatment; awareness of the patient’s contribution to her dysfunction; an understanding of the process of denial and knowing that it should enter her physician’s conscious-

What Is Primary Care?

[27]

ness every time she calls or visits; appreciation of the current vogue about the dangers of “drugs”; and alertness to the need to change her knowledge of the disease and use her appreciation for prevention to help teach her the importance of treating it properly. All of this will take perhaps ten or ¤fteen minutes of conversation at the end of the visit—or perhaps over several visits.

A Mistaken Notion: Primary Care Medicine as Entry-Level Medicine A common error in thinking about primary care is to see it as entrylevel medicine (as for this patient) and, because of this, rudimentary medicine—for mostly (say) the common cold and imaginary illnesses. This is a false notion that is quickly dispelled, as this simple case suggests. If primary care physicians are to separate out patients with serious diseases and pass them on to specialists, then they must be able to recognize serious diseases. They must also recognize their role in directing the specialist to focus not only on the disease problem, but also on the problems posed to the sick person by the disease. This is necessary to avoid the common situations in which consultants refer only to their own notion of the problem, with the result that the goal of the consultation is never realized. Alternatively, they may fail to diagnose the important diseases until their manifestations are obvious and then refer the patient to a specialist. If so, we would not think them very good doctors. Perhaps we want them to refer patients when they don’t know what is the matter. Everyone knows, however, that knowing when you don’t know requires sophisticated knowledge. Then there is, for example, the fact that the list of diseases that start with ¡u-like symptoms stretches from acquired immune de¤ciency syndrome (AIDS) and bagassosis to varicella and Wegener’s granulomatosis. In many of them, such as Lyme disease, inappropriate use of antibiotics does diagnostic and therapeutic mischief. Finally, however, there is the illustrative case of Myrna Paolita, in which knowledge of the disease, the patient, the social setting, appropriate goals of treatment, and the perspective of chronic disease all have an impact on how she should be cared for. This paragraph should be kept in mind when we think about the effects of how we organize medical care; the rising importance of corporate managed care; the impact of varied methods of payment, from fee for

[28]

doctoring

service to capitation; the time allotment for visits; and the training and deployment of doctors. From the perspective of training physicians and the knowledge bases required for adequate performance, the higher we go on the scale of specialist training, the less complex the medical problem becomes. Since this conclusion is the opposite of that usually drawn, some justi¤cation is required. One should not confuse highly technical, even complicated, medical knowledge—special practical knowledge about an unusual disease, treatment (complex chemotherapy, for example), condition, or technology—with the complex, manysided, worldly-wise knowledge we expect of the best physicians. The former can be imparted to technicians, given enough time; the latter is learned through experience resulting from the expectations of patients, colleagues, and the heritage of medicine. The former is a matter of considerable training designed to meet the unexpected and expected within a narrow range of knowledge. The latter is a matter of developing judgment through experience and a grasp of the human condition in matters of sickness and health, with deep roots in a solid foundation of medical science and technology. This is the complexity that the sophisticated specialist is capable of concerning the disease, the patient, and the context—not merely arcane science. We expect this of the best specialists and are disappointed when it is absent. It is not demanded of or taught to primary care physicians and, not surprisingly, it is too often not there—yet it de¤nes their role. Given the demands inherent in de¤nitions of primary care that go further than merely ¤rst-contact medicine, we should expect that training to meet them will be complex, dif¤cult, and based on more than the science of medicine. A generation ago, during the heyday of specialty medicine, there was discussion about solving the problem of the (then) shortage of physicians by training persons to perform some of the simpler functions of physicians—like the feldshers of old Russia or the barefoot doctors of emerging China. A new group of well-trained nurse practitioners with that idea stepped forward, eagerly volunteering to assume the role of primary care practitioner.6 Existing data suggest that, measured by patient satisfaction and outcome appraisals, nurse practitioners have done as well as doctors. There is no reason why these nurses should not have equal standing as primary care practitioners if they accomplish the same ends. In parts of the country they and physician assistants serve in this role, providing much of

What Is Primary Care?

[29]

the medical care. If primary care is merely basic medicine, why not? If it is all a matter of simple diseases, family, social milieu, and community, nurses and physician assistants can unquestionably be as well trained as physicians. So, why not use nurses or physician assistants? I see no reason based on category alone. My quarrel isn’t with the other patient care professions, but rather with an obsolete theory of what care is and what any practitioner needs to know and do. Given the goals of primary care that have developed in the last decade and were discussed in the Introduction, it isn’t simple medicine. Once again, the conceptual trap is thinking of the world of medicine as a world of diseases, especially acute diseases. If you think about chronic conditions—for example, rheumatoid arthritis, multiple sclerosis, mental retardation, diabetes, stroke, cerebral palsy, chronic lung disease (emphysema), schizophrenia, human immunode¤ciency virus (HIV) infection and AIDS, congestive heart failure, certain cancers, blindness—it is clear that their care is not based on just knowledge of disease. Nor will only knowledge of the family, the social world, and the community do. Adequate care of persons with these conditions requires knowledge that starts at the molecular level and reaches the community. It is a knowledge of function and malfunction and their basis in the anatomical-pathoanatomical, physiological-pathophysiological, psychological-pathopsychological, interpersonal-pathointerpersonal, social-pathosocial, and communalpathocommunal nature of persons. This is knowledge of sickness in the community, the society, the interpersonal world of the sick person, the person, and the body. From the perspective of medicine, these understandings can never be separated from knowledge of the body in sickness and health because they are as necessary to comprehend and treat the sick person as to keep the well person healthy. It is the appropriate knowledge of primary care necessary for maintaining or returning function, and it is inherent in the modern de¤nitions of primary care medicine as generalism.

Primary Care Viewed in Contrast to Secondary or Tertiary Care The most common way in which primary care is considered is that it is not secondary or tertiary care. In this sense, a primary care physician is neither a surgeon nor a (disease) specialist, nor a physician

[30]

doctoring

with training as a specialist—for example, a gynecologist—who is not functioning as one. Specialty training in internal medicine is directed at an organ or a category of disease—for example, cardiologists for the heart, hematologists for the blood, oncologists for cancer, and infectious disease specialists for infectious diseases. Specialists have been trained in the basic science and clinical skills speci¤c to the category. Their knowledge is expected to be comprehensive but exclusive. Increasingly, specialists command the expensive and select technology of their specialty—for example, invasive cardiology, gastrointestinal endoscopy, rheumatology’s joint injections, and pulmonology’s bronchoscopics. Specialized medicine has both technical and economic importance because specialists tend to use their technology. When new technology appears in their specialty, they generally employ it beyond previous evidence of its need.* Specialists are not expected, and do not expect themselves, to have knowledge outside of their specialty. When de¤ned by contrast to specialty medicine, primary care medicine (in confusion with primary health care) is an organizational term often used to connote a lesser form of medical practice set up to pass patients on to secondary, tertiary, and quaternary levels of care. It is seldom realized that the hierarchy described in the previous paragraph depicts a social, technical, institutional, and organizational structure developed for acute disease and for the treatment or diagnosis of disease episodes rather than the ongoing care of patients with chronic diseases. Consequently, the patient is ¤tted to the medical structure rather than the other way around. Most commonly, this means that a patient with several problems has several specialists but no one identi¤ed as “the doctor.” The following case may help clarify the problem. Magda Haslik is a forty-seven-year-old mother of six who came to the doctor because she wasn’t sure what to do. Two years ago, because of shortness of breath, she saw a cardiologist who discovered that she had a severe cardiomyopathy whose cause was unclear even after * New technologies also spread because they allow patients to become candidates for procedures that improve their quality of life. For example, elderly patients with severely symptomatic coronary heart disease and other diseases would not survive coronary artery bypass surgery but can be helped by coronary angioplasty. New technologies may ¤nd increased use when they change medical attitudes. An example of this is the increasingly widespread use of laparoscopic techniques in abdominal and gynecological surgery as their ef¤cacy and safety are demonstrated.

What Is Primary Care?

[31]

cardiac catheterization and other studies. She is able to do housework with assistance, but she still does all the cooking. Recently, her liver was found to be enlarged, and she was referred to a gastroenterologist who suggested a liver biopsy. About a year ago, she developed pain in her right shoulder that has become progressively worse. Despite physical therapy, which she attended infrequently, the shoulder has gotten worse and is now frozen. The orthopedist suggested mobilization of the shoulder under anesthesia, with several follow-up treatments under sedation before resuming physical therapy. She says that she is “fed up” with all the doctors, appointments, treatments, and so on. She doesn’t see much point in the process, although she is aware that her heart disease is severe. Her aged father died a year ago, and her mother has heart disease. Examination reveals a woman who does not appear to be ill. Her right shoulder has markedly restricted range of motion, and the muscles around the joint are atrophied. By any measure, her heart is very large and her liver is moderately enlarged (about 8 cm below the costal margin). A chest X-ray reveals a modest (1 cm) increase in the transverse size of her heart compared to the size on a ¤lm taken two years ago. Ultimately, Mrs. Haslik will probably require heart transplantation. When this will be necessary is not clear, in part because the etiology of the cardiomyopathy is not known. This is one reason why the liver biopsy is necessary. Does she have passive congestion of the liver from her heart failure or another disease? If so, her heart disease is progressing more rapidly than if the liver is not congested. The patient has chronic heart disease. Treated in one fashion, she can be a sick woman with a sick heart who will become unable to perform her family functions. Treated in another manner, she will be a well woman with a sick heart who, when the heart is bad enough, will require transplantation to return to function. To be well her shoulder function must be restored. The current problem to be solved concerns her negative attitude. She must be motivated to get all these things done and get back to healthy functioning. For this she requires the ongoing help of a primary care physician, as well as the assistance of her specialists.

Today this kind of patient almost always remains ill; the institutional structure of specialist care ensures it. At best, her case is dif¤cult. She requires expert coordination of care that is possible only if her own doctor is devoted to that goal. It is a task for a primary care physician.

[32]

doctoring

Another Mistaken Notion: Primary Care Medicine as a Gatekeeper Primary care physicians are often thought of as gatekeepers—a strange metaphor for a doctor in a democracy. It suggests the guy in the little house by the gate who looks folks over and then lets the rich in and sends the poor and others to the tradesman’s entrance. What is meant by this term, I suppose, is that primary care physicians, when seen simply as nonspecialists, are ¤rst-contact doctors— the ¤rst doctors people see. This concept is espoused especially by health maintenance organizations or clinic-type medical care organizations, but it could be extended to all medical care under some health care reorganization plans. Patients enrolled in these kinds of insurance plans must see a primary care physician before they may see a specialist or use expensive technology. Most primary care physicians are family physicians, internists, or pediatricians. Some have suggested that gynecologists should also be considered primary care physicians.7 It is often believed that any physician or surgeon, regardless of the specialty, can function as a primary care doctor. As a ¤rst-contact doctor (I hope the word gatekeeper will acquire a progressive and fatal ridiculousness) the primary care doctor performs both care and triage functions. If the problem can be handled by the primary care doctor, then the patient will not be referred on. It is expected that the primary care physician, as the ¤rst-contact doctor, will have more patients than specialists and will see the largest number of patients per unit time. This doctor will order tests and utilize technology if appropriate, not simply because the patient so desires. An important goal of this view of primary care is to save money because of ef¤ciency and denial of costly services. Here the de¤nition of primary care is primarily ¤nancial and political. It has been pointed out repeatedly that there is a con¡ict of interest in feefor-service medicine. Physicians make more money by performing more services, yet they are expected, by the norms of their profession, to subordinate their monetary interests to the good of the patient. A con¡ict of interest also exists in health plans in which the physician bene¤ts when the plan spends less money. Here the doctor makes more money when the patient is denied services. The basic con¡ict between what is in the best interests of the doctor versus the best interest of the patient has always existed. It

What Is Primary Care?

[33]

cannot be abolished by regulation. It is part of the moral basis of medicine and can be effectively dealt with only by moral self-regulation by the physician. The idea may be dif¤cult to accept, but good medical care requires unnatural behavior on the doctor’s part. To take the best care of their patients, doctors must often work longer and harder than they wish, postpone the grati¤cation of their other interests, do without adequate sleep, and even reduce the time spent with their families. Money is only one of many interests of the physician that must be subordinated. Identifying con¡icts of interest is important, but how physicians resolve these con¡icts is the vital issue. The idea of primary care physicians as gatekeepers is problematic, as Barbara Star¤eld makes clear: Thus, the issue of gatekeepers is controversial, and is becoming increasingly so as more types of health care organizations adopt it. There is widespread suspicion that in some plans the basis for gatekeepers is cost-containment rather than rationality of organization. Gatekeepers are increasingly being used to deter utilization of specialists, thus raising the possibility that needed care will be denied. Failure of the gatekeeper to refer might even lead to inappropriate care if the gatekeeper is not suf¤ciently skilled or knowledgeable in the management of the problem. If the purpose of gatekeepers is to provide more rational use of resources, there should be a scienti¤c rationale for the belief that primary care practitioners can ef¤ciently and effectively judge who should be referred to a specialist and who should not. Without a clear rationale for what should be handled in primary care and what should be referred, there will always be suspicion that factors such as costsaving are controlling the decisions.8

Family physicians are not de¤ned as gatekeepers by the nature of their training. Norms of service developed in training, however, may take a back seat to the service-denial and cost-saving functions required by some health care organizations. Physicians in many health plans are so heavily scheduled that they have dif¤culty taking the time necessary to work with patients on emotional or family problems. Thus, when the role of gatekeeper is imposed on a doctor, ideals and standards of good medical practice can become secondary. The primary obligation of the doctor becomes the good of the organization rather than the good of the patient.

[34]

doctoring

Primary Care as First-Contact Care Walsh McDermott, a famous and wise American physician, used to say that it takes a good doctor to know when a good doctor is needed. Illnesses large and small, trivial and fatal, often start the same way— perhaps a little fatigue, achiness, and fever. If they are dismissed as a “virus” or inappropriately treated with antibiotics, disaster can follow. In chronic illness, the past always intrudes heavily on the present and the present determines the future. It makes sense, therefore, that the ¤rst doctor seen by a patient with a new illness should have broad diagnostic and therapeutic abilities. It is also reasonable that patients, both sick and well, receive continuing care from such physicians. Today the training of general internists, family physicians, and pediatricians (for children and adolescents) most closely meets this ideal. It has been suggested, however, that ¤rst-contact care can be provided by any physician acting in that capacity. The idea that ¤rst-contact care does not require special training appears to contain an implicit assumption, as discussed earlier, that patients either have minor problems, easily handled by general internists, family physicians, or pediatricians, or major problems, in which case they ought to be referred to a specialist. Gynecologists, for example, will be expected to take care of upper respiratory illnesses, minor medical dif¤culties, and psychologically based illness. General preventive services, such as monitoring cholesterol and other cardiac risk factors, weight control, and immunizations, should also be within their capabilities. The narrowest subspecialist, the reasoning goes, should also be able to provide this range of medical services. This naive idea arises, as do so many other wrong beliefs about primary care, because of the concept that doctors take care of diseases. Diseases, the idea goes on, form a hierarchy from simple to dif¤cult. Specialists take care of dif¤cult diseases, so, of course, they will naturally do a good job on simple diseases. Wrong. Doctors take care of people, some of whom have diseases and all of whom have some problem. People used to doing complicated things usually do complicated things in simple situations—for example, ordering tests or X-rays when waiting a few days might suf¤ce—thus overtreating people with simple illnesses and overlooking the clues about other problems that might have brought the patient to the doctor. When a thirty-two-year-old man comes into my of¤ce for a physical examination and has no symptoms, the real question is

What Is Primary Care?

[35]

why is he here? Did his wife or family send him? If so, why? Is he worried about something unspoken? Is depression making its entrance as somatic distress? This is not the kind of thing specialists know. Why should they? It isn’t their work. If general internists, family physicians, or pediatricians are not doing a good job as ¤rstcontact physicians, they need better training. If there are too many specialists and it is believed that they should be redirected to primary care, they will require vigorous retraining.

What Is Wrong with These Simple Notions of Primary Care? 1. On any single visit, a patient does not present symptoms of serious versus nonserious illness. One of the moving ¤gures of Kaiser-Permanente in its early days expressed the belief that the organization would work wonderfully if the “worried well” could be kept from clogging the system so that physicians could deal only with real disease. More frankly stated than most, this view expresses a common but super¤cial assumption about ambulatory medical care: that only a minority of patients come to a doctor because they have serious disease—for example, cancer, multiple sclerosis, or pneumonia. These people are truly sick, not merely well people with imaginary symptoms that worry them into thinking that they have cancer, multiple sclerosis, or pneumonia. The majority of patients, the idea continues, are the worried well, those who have psychological problems that are expressed somatically. The problem with this view is that with the exception of far-advanced disease or catastrophic illness, all patients present in the same way. They have perceived discomforts or disturbances of function (symptoms) that they interpret as illness rather than, say, fatigue, overwork, or aging.9 A disease is not a thing that sick people have that can be discovered, while the worried well don’t have such a thing to uncover. All symptoms have a cause, but only some ¤t into currently acceptable categories of disease. It is a basic error to consider diseases as though they were natural categories like oak trees or tent caterpillars. They have no independent existence and can only be found as they are instantiated in a particular patient at a particular time. The idea of diseases as natural, free-standing entities has been extremely useful in simplifying the clinician’s job and in medical research. As cate-

[36]

doctoring

gorized in ICD codes, the concept is also important for health-risk underwriting, risk management and reimbursement, and other bureaucratic functions. Disease categories are less useful for understanding medical care. People with multiple sclerosis do not arrive with a sign on their chest that says “multiple sclerosis.” Instead, they usually present with vague discomforts—for example, tingling of their ¤ngers or the belief that they walk oddly. There are a multitude of reasons for such symptoms, most of which are unimportant or transient. The diagnosis is usually missed, and the patients are dismissed with trivial diagnoses such as “nerves,” overwork, or fatigue until more severe symptoms occur later in the disease. Similarly, early congestive heart failure is heralded by mild shortness of breath, fatigue with effort, or tired legs, but so are many unimportant illnesses. Virtually everyone in our society thinks of colon cancer when bowel habits change, of a brain tumor when headaches persist. Most do not have cancer, but some do. Patients or their parents have no way of knowing which is which, and when they come to the doctor they may all ¤t into the category of the worried well. At the same time, any one of these patients, even those with no symptoms, may have serious disease. Hypertension, diabetes mellitus, and HIV disease, for example, are most often found in asymptomatic people; all three have long periods of latency before their consequences become manifest. 2. Most ¤rst contacts are not ¤rst contacts. First contact generally means the ¤rst visit to a doctor for a particular problem. The ¤rst contact, then, may not be the ¤rst of¤ce visit with a particular provider; often the patient has had experience with the provider and vice versa. This history inevitably in¡uences the nature of any further contacts, and will in¡uence both the patient’s symptoms and the doctor’s response. For example, patients who ¤nd their medical care organization impersonal or impatient with what are believed to be trivial symptoms will alter their behavior in the future, most commonly waiting until symptoms become more severe or avoiding contact altogether. They will be the patients afraid to bother the doctor with questions about childhood nutrition or worries about growth and development. On her second visit for a checkup more than a year after her ¤rst visit, the married young patient asked, “Are you supposed to have pleasure when you have sex?” After the discussion was over, she said that she had wanted to talk about it on her ¤rst visit, but that a

What Is Primary Care?

[37]

visiting doctor was observing me. This embarrassed her, so she put off the question.

The nature of the encounter also in¡uences how patients follow up on the advice or treatment provided. Changes in health-related behaviors also depend on personality, family, and other social factors, as well as the environment in which medical care is delivered. In general, the concept of ¤rst-contact care disregards the relationship between patient and provider as well as the in¡uence of that relationship on the accuracy and timeliness of the diagnosis, the effectiveness of treatment, cooperation with treatment programs, and continued utilization of services. The well-being of patients is related not only to their diseases but also to the adequacy of care, which is strongly in¡uenced by relationships between patients and providers. 3. An illness is not an event; it is part of a process. Although an understanding of the differences between acute and chronic diseases is important in every part of this discussion, it is most important here. As noted earlier, the infectious diseases provide the model of disease that still underlies a great deal of thinking about illness and health care. Infectious diseases—for example, strep throat, pneumonia, urinary tract infections, and gonorrhea—may occur in a previously healthy person. They have a distinct onset, a time-limited course, and a discrete ending. They are strongly in¡uenced by the timing and adequacy of treatment. Many common infections, presumably viral in origin, cannot be de¤nitively diagnosed and are not changed by treatment. Additionally, virtually none of the important medical diseases that lead patients to seek medical care, that cause disability, and that consume a primary care doctor’s time ¤t the pattern of the classic infectious diseases, including AIDS, Lyme disease, and poliomyelitis. None of the chronic diseases ¤t the model on which medical care is too often constructed. Rather than seeing patient care as a continuum wherever it takes place, the infectious disease model encourages the idea that each ¤rst contact with a physician is an event. In this view, each visit is like a knot on a string and it is the knots that count. Clearly, it is the string, knots and all, that is important and that provides the measure of success. Just as most ¤rst contacts are not ¤rst visits, every visit is a prelude to the next. The future illness experience of a patient, con¤-

[38]

doctoring

dence in and approach to medical care, future cooperation with treatment, and understanding of good health behavior are to some degree conditioned by each visit. All of these factors in¡uence the utilization and cost of medical care. Seeing each visit as an event, rather than as a moment in the continuum of care, directly contradicts the fundamental assumption of preventive medicine—that appropriate action in the present can prevent or reduce illness and disability in the future. 4. A ¤rst contact is not a commodity to the patient or doctor. In health care planning it is natural that each service might be seen as a commodity or product. The calculus involved in determining the cost of providing the service, the factors affecting reimbursement, the required number of such services, and other factors all promote the commodity view. It is also not surprising that certain high-technology services such as bone marrow or other organ transplantations are often thought of in industrial terms. In the center is the specialty provider, while in concentrically larger rings are the specialists, primary care doctors, and health centers all feeding eligible patients to the center. These ideas should not be carried over into thinking about the nature of medical care. It is perfectly acceptable to use different perspectives on the same thing to serve different purposes. Medical care—in all of medicine, not just primary care—is a human interaction between patient and doctor within a context and in a social system. As such it is not a commodity. 5. Not all doctors are the same. The idea that any doctor can provide ¤rst-contact care arises from a simplistic view of the diagnostic and therapeutic process. It requires clinical skill to diagnose at an early stage, and the correct diagnosis often makes an important difference in the outcome of the illness, if not necessarily of the disease. The skills required are primarily the ability to take a detailed history and perform a careful physical examination—precisely those abilities that have lapsed in recent medical school graduates.10 These are the central skills of medicine on which everything else is built. They take time and careful attention, and yet they are often brushed aside as inef¤cient or disparaged by patients who have been conditioned to think that the tests make the diagnosis. An alternative diagnostic strategy is to do a magnetic resonance imaging (MRI) scan on every patient with neurological symptoms, cardiac studies on all who are short of breath, or colonoscopy on every patient with bowel symptoms, or to refer

What Is Primary Care?

[39]

these patients to appropriate specialists in the futile pursuit of certainty. This approach is one obvious source of the excessive cost of medical care. It also creates a lot of trouble for patients when they are subjected to unnecessary diagnostic studies, or when incidental ¤ndings or diagnostic errors lead to further unnecessary tests and visits. All diagnostic effort starts with thinking of the possibilities and then employing clinical logic to ¤nd evidence for or against the alternative.11 Training in these methods varies from specialty to specialty and is often not explicit. The “diagnostic reference library” that physicians hold in their heads—especially early in their careers—is different in different specialities. Not surprisingly, gynecologists know better what questions to ask about gynecological disease than about neurological or cardiac conditions. Family physicians are better at looking for family determinants of an illness than are neurologists, and so on. In general, specialists cannot provide adequate primary care. Their hospital-based training exposes them to different stages and spectrums of diseases. As a consequence, their interests, approach to patients, and way of framing and acting on diagnostic questions are different from those of, say, general internists. The opposite problem is that if seeing ¤rst-contact patients keeps specialists from taking care of an adequate number of patients in their own specialty or doing suf¤cient numbers of procedures, they lose their edge. Specialists maintain their expertise by constant exposure to patients with problems addressed by their specialty and the performance of procedures that are part of the training in that specialty. Surgeons do not simply do different things; they generally think about diseases and their treatment in ways different from those of physicians. They are unused to long-term illness, often impatient with or unsophisticated in handling emotional problems, and in general act more quickly than their nonsurgical colleagues. The problem of maintenance of expertise among surgeons is similar to that of other specialists. It has been repeatedly demonstrated that superior surgical outcomes are associated with greater operative experience.

Who Should Deliver Primary Care? As the preceding paragraph suggests, while virtually all physicians sometimes provide primary care, the training of some may better

[40]

doctoring

suit them for the task than others. Internists, family physicians, pediatricians, and obstetrician/gynecologists have been suggested as most appropriate. The traditions of the ¤rst three specialties seem appropriate. Obstetrician/gynecologists have been suggested because they might be additionally trained to meet the general health needs of women. It might seem natural, in the present social climate, to suggest that the health needs of women are so special that they require a special class of physicians. This would mean a drastic change in the orientation of obstetrician/gynecologists, whose primary training is surgical. While such a change is possible, it appears simpler to train internists and family physicians to be cognizant of the special problems of women and to know when to refer their patients for a gynecological problem beyond their expertise. Good new textbooks have appeared that speci¤cally address women’s health problems.12 In the discussion of the speci¤c ¤elds of primary care that follows, it is evident that they have many similarities. Their borders are not distinct in the descriptions themselves and in the work of individual physicians, who may change their approach as their experience draws them in one direction or another. The urban–rural differences determining which kind of generalist predominates have been mentioned, but there are also regional preferences for one type of practice over another. It is vital to remember, however, the unity in the differences. The skills and knowledge discussed in the next several chapters are common to all generalists, as well as to many specialists. It is important to stress that the collegiality that should be present among those who deliver primary care medicine extends to nurses and physician assistants as well as to other caregivers. The ¤eld of family practice has evolved from a deeper understanding of the nature of primary care, rather than in response to current political and economic trends. For this reason, its practitioners have a better grasp of its mission and of the training necessary for a family physician. The range of skills, including surgery, pediatrics, and obstetrics, ideally suits the family physician to rural or underserved areas. Unfortunately, family practice training programs, despite their clearer self-de¤nition, until recently have not grown as much as the leaders in the ¤eld anticipated, at least in part because of society’s longstanding overcommitment to specialized, technologically driven, disease-oriented care. Family physicians are trained in a broad range of medical skills, so that they have greater breadth than

What Is Primary Care?

[41]

depth of knowledge. Their patients are of all ages and are both well and ill, but the typical illnesses they encounter are less well differentiated—less often “name” diseases—than those of general internists or general pediatricians. They have surgical knowledge, and while they do not necessarily hold the knife, they are comfortable with surgical diagnosis and care. They do not exclude other specialty knowledge such as that of dermatology or of the ear, nose, and throat. They are involved in the community, and in public health and outreach efforts such as community programs or senior citizen centers, more than other physicians. This aspect of family medicine is growing in importance, but the specialized knowledge that will be necessary to speci¤cally train family physicians to make good on their commitment to communities has not yet been developed. Pediatricians handle all the illness and health problems of children and adolescents. Like other primary care doctors, they focus on the child rather than the disease. By choice and necessity, they are knowledgeable about parenting and about family issues that have an impact on children. Depth of knowledge about children, particularly their growth and development, is characteristic of the specialty. And there are regional differences in the United States that in¡uence whether children will be taken care of by pediatricians or family physicians. The idea of the general internist has undergone profound changes over the years. Well into the 1960s, general internists conceived of themselves as highly prestigious physicians who were diagnosticians and consultants, the primary physicians to the sick. The rise in subspecialization and high-technology medicine toppled them from their pedestal, but many general internists became defensive about this result and fought back. It used to be that the orthopedist queried an internist about abnormal liver enzymes in a patient about to be operated on. And the internist’s okay was enough to go ahead with the surgery. Nowadays you might hear, “Are you a gastroenterologist?” “No, an internist?” “Oh, just an internist?” With the advent of the primary care revolution, internists are now being advanced as the ideal gatekeeper and triage of¤cer, but “some would consider this a fall from grace.”13 The development of programs for training general internists in primary care, on the other hand, has been ongoing since the 197os.14 These programs have improved care and learning outside the hospital and have made important contributions to faculty development,

[42]

doctoring

curricula, and new knowledge about and for the practice of primary care medicine.15 Successful academic fellowship programs in general internal medicine have also been developed, a necessary step to providing suf¤cient faculty to further the ¤eld.16 Discussion continues, however, about what general internists are and the best method of training them,17 and general internal medicine has been a concern of the American College of Physicians.18 The uncertainties about the nature of general internists, and the emphasis on the place of training rather than on the conceptual content of training, re¡ect a failure of doctrinal cohesiveness in medicine. This problem will not be solved, I believe, until the more fundamental questions about the function and nature of primary care medicine are resolved. What appears to be evolving is the idea of general internists as physicians whose practice is marked by the diagnosis and management of adult illness that does not require surgery. Their expertise is in the recognition, treatment, and management of complex multisystemic illness beyond the capabilities of the individual internal medicine subspecialists. Their distinctive know-how is not de¤ned by a body system, but rather by the knowledge, practice, and integration of a complex set of skills including medical interviewing, physical examination, clinical judgment, clinical pharmacology, interpretation of laboratory and other special studies, the process of health care delivery for their patients (including access to nonmedical providers), and the optimal use of resources. It is the depth of knowledge and exclusive experience with adults that makes the general internist ideal for cross-disciplinary consultation with other practitioners in medical, surgical, and psychiatric ¤elds. I believe the failure of primary care programs to attract suf¤cient numbers of applicants, which existed until recently, re¡ects these more basic issues rather than merely problems of money—important as they may have been until recently.19 The new surge in student interest and the rising number of applications for training programs in family practice, general pediatrics, and internal medicine suggests student awareness of the economic realities favoring primary care.

The Heavy Hand of the Past

[43]










2 The Heavy Hand of the Past Thinking About Diseases Versus Thinking About Persons










M

any of the functions we want primary care physicians to perform do not logically follow from or are contradicted by medical science as it is still taught to students and house of¤cers. Physicians and students have a broad knowledge of the behavior and beliefs of their patients. They must, because everybody has a great deal of information about other people that is necessary to make it through everyday life. Why, in an era of the rising importance of patients, does this knowledge not ¤nd its way into practice? Why has knowledge of persons not won the ¤eld? Because physicians have great dif¤culty discovering the necessary information about the sick person and entering it into the calculus of their medical judgments so that it has equal weight with information about disease, pathophysiology, and technology.

Educational Problems to Be Solved Here is the ¤rst set of educational problems to be solved. They will not be resolved, I believe, by merely reiterating the principles of primary care or presenting examples of how this psychological factor or that social element had an impact on the patient’s illness—the [43]

[44]

doctoring

point of too many case conferences or bedside exercises. This is like assuming that if you tell the blind often enough the path they must follow, they will be able to navigate without vision. Demonstrating for students the effect on a person’s illness of, for example, a life of defeat in an unforgiving world, is an important ¤rst step. It does not, however, tell the student how this information is to be used in caring for a particular patient. This knowledge is very different from telling them that nitroglycerin exhibits tachyphylaxis, after which they know how to modify their use of the drug. Physicians have a body of knowledge and a way of thinking into which an understanding of tachyphylaxis ¤ts logically. Training must speci¤cally teach how information about persons is acquired and integrated into decision making so that they have a structure into which new knowledge ¤ts as easily as the fact of the tachyphylaxis of nitrates ¤ts in its system. Two barriers exist, I believe, thwarting the best intentions of physicians and their teachers to put into practice the foundational ideas of primary care. The ¤rst obstacle is the con¡ict between the kind of knowledge by which physicians know disease—the science of medicine—and the kind of knowledge by which they know and act on their patients as particular individuals. The con¡ict is manifest in the world of medicine, where the scienti¤c ideal of knowledge— objective and measurable—disparages that which is subjective or nonmeasurable. Diseases are best known by the largely measurable, objective, discursive facts of medical science. Persons are known in part by subjective, nondiscursive, representational means. In practice, both kinds of information, objective and subjective, mix and mingle to produce a picture of this person’s illness and form a basis for doctors’ actions. Despite this fact, contemporary medicine values objectivity and belittles subjectivity. The society at large reinforces medicine’s bias because it loves facts, and scienti¤c facts most of all. As a culture, we have collectively become Dragnet’s Sergeant Joe Friday, asking for “the facts, just the facts.” The unselective embrace of objectivity and the disparagement of subjectivity often thwart attempts by doctors to put the basic ideas of primary care into practice. The second barrier to many of the goals of primary care exists within physicians themselves because different, even contradictory, kinds of thought are required of them when thinking about the science of medicine and thinking about the individual patient. Further, the ideal image of the scienti¤c physician who knows

The Heavy Hand of the Past

[45]

disease and the science of medicine is of a doctor dedicated to the objective knowledge of patients and their diseases, someone who removes himself or herself as an individual and a subjective force from the medical equation. By contrast, the ideal image of the caring, compassionate physician is someone immersed in the personal problems of patients’ illnesses. You cannot be absorbed in the patient’s personal problems and at the same time be removed. Knowledge of a particular patient is, in fact, necessarily the exact opposite of scienti¤c knowledge. The more immediate the perceptual and intuitional information ¡owing from the patient, the truer is the knowledge of that patient. Any abstraction produces an inaccurate picture. Premature judgments, categorizations, preconceptions, biases, and stereotypes are misleading abstractions. Complete knowledge of the person is impossible. Individuum est ineffabile: the person is unknowable. The only instrument that can come close to knowing a person is another person—in our instance, a physician. Doctors, as doctors, must become their own instruments; it is their subjectivity that seeks out the subjective nature and responses of their patients. Remember the slogan of two generations ago, “Treat the patient as a person”? Keep in mind the parallel statement, “Teach the doctor as a person.” The central agency of primary care medicine is a person who is a doctor. Ages have been spent learning to teach doctors how to apply medical science to disease. Our task is to learn how to teach doctors how to apply themselves to the care of the sick. The physician comes to know the patient through listening to what is said and unsaid; seeing what is manifest and not manifest in the patient’s ongoing presentation to the world; feeling with the examining hands what the body has to tell; and, ¤nally, being aware of feeling and emotion. This ongoing process (it is not a static event) produces information that, like all information, is true—approximates reality— within levels of probability. Consequently, it is the physician’s job to constantly assess how accurate the information is. Here, too, there are differences between the ideal of truth as the objective facts produced by medical science and the ideal information about a person as a person produced by a physician’s subjectivity. Whatever teaching methods are chosen must prepare physicians not only to use themselves as diagnostic and therapeutic instruments, but also to be enhanced in the process.

[46]

doctoring

Twentieth-Century Error It used to be said that medicine is an ennobling profession. One doesn’t hear that often today, but it is not because the task of medicine has changed. It is because we have fooled ourselves into believing that in the face of medicine’s current technological effectiveness, doctors as persons (who may or may not be ennobled by their work) are no longer a crucial part of medicine. For example, in a recent editorial, a well-known and respected editor of a widely read medical journal discusses what he believes will be the next transformation in the delivery of health care. “On-line, computer-assisted communication between patients and medical data bases and between patients and physicians,” he states, “promises to replace a substantial amount of the care now delivered in person.” Examples given of the things patients might accomplish this way are changes in the dosage of insulin, anticoagulants, antihypertensive drugs, or diuretics; management of many childhood diarrheas; and decisions about when during labor a woman should go to the birthing center. There might be problems, the editor acknowledges, arising from con¡icting reports, inaccurate information, varied opinions, and contradictory information. Further, “Some people will object to receiving some of their care this way. They will still want to hear a human voice on the telephone even though it takes hours to make contact. No doubt, something important will be lost if physicians no longer see patients in person for everyday problems. The ‘laying on of hands’ seems to have therapeutic value.”1 This suggests not merely that computers and information systems might happily create more knowledgeable patients who are active participants in their own care, but also that they will render the physician unnecessary; the information itself will do the work of doctors. The belief that medicine involves the application of impersonal facts to an objective problem that can be seen separately from the person who has it is the cardinal and emblematic error of twentieth-century medicine. The denial of its truth can be seen in any patient with chronic illness, in the family of any child with leukemia, in any aged person and in all nursing homes, in persons with disabilities, and among the poor, sick, or well. The impetus to the development of primary care medicine arises in part because of this mistake. The training of primary care physicians must not participate in or further the same error. This does not deny

The Heavy Hand of the Past

[47]

the importance and the challenge of using information services more effectively. There is a further problem to be solved. We like to believe that doctors gather evidence and reason from the facts and from their knowledge of medical science to a decision that is best for a particular patient. To the contrary. The modes of thought in which physicians are trained and that govern much of their behavior are based on virtually automatic skills and rules that are preprogrammed and learned early in their training.2 These skills and rules concern disease and technology, not persons. Doctors have not been trained to acquire similar rules and skills about persons. Until they are, their best intentions in treating a sick person may be overridden by an ingrained automatic rule.

Historical Origins of Our Dif¤culties I believe it is necessary to know the historical roots of the barriers to be overcome in training primary care physicians. Why is this necessary? If we know what the problems are, why bother with the history of our current beliefs or the theoretical underpinnings of contemporary medicine? Physicians and medical educators (and policymakers) are characteristically impatient with anything that smacks of theory or philosophy. Doctors claim that their intolerance of theory arises from the urgency of the practical issues that confront them. They simply have no time to ¤ddle with philosophy. They are like the patient who says, “I don’t need psychotherapy; I know what my problems are.” That is not the issue. It is how to solve the problems that counts, and to do that, you must know the underlying reasons. Like it or not, physicians and everybody else awaken every morning with philosophy ¤rmly in command of their thinking; they have ideas about how the world works. It doesn’t matter, most of the time, that these beliefs are largely unexamined because they work. In times of transition or fundamental change, however, older assumptions may become a hindrance to effective education or good medicine. We are in such an era now. Greater emphasis on primary care in American medicine and changes in training programs do not have to wait until new ideas about science and medicine evolve. They do require an understanding of the role played by foundational assumptions of current

[48]

doctoring

medical science and practice. A central example is the presupposition that diseases are objects of nature like (say) snow¡akes, African violets, or genes. It is dif¤cult to overestimate the brilliance of the Paris physicians of the 1830s, the discoverers of scienti¤c medicine. An example is Theophile Rene Laennec, who was able to see one disease, tuberculosis, despite its diverse manifestations and anatomical distribution, before the discovery of the tubercle bacillus. The method of these physicians was empiricism at its best. They brought knowledge of their patients’ symptoms and ¤ndings on physical examination into the autopsy room. The results of the autopsy could then be used to determine the cause of the patient’s sickness. By establishing disease as the biological entity responsible for illness— the myriad details the sick person presents to the physician—these initial discoverers swept away layers of confusing social, psychological, and personal issues as irrelevant to scienti¤c medicine. As a lesson for our time, the discoveries based on a careful clinical method preceded the development of disease theory. The diseases were real; that the illness is entirely explained by the disease is theory. For more than a century, disease has had logical—conceptual— status as a biological entity within the system of medical science. It is, in fact, the logical focus of the system. This means that in thinking about medicine, one always ultimately gravitates to the concept of disease. A system, in the sense of the system of medical science, is a group of related ideas and beliefs that circumscribe and contain within them all the pertinent aspects of the system being considered. The word model, as in medical model, is commonly used to express the same idea. In a coherent system, all statements about its parts should necessarily—logically—follow from one another. Whatever else one might say about illness, pointing out possible psychological or social elements, will remain peripheral to medical thinking because it lies outside the system of medical science. Working within the scienti¤c model of medicine, the social, psychological, and personal elements found in all illnesses do not logically follow from considerations of disease because of a fundamental assumption of the medical system: since diseases are biological entities, they are part of nature. From this assumption, it follows that they can be understood and investigated as material things—matter—just like the rest of nature. The legacy of the originators of modern science in seventeenth-century Europe is an understanding of nature and its parts as mechanisms. The function of living organisms, especially

The Heavy Hand of the Past

[49]

mammals, is of particular interest. Medical science, therefore, can examine the diseased body as a mechanism gone wrong. From this perspective, pathophysiology is independent of culture, consciousness, the meaning of things, morality, or spirit. This notion excludes thinking logically about social, psychological, ethical, or personal issues within the same system—at the same time—as the mechanistic elements of disease. The point is made by the fact that when George Engel, an internist at the University of Rochester, who was an early contributor to understanding the place of emotional factors in physical illness, appealed in 1977 for an understanding of disease as a biopsychosocial entity, his paper was widely read. It is frequently quoted still, but it has had distressingly little impact on practice.3 When the disease is seen as an entity of nature, then the methods of science can be used to understand it. A fundamental feature of science is its ability to create abstractions. In the best science, a conceptual model is built that provides an understanding of the thing in question, from the hemoglobin of sickle cell anemia or the genetic determinants of breast cancer all the way to the origin of the solar system. The model is a complete abstraction from the naturally occurring object. The adequacy of the model—its approximation to the mechanism of the naturally occurring object—is tested and re¤ned by the well-known processes of forming hypotheses, designing experiments to test them, and gathering data. It is these steps that most of us think of when we think about what science is and how it works. In considering the work of clinicians, however, it is the fact that diseases are considered objects of nature and the process of abstraction that matters most because it is the effect of these two factors on our consideration of illness—the occurrence of diseases in particular persons—that must be taken into account. Thus, when clinicians consider individual patients with pneumonia, they are apt to think of pneumonias in general—the way they learned about them—rather than of this person’s pneumonia. Clinicians commonly fail to repersonify their ideas of disease, staying instead within the abstract concepts of medical science.

Other Obstacles to Change There are also reductionist and atomistic assumptions that underlie medical science and that have become a central element of the thinking

[50]

doctoring

of all Western physicians. In fact, they are a familiar ingredient of the thinking of almost everyone in the Western world. The reductionist presumption is that a whole is best understood by reducing it to its parts and examining them separately. The parts determine the whole, in this way of thinking, not the reverse. Each part of a biological system, the atomist assumption holds, can be understood separately from any other part or from the whole. There are two further atomistic assumptions: that the body and its diseases can be understood apart from its context—Paris, the Russian steppes, East Harlem, an igloo, or the New York Hospital—and that time is irrelevant because clock and calendar time are not part of science’s understanding of nature.4 Because reductionist atomistic science seeks not merely to understand (say) the liver or the genes but also to comprehend humans, it concludes that when you know the parts, you will know the whole. It does not require much re¡ection to become aware of the fallacy of these assumptions in the care of patients.5 We all know the dif¤culty of predicting the behavior of people whom we know very well. The whole, it is commonly said in speaking of humans, is greater than the sum of its parts. Some of the disappointment that a century of the behavioral sciences—particularly psychology—has failed to teach more about everyday human behavior comes from their modeling themselves on these classical methods of science. The temporal and physical contexts have great in¡uence on the presentation, course, and outcome of disease because they determine the environmental and cultural in¡uences on the behavior of illness, and of doctors and patients. This is why historical controls are no longer employed in clinical research. You may object that I have set the problem in terms that are too stark, that physicians do include nonbiological ideas along with pathophysiology when thinking about patients. People often entertain contradictory notions simultaneously in order to solve practical problems; there is, after all, no way of excluding the fact that sickness has social manifestations. In moments of con¡ict or uncertainty, when decisions are dif¤cult, physicians, like others, usually fall back on the path of least resistance: logical coherence. Training programs can be devised, research grants proposed, and institutions planned that focus on the sick person, but undermining these are the more basic concepts that rule the roost of legitimacy hidden in the minds of students. It cannot be otherwise. The major contribution of late-twentieth-century medical science is a growing literature on the

The Heavy Hand of the Past

[51]

molecular mechanisms of disease. So, it should not be surprising that many physicians have dif¤culty factoring what they know of their patients into their medical thinking.

The Person as the Logical Center of Medicine Overcoming the logical obstacles to thinking about sick or well persons that exist within the minds of physicians-in-training, then, is one of the basic problems faced by training programs. The solution to the problem is obviously the place from which we started. The sick or well person must become the logical—conceptual—center of the system of medicine, displacing disease. The center of the system of this medicine is the patient, not the disease; thinking about this kind of medicine should always bring the mind to the patient. The patient is a conceptual center that reaches up toward the community and down to his or her molecules; everything is logically related to this patient. For a medical scientist, or perhaps a biologist, the patient is a container of objects of nature ranging from diseases to molecules. And in terms of these objects of nature, the patient is logically related to, for example, tissue cultures, DNA from any source, and heart-lung preparations. For a primary care physician, a patient is an exemplar of humankind, logically related to all the dimensions contained within the idea of being human—including the idea of diseases. The clinician’s understanding of the patient is in stark contrast to the medical science idea of the patient as a container of objects of nature such as cells, organs, or diseases. Persons are not like other objects of nature with which physicians are familiar, such as genes or mitochondria. The latter are physically boundaried objects that do not (for science) exist in time. A patient is more aptly described as a history6 or as a historic route of a society of personal moments.7 None of these descriptions deny the biological or the pathophysiological; that is impossible or, worse, silly. For clinicians, the issue is not either/or. Is a patient primarily a part of nature or primarily a history? Patients are historical in the sense that what they are at each moment includes both the past and the future, and neither past nor future excludes the body and what has happened to it or may happen. Patients are not historical in some passive sense, the recipients of events around or within them over which they have no control. They are active participants, adding

[52]

doctoring

their own purposes, small and large, to each moment. This is the reason that historical beings include the future in each moment, because the future, although uncertain, arises out of each moment, following from the person’s purposes and actions. What happens to the body in the past or in the future cannot be exclusive of or separate from the multiple other dimensions of a person. Patients, persons, are of a piece; what happens to one part of them happens to all of them. It cannot be otherwise. There is nothing psychological that is not also physical, nothing physical that is not ultimately psychological, nothing social that is not physical, and so on. When doctors see patients, we want them no longer to look primarily for the disease or its absence to explain why patients present as they do. Put another way, the automatic answer to the question “What is the matter with this patient?” should not be “What disease or pathophysiological or pathoanatomical entity exists here?” In fact, we would like clinicians to remember that their primary function is not to ask questions but to answer them. Clinical logic is the logic of a chain of questions and answers following from the nature of illness as a process, whereby each step inside or outside the body offers an opportunity for intervention. Everything a clinician does attempts to answer the patient’s question “So-and-so is not right with me; what is the matter?” The answer “You have suchand-such a disease” is inadequate and incomplete. The name of the disease should be merely a shortcut for thinking about the complex process within and outside the person and body that has led to the present state. Patients commonly show up with complaints framed in diagnostic language: “I have bronchitis” rather than “I have a cough and phlegm” or “I think I have an ulcer” rather than a description of discomforts or upsets that may or may not be abdominal. Medicine is always part of its larger culture; as a result, its language and system of answers pervade the thinking of patients. The disease orientation of patients is encouraged by the pervasive focus on disease in the media, which is, naturally, behind the times. The fact that patients employ the language of disease should not fool us into making the disease the logical center of our work. Years ago, H. Tristram Engelhardt said: I will thus be in disagreement with the notion that the world of illness, the world of the clinician, can be deduced from or reduced to the world of the pathoanatomist. I will, in fact, be contending that the reverse is the case—the world of the pathoanatomist, the pathophysi-

The Heavy Hand of the Past

[53]

ologist, and the pathopsychologist is always dependent upon that of the clinician for its sense and direction. Further, the world of the clinician is de¤ned by people—by their complaints and vexations.8

The patient is the logical center when the physician tries to de¤ne in the patient’s terms the nature of the loss of function, trouble, complaint, or vexation before beginning an exploration in (current) doctor’s terms. Years ago, medical students were told to start each history with the patient’s “chief complaint” in the patient’s own words. Over the years this rule has lost all meaning; now the “chief complaint” is virtually always some phrase with pathoanatomical or pathophysiological overtones. Doctors should be looking for what the patient thinks has gone wrong, why it is considered threatening (to whatever degree), why it has happened now, and what would be considered an adequate solution. At best, the chief complaint was never meant to contain all these questions. If one brings to mind only acute illness, most of what I have written seems unnecessary fanciness. A cough is a cough, fever is fever, and so are broken bones, acute abdominal pain, bleeding, and all the other symptoms of acute illness. In an emergency room view of medicine, acute illness rules the roost and the disease remains the center of the process. In the hospital and in the doctor’s of¤ce, however, the acute illnesses of times past are not the major problem; instead, chronic illness and its crises reign supreme.

Dif¤cult Cases A case may make the point: Irene Stor is a ¤fty-two-year-old childless woman with insulindependent diabetes since age ¤fteen. She has been married for twenty years to Marvin, an accountant, who is her major caretaker. Despite her diabetes and the considerable dif¤culties with its control in her early years, Irene completed college as a biology major and worked for a few years as an occupational therapist. In her thirties, she began to develop diabetic retinopathy, endangering her eyesight. Her life became consumed by fear of blindness that pushed everything else aside. Because her diabetes was dif¤cult to manage, she went to various diabetologists and internists, seeking a regimen she could live with and to avoid blindness. In addition to diabetic ophthalmopathy, she developed sympathetic neuropathy,

[54]

doctoring

and abdominal symptoms became very troublesome. She tried to ¤nd specialists, even if in distant cities, who might help her. Given the problems with her eyes, gastrointestinal tract, and ordinary diabetes control, her life was diabetes. In her early forties she took a volunteer job with a museum, guiding the blind around the natural history exhibits. This grew into a career in which she lectured and published extensively. For the ¤rst time in her life, she was busy all the time and traveling widely. She received considerable recognition for her achievements and ideas. While frequently tired, she was excited by what she was doing. She worried that Marvin was envious of the attention she received and at the fact that she was not as dependent on him as in the past. Although she had less time to see her doctors, her diabetic control did not worsen, nor did her other symptoms increase. After ¤ve years, she decided she was getting too tired to continue and stopped working abruptly. She returned to a life dominated by diabetes: the continuing failure of her eyesight and the diarrhea, abdominal bloating, and postural hypotension from sympathetic neuropathy. In May 1994, she developed bleeding in the left eye. A gas bubble was introduced to maintain pressure against the retina. She was required to lie face down for six weeks. The bleeding was controlled, but the eye was effectively blind. She seemed to lose interest in events, and antidepressants were started, without improvement. In September, she started vomiting. The vomiting could not be controlled and continued almost all day every day thereafter. She remained bedbound at home, vomiting numerous times during the day. She received her medications and nutrition (total parenteral nutrition) through a Hickman catheter, which her husband managed under the physician’s direction. Numerous professionals—for example, visiting nurses, nutritionists, occupational therapists, and social workers— participated in her care at home. Her parents and in-laws were very helpful and supportive. The gastroenterologist had never previously encountered vomiting of this severity and duration, nor had numerous consultants. In February 1995, she decided to kill herself, but she wanted her doctor’s help.

Despite excellent physicians who have remained attentive, Irene Stor represents a failure of medical care. No consideration of this case is possible without understanding diabetes and its complications. She continues to require an impressive amount of technical expertise and patient care resources. Her diabetologist and gastroenterologist are aware that they are not attacking her central

The Heavy Hand of the Past

[55]

problem. The psychiatrist believes that she is depressed; so does everyone else, including her family members. What should be done? Lonely, isolated, without any show of emotion, interests, or purpose, no longer feeling connected to her husband, whom she believes is angry with her, Irene is not merely depressed, she is suffering. (Numb is her term for it.) A physician who had seen her in the past managed to make a connection to her and evoke once again a responsive, purposeful woman. The vomiting stopped almost completely, and she started taking small amounts of nutrition by mouth. The doctor did not contradict her reasons for killing herself or argue with her contention that even if she no longer vomits and can eat again, “something else will happen soon enough.” The pathophysiological dif¤culties of Irene’s diabetes at this time are manageable. There are no pathophysiological reasons why she cannot increase her oral intake to the point of self-suf¤ciency, learn to walk again, and participate in the activities of daily living. As in all complicated cases, a sizable bureaucracy has grown up around her that will interfere with her ability to solve her problems, but this problem, too, can be overcome. Even the reason for committing suicide, that something will just happen again, suggests that suicide is not necessary until something else does happen. This optimistic stance in the face of awful troubles is the hallmark of clinicians. In the case of Irene Stor, however, it is too late. The suffering of the last eight months has already occurred and will never leave her mind. Every intervention, no matter how successful, will occur against that backdrop. She is correct: something else will happen, and her expectation is of more suffering, not more function. This case could only have been changed in a fundamental way years ago. Irene has a chronic disease in which it is inevitable that dif¤culties will occur ten, ¤fteen, twenty, or thirty years later. The speci¤c problems of 1995 might have been dif¤cult to predict in 1960 or 1970, but their general shape was evolving even then. The problems of today are the outcome of the interaction of her person with the disease, modi¤ed by technical medicine and by the changing context of life and medical care. It is necessary to realize that every medical intervention, even in acute disease, will bring about some change in a person’s life in every moment of the future, near or far. Consciousness of this fact may not be necessary as physicians decide what to do in a particular acute disease. On

[56]

doctoring

the other hand, the importance of future change is one of the de¤ning features of chronic disease. As a person is a history, so is a chronic disease; chronic illness is the interaction of the two histories. It isn’t a crystal ball that is wanted, some magical way of predicting the nature of these distant futures. As the history of the chronic illness unfolds, some futures become more probable and others less so. Irene Stor’s diabetic complications are not cardiac or renal; they are ocular and autonomic neuropathy. The autonomic neuropathy has been expressed more fully in gastrointestinal complications than in postural hypotension. She has had ongoing support from her parents. Her marriage, like theirs, has been stable. She has been dependent on her husband over the years and has rarely been separated from him. She has never been adventurous, but she has shown independence of judgment. Her success in the museum venture was a surprise, despite her intelligence. Its ending was not a surprise in view of the changing marital and other family dynamics it brought about. On the other hand, her diabetes and its complications were much less intrusive during the museum interlude. In the interest of her future, her career should have been supported and the family problems resolved early in those years, rather than after it ended. Even now, her intellect, curiosity, and interest in challenges provide a possible way out of her dif¤culties. Are these the jobs of her doctors? Who else? Are the problems psychological? If learning to ride a bicycle is psychological, then these are psychological issues. Are they social, marital, interpersonal, family, physical? If the person is the logical center of medicine, they are medical. What doctors have these skills? Where are they learned? What training programs are necessary to teach them? These are distinct and separate questions from whether they are medical problems. We have located them within the purview of the generalist. This points the way to the necessary training. Some might say that this case is impossible to treat. If hard cases make bad law, as the saying goes, maybe they also make bad medicine. Here is a simpler example. Joy Barker is a ¤fty-two-year-old divorced, childless teacher who has had low back pain since late adolescence following a bicycle accident. She has seen chiropractors, acupuncturists, orthopedists, and an Alexander therapist. None of them helped her except brie¡y. She didn’t stay with any modality very long. She had reactions to virtu-

The Heavy Hand of the Past

[57]

ally every medication that was tried. Inability to tolerate medications has been her experience over the years. In recent months she has had right upper quadrant pain that was extensively investigated by X-ray studies and endoscopy without producing evidence of disease. A physician told her that the pain was referred from her thoracic back, where a particularly tight, tender muscle mass was present. She didn’t believe the doctor but consented to an injection in the trigger point. The procedure did not relieve the pain and was, by her account, extremely painful. She believed the doctor had been “mean.” She recently came to her primary care physician because she had begun to hurt “all over.” Careful questioning revealed that the low back pain was worse than usual. She also had pain on motion in her left shoulder, neck, and left knee. There was no pain in the muscles of the shoulder girdle or hips. She continued to work regularly but feared that she would become disabled and unable to work or care for herself. Careful examination revealed some diminished range of motion in the right shoulder. The paraspinal muscles of the neck were tender, as were the periarticular muscles of the left knee. There was no evidence of in¡ammatory arthritis. The sedimentation rate was normal.

One does not require much experience in patient care before meeting a patient like this. It is likely that no matter how extensive the investigation, this patient will be found to have no speci¤c disease whose treatment will make her feel better. Trials of medications, such as nonsteroidal anti-in¡ammatory agents will probably produce more side effects than relief. Referrals to physical therapists will be dif¤cult because the patient will object on the basis of geography, ¤nances, or personality differences. Nonetheless, most physicians will follow the classic diagnostic (magnetic resonance imaging [MRI] or computed tomography [CT] scans, neurologists, or orthopedists) and therapeutic (nonsteroidal drugs, muscle relaxants, a trial of steroids, and physical therapists) route before giving up in frustration. Doctors have dif¤culty doing otherwise because their current training brings them back to the logical center of the case: the disease process. If there is no disease, they will ultimately withdraw. If the patient had an interesting disease like scleroderma, on the other hand, no matter how troublesome the patient’s behavior, they would probably continue to treat her. This patient’s future will be increasingly marked by pain and disability, unless a regimen for care of her back is soon started. Her present condition could have been predicted years ago, since it is

[58]

doctoring

the common outcome for persons with recurrent back pain who do not start an adequate regimen of exercise and self-care of the back. I chose this example, not only because the patient’s behavioral pattern makes care dif¤cult and thus poses a problem for her doctors, but also because it involves back pain. Back pain—musculoskeletal impairment in general—although very common (the most common cause of work disability)—poorly ¤ts the understanding of disease in which physicians are trained. Musculoskeletal pain often does not have precisely de¤nable anatomical or biochemical characteristics; it is usually not even a disorder of a distinct tissue such as cartilage, bone, joint, nerve, or muscle. Instead, these are disorders of a functional whole—the back, in this instance. Seeking the molecular basis of these problems, the path of virtually all the recent successes in medical research, will not result in understanding. A different fundamental basis of dysfunction is present here. Chronic musculoskeletal disorders, such as garden variety back pain or knee pain, are disorders that arise or are perpetuated because the relative contributions of each part of the functional unit to the whole are out of balance. One disorder may start as a result of muscular strain, another as a consequence of cartilage or disc injury, another secondary to postural imbalance; in yet another case, the precipitating factor may be a gait abnormality. However these disorders start, when they become chronic it is because of distorted relationships of the parts to each other and to the whole, perpetuated by the life pattern of the individual—for example, the job, recreational exercise, relationship to activity in general, emotional factors, and medical care. For this reason, I could have exempli¤ed the problem by a case of recurrent back injury in someone whose occupation—for example, foundry worker—involves major back stress. There again, the problem is not just the patient at this moment but also two, ¤ve, ten, or twenty years hence.

Going Back to Empiricism It is common for patients, as in these examples, to have problems for which the established theory of medicine is inadequate. This has been true throughout the history of medicine. From the time of Hippocrates, medicine has been marked by a struggle between two major ideas, rationalism and empiricism. Rationalism focuses on theories of diseases, their origins in nature, biological mechanisms,

The Heavy Hand of the Past

[59]

and science. It bases the actions of physicians on these theories. Empiricism is concerned with patients: what it is about them and their interaction with their environment and with nature (in the form of the mechanisms of disease) that produces this illness at this time. Empiricism focuses the actions of physicians on patients themselves rather than primarily on their diseases. Hippocrates was an empiricist; Galen was a rationalist; the Paris school of the early nineteenth century, where our modern concepts of disease originated, was empiricist; the last 150 years have been increasingly rationalist; and the dominant forces in current Western medicine are strongly rationalist. Since the 1930s, however, there has been a growing empiricist trend—the patient has become increasingly central. Clinical empiricism is based on the observation of patients and on the re¡ective appreciation of the physician’s experience with illness and the care of patients. Scienti¤c empiricism is the discovery of the facts of nature through experiment. This is so much a part of medicine that many currently believe medicine to be an empirical profession. Clinical and scienti¤c empiricism are distinct. Clinical empiricism is primarily observational, descriptive, and historical. This is Thomas Sydenham’s “plain and open method” and his student John Locke’s “historical, plain method”: “historically fastidious, observationally acute, free of unwarranted conclusion, and plainly argued.”9 As a method, clinical empiricism may be closer to history than to science. It is concerned with the observation of the patient not only at this moment and as the doctor continues to treat the patient, but also with the history of the illness, the story of the patient up to the ¤rst contact with the physician. Like the historian, the doctor can know what happened in the past only through the senses of the participant, the traces of past events still imprinted on the present, and documents that record previous events—test results, reports, operative summaries, and the like. It is not the patient who is the historian but the physician. The patient tells the history (in response to questioning), while the physician elicits, records, and interprets the history based on a critical examination of the source(s) and the particulars elicited from the patient, other records, and historical materials. This may be plain in the observational and descriptive sense, but it is not simple. As I noted in the Introduction, the patient as a sick or well person is in many ways a new focus of interest. Seeing the patient becomes a technical act. (Actually, all of the senses are involved.) When patients come to physicians, they present

[60]

doctoring

thousands of facts in a blur of information. It is the physician, as historian and interpreter, whose perspective winnows the shower of information to determine what will be acknowledged and attended to. Anything that falls outside of this point of view will not be heard, felt, smelled, or seen. We become aware of that for which we have concepts, and our concepts direct our senses. Percepts have what Alfred North Whitehead called causal ef¤cacy; they direct the process of perception moment by moment, thus building up a structure of meaning that constitutes what the percipient “sees.”10 Ernst Cassirer has convincingly demonstrated the symbolic nature of knowledge as it is acquired in everyday life.11 Susanne Langer makes clear that all of these understandings represent a change in our fundamental understanding of knowledge. The clinician is not gathering atomic facts, each separate from the others or from the whole, as in the positivist science model. Rather, the facts are tied together by the meanings in which they are embedded, meanings speci¤c to the patient, the physician, and the culture they hold in common. It is the meaning of objects, relationships, and events that drive the actions of the patient and physician rather than atomic facts.12 This should make it clear that the clinical empiricism that must be developed in order to teach primary care physicians is an extremely sophisticated approach to gathering clinical information. Simply teaching students or house of¤cers to attend to, for example, social or psychological factors will have no lasting effect on practice because the students do not know how to see or hear the information and because it is outside the conceptual structure guiding their thinking and, therefore, their perception. Clearly, there must be a return to careful observational and physical examination skills, as well as ¤nely tuned history taking and communication skills. If the discipline of clinical medicine is related to the discipline of history, then physicians must come to understand what guides historical inquiry. At a minimum, the physician wants to know not only what the patient experienced in the past but also about the patient as an observer. Since perception is symbolic and guided by concepts, then patients’ reports of their experience will be directed by the meanings they attach to events. An adequate history must elicit not only what the patient experienced but also what set of concepts or meanings guided the experience. In this century, the demand on empiricism—that it ¡esh out the inadequacies of theories of medicine—has been met by bedside

The Heavy Hand of the Past

[61]

training. Medical students start on the hospital ¡oor in their third year, and graduate medical education is almost entirely hands-on training. What I believe to be the current requirements of clinical empiricism cannot be met by the didactic methods employed in the past. With a changed understanding of the nature of contemporary clinical problems, the information necessary for clinical action, and the responsibilities of primary care physicians must come changes in educational methods. This is especially true now that hospitalization is less common and of shorter duration than in the past. We now know the educational goal: teach physicians how to discover the facts about the sick or well patient and enter them into the diagnostic and therapeutic processes in equal partnership with information about disease, pathophysiology, and technology. We also know some of the barriers to this goal: (1) the seeming dif¤culty of speaking about scienti¤c facts of disease and the psychological, social, or ethical facts about patients in the same breath; (2) the different kinds of thought necessary for clinical thinking—analytic, reductionist, and discursive for thinking about diseases, synthetic, representational, and nondiscursive for thinking about persons; (3) the con¡ict between subjective and objective knowledge and information; (4) the fundamental difference between the impersonal instruments that discover objective facts and the personal instrument of the individual physician that uncovers subjective information; (5) the automaticity of much medical decision making and action; and ¤nally (6) the failure of many of the illnesses with which primary care physicians deal—for example, musculoskeletal impairments and chronic illness—to ¤t the contemporary ideal of states that are explained by molecular mechanisms. On the face of it, the goal seems straightforward enough but the hurdles seem insurmountable. The object is not to have completely worked out or ¤nished programs tomorrow, which is impossible, but to start down the correct road that will give us something to teach now and ultimately lead to the appropriate development of theory, clinical methods, and didactic programs. There is one further barrier between physicians and their patients, the wonderful array of diagnostic and therapeutic technologies that have changed the face of medicine in this century. We must examine what is special about this problem and what can be done to allow technology to enhance the physician’s appreciation of the patient, as well as of the disease process.

[62]

doctoring










3 The Special Problem of Technology








P

rimary care medicine will never achieve its goals without solving the problem of technology, a thing unique unto itself.1 Technology will confound almost any attempt to change the health care system or redirect its fundamental goals. Further, if there is one thing that can be singled out as the engine of the medical cost in¡ation now occurring everywhere in the world, it is the seemingly irresistible spread of technology into every level of medicine—irresistible to doctors, patients, and nations alike. Evidence that technology is a problem is everywhere in medicine. In intensive care units the world over, the technology of monitoring, organ support, and resuscitation is used where it is appropriate—in relation to the purposes of the sick person. But it is also used where it is inappropriate, de¤ned by the abilities of the technology and the consequent expertise of physicians, rather than—or even contrary to—the best interests of the sick person. Another example is the increasingly common diagnostic parade of one test after another, whether or not they are warranted by the patient’s symptoms or improve diagnostic accuracy. The thesis of this chapter is that, like the broom in The Sorcerer’s Apprentice, technologies come to have a life of their own, not only because of their own properties but also because of certain universal [62]

The Special Problem of Technology

[63]

human traits. Technologies come into being to serve the purposes of their users, but ultimately their users rede¤ne their own goals in terms of the technology. Like the twentieth century, when they have come to dominance, technologies are reductive, oversimplifying, intolerant of ambiguity, and democratic. It would be simple to say that technologies are just things, neutral in themselves, employed solely at the will of their users, but that is not true. They are developed, made, and marketed for purposes meant to match the intentions of their potential users. Thus, their deployment will not change until we understand the singular category they occupy in medicine and speci¤cally train physicians to control as well as employ them. The de¤nition of technology presents problems for which dictionaries are no help because the term can be used so broadly as to defeat understanding. Thus, any tool employed in a craft could be said to be that craft’s technology. In medicine that would label any means a physician employs to achieve diagnostic or therapeutic goals as technology. Taking a history or doing a physical examination would, by that de¤nition, be technologies (by extension from techniques). So, too, would stethoscopes, ophthalmoscopes, and scalpels. These are not the kinds of things that trouble us because they do not divert the purposes of their users to the purposes of the technology. In this discussion, I limit the term technology to modalities and instrumentalities that greatly extend the power of human action, sensation, or thought independently of their user. In addition to the instruments and devices usually considered as technology, we should include, for the sake of understanding, powererful medications—such as cardiac, antimicrobial, or psychotropic—that greatly extend our therapeutic power. It is our power that technology expands. Technology is not science. The two are frequently lumped together—as in sci-tech—but they are distinct. Science is not my topic. The purpose of this chapter is to see what there is about positron emission tomography (PET), magnetic resonance imaging scanners (MRI), angioplasty, endoscopy, automated chemistry machines, and so on—the whole wondrous parade, not the science that spawned them—that poses problems for medicine. Technologies are reductive and oversimplifying. Much of their hold on medicine, however, is a result of two prior reductive steps in the history of medicine. The ¤rst step was reducing the problem of human illness to the biological problem of disease, as discussed earlier. The second reductive step follows from the scienti¤c investigation

[64]

doctoring

of diseases. Here the ¤ndings of science become the accepted picture of the disease, further oversimplifying the problem. There is, however, a certain circularity whose presence should be acknowledged. Disease de¤nitions allow the entrance of science. Science increases knowledge of the disease by employing technologies and promoting the development of further technology. These technologies come about because of the scienti¤c understanding of the disease and reinforce the original picture of disease that started the cycle. This circle also contains the values that direct the technologies toward the facts that support the values. Breaking out of such a circle is one of our tasks—but not an easy one. The process just described may be the way technology entered medicine, but knowing this will not end its almost autonomous growth. We will not solve the problem of technology without providing other solutions or defenses against the human characteristics that lead to our dif¤culty. I will discuss six such characteristics: fascination and wonder, the lure of the immediate, unambiguous values, the avoidance of uncertainty, the self-perpetuating nature of technology, and the human desire for power.

Fascination and Wonder The ¤rst hold that technology has on us I call wonder and wonderment. Everybody loves the new and the shiny, especially when it does fantastic or seemingly inexplicable things that enthrall us. If I were to put a device that looked different and had a screen and a strange keyboard on a table, people would soon start poking at it, manipulating the keys to see if they could bring up the control system, ¤nd out what it is and how it works. Fascination and wonder are not easily put aside and are quickly reawakened; one taste leads to a desire for more. Consider the American romance with the automobile. For this reason, physicians may employ technology in the care of patients because the images and ¤ndings are so fascinating. Yet, the attraction transcends mere fascination. The awe that properly belongs to contemplation of the body becomes attached to images distanced from the body, such as a cineangiogram or an MRI scan of the spine. Wonder and fascination cause physicians to use and overuse their technology. They like to see it in action—and they want a new

The Special Problem of Technology

[65]

model now. Wonder may seem a childish motivation in a very serious pursuit. It is childish, which is one of its attractions. It helps solve the problems of boredom, absence of meaning, and loss of motivation. The human body is wondrous and so is the psyche, which is why some doctors love to take the body apart and others love to pry into the psyche. Yet surgeons like psychiatrists are taught never to cut into the body or probe the mind unless it is in the patient’s good. We know that curiosity—an aspect of wonder—is not easily held in check, but much time in medical education is spent successfully socializing doctors to do precisely that. Therefore it can be done.

The Lure of the Immediate The second reason for technology’s hold on physicians is that it roots us in the immediate, the now of its presence. The numbers of the readout, images on ¤lm, dexterity required for its deployment, technical complexities, tubes, wires, plugs, valves, needles, gauges, mirrors, focusing devices, and so on exist in the here and now—the immediate moment. These things are immediate in another related, but perhaps more important, sense: they are unmediated by our own reasoning. Technological output itself is a thing. Computer jargon even has a name for it: WYSIWYG, What You See Is What You Get. The user doesn’t have to reason from one output to another; each is distinct. How different technology is from the patient: a bundle of large questions, a life that exists only in the most fragmentary sense in the here and now. Look at a patient, see only the here and now, and you have missed the truth of a sick person. Any one moment of life—in an intensive care unit or a nursing home—contains only a bit of something much larger. A human life is a trajectory through time, the historic route of a society of complex parts. Sick persons, all persons, are dif¤cult to understand. The doctor is also a complicated life through time that interacts with the patient. God bless the immediate; no need to get caught up in all that dif¤cult sick person stuff. This is why, given a complicated human question in the care of the sick, doctors love to start talking about physiological parameters, calling up diseases or planning some tests. For example, the attending physician and medical students stood outside the room of a dying patient with uncontrollable suffering.

[66]

doctoring

Did they speak about her suffering, or about what to tell her or do for her? No, they were reading her test results and X-ray ¤lms— irrelevant to her present problem but simple and immediate. Why isn’t the examining hand on the abdomen just as immediate as looking at a readout or a computer-generated image? Because it isn’t just a hand or sensations in the ¤ngers; it is a doctor feeling responsible for the approximation of what ¤ngers indicate about an unseen reality—and what it means. Why isn’t the same thing true of the image on the ¤lm? It can be true; it should be true. The physician viewing the image should be reasoning about what must have come before and what will follow from the information contained in the image. And then how that information ¤ts in with what he or she knows of the patient and the patient’s interests, desires, purposes, fears, and concerns. As the technology gets better, it becomes more autonomous; it tells you directly what it means in immediate terms— like the computer-generated electrocardiographic (EKG) interpretations. Or a specialist, whose sole job is to interpret the image, tells you what it means in unmediated terms. As we all know, physicians today less often read their own X-rays (even with the interpretation in hand); or go to the pathology department when the biopsy is being reported; or question the precision, accuracy, or validity of the automated chemistry report. It is not that doctors are lazy; they have come to accept these technologies and their output as the equivalent of the thing being tested. There are specialties in medicine that have always lived more in the immediate present than others; surgery is the best example. The open wound, ¡owing blood, and exposed viscera are more immediate than the evolution of a drawn-out illness. The special attraction of the immediate is one reason that surgeons are different from internists (or the reverse). Systems of answers that we teach physicians about diseases are ill-suited to frame the longer-term, larger questions raised by the sickness of the person to whom the monitors are connected. Science has ruled out of court the information from values and aesthetics by which we live, allowing only brute facts. One advantage of the immediate is that it provides answers—information—when more relevant understanding would require deeper reasoning and greater involvement from doctors as persons. Unlike the computer, understanding cannot operate separately from the reasoner. Immediacy, and its less urgent requirement for reason, facilitates detachment

The Special Problem of Technology

[67]

from the suffering of a patient. Thus, technology holds us in thrall as much by the seeming advantage of the immediate as by its wonder.

The Lure of the Unambiguous The third aspect of technology, unambiguous values, keeps it employed sometimes even when it is inappropriate. Watch the movie of a coronary arteriogram. It’s like a western where you can quickly tell the good guys from the bad guys. The values are clear and unambiguous. A good coronary arteriogram with adequate dye in the vessels is anatomically clear. You can compare it to those taken previously and subsequently. A good coronary artery is open and a bad one is obstructed, although there are criteria for degrees of good and bad. A good obstruction is short, with adequate runoff, and not so tight that an angioplasty balloon won’t get through it or allow the surgeon to bypass it. Cardiologists, although they may disagree with one another about details, are absolutely clear and straightforward about such things. When they are not, they develop new criteria to remove ambiguity. Unambiguous values mark virtually all technology. In fact, lack of ambiguity is essential to good medical science. Progress in medical science is impeded and international research is impossible if cardiologists at Cornell cannot speak the same language and mean the same thing by technical terms as those at Stanford, Oxford, and the Hôtel de Dieu. So, on the face of it, the unambiguous seems reasonable—except that we physicians do not generally know how or when to abandon it. Many, perhaps most, of life’s simple pleasures are also unambiguous. We generally know what is good and bad behavior, food, wine, and sex. On the other hand, the development of sophistication in nontechnological pursuits involves appreciation of complexity and ambiguity. Sophistication in technology, I believe, goes in the other direction. More sophistication means less ambiguity; the better the equipment, the clearer the values. Good or bad as measured by technology is not necessarily the same thing as good or bad for patients. Coronary artery disease and its technologies are a case in point. Consider a middle-aged man without symptoms who wants to join an exercise program. He is required to have a treadmill exercise test. The test (following the usual Bruce protocol which approximates no exercise you have ever

[68]

doctoring

done) is positive by published criteria. He is advised that this unambiguous criterion often indicates coronary artery disease and should be followed by a thallium stress test. This test is also positive, and he is advised to have a coronary arteriogram. (Often the thallium scan is considered redundant; the patient goes directly to the arteriogram.) The arteriogram shows signi¤cant obstruction of a coronary artery. Subsequently, a coronary artery angioplasty is done to reduce the obstruction. This little scenario is extremely common in the United States and increasingly so elsewhere. There is no good evidence that the outcome of this chain of events makes a positive difference in the life of such a patient—that asymptomatic patients with a positive test, as described, who go on to angioplasty or bypass surgery do better than patients who are not so treated. The relationship between what is considered good and bad in the test results and what is best for the patient is, at the very least, obscure and, at the worst, just plain wrong. What has taken place is that because available technology permits visualization of the major coronary arteries, atherosclerosis of these vessels, which can be unambiguously demonstrated, is now considered the equivalent of coronary heart disease in its de¤nition. However, coronary heart disease is a more complex entity than merely atherosclerosis of the major coronary arteries, although the two are often associated. For example, the degree of obstruction of a coronary artery does not necessarily correlate with coronary blood ¡ow; this can be seen even in the data from papers that report a positive relationship.2 The reason is that coronary arteries are dynamic functional entities, not merely passive pipes. Endothelial dysfunction is a more important factor than structural obstruction.3 As another example, at autopsy one commonly sees old people whose coronary arteries are so ¤lled with the calcium deposits characteristic of advanced atherosclerosis that one wonders how blood ever got through them. Why did they have coronary artery disease but show no evidence during life of loss of everyday function? Because they did not have heart disease. Unfortunately, physicians infrequently attend autopsies and thus are not exposed to this common phenomenon. Conversely, sometimes one sees a patient with clear-cut signs of coronary heart disease but little evidence of obstructed coronary arteries. I am aware that cardiologists understand that coronary heart disease is more complex than mere stenosis of a coronary artery.

The Special Problem of Technology

[69]

They know the contribution of cigarette smoking to coronary events, the importance of dyslipidemia in its genesis, the impact of a strong family history, the importance of population variables that mitigate the effect of dietary fat (in Eskimos, for example), and the contribution of other disease variables such as diabetes or hypertension. In the evaluation of the previously described patient with technologically demonstrated coronary artery stenosis, however, what I said above of its unambiguous hold on the cardiologist is too often true. This is exactly the point of understanding the grip technology has on physicians. As I suggested previously, human sophistication is marked by tolerance for ambiguity, whereas sophisticated technology removes ambiguity. It does this by narrowing down the ¤eld of difference between what is good and what is bad, so that ultimately one test result is taken to be good and another result bad. That is how the state of the coronary arteries became accepted as the equivalent of a disease of the heart itself in the circumstances I have described. We must not forget that technological measures of value, even as they achieve a life of their own, are derived from human values. When medicine’s priorities (another word for values) are simplistic, they will be represented by a technology that also exempli¤es simple values: for example, giving a part priority over the whole—sustaining an organ but losing sight of what is best for the whole person. We value the preservation of structure over the preservation of function—fortunately less of a problem nowadays. We value the body over the person; we value survival over maximum function and length of life over quality of life. The development of technology is not an event but a process. Technology is invented to solve problems arising out of the pursuit of medical values. Technological values, however, foster and reduce the ambiguity in medical values, which subsequently leads to a new stage of technology. The result is that the sophistication necessary for physicians to tolerate the ambiguity that would inevitably follow attempts to break out of the circle is sti¡ed. Perhaps the development of MRI scanning of the coronary vessels is an example. The odd thing is that if one faulted modern physicians for lack of sophistication, they would most likely dismiss the criticism by pointing to the sophisticated equipment they use. So, to fascination and the lure of the immediate, we add unambiguous values as a reason technology runs doctors rather than vice versa.

[70]

doctoring

The Pursuit of Certainty The next reason for the dominance of technology is medicine’s pursuit of certainty. The central problem that physicians confront is uncertainty. It is doubt that grays hair. Many years ago, a classic study of medical students identi¤ed two reasons for uncertainty:4 ¤rst, defects in the knowledge of the individual physician and, second, the inadequacies of the profession’s knowledge. Even if I, impossibility granted, knew everything medicine knew, I would not know everything. There would still be uncertainties. But in an ideal world of complete knowledge, in this view, we could be certain. Unfortunately, as Sam Gorovitz and Alisdair Maclntyre pointed out years ago, there are two other roots of uncertainty that can never be removed.5 The ¤rst is that every decision, small or large, is made about the future and the future is ineluctably uncertain. All medical decisions are about the future, since the future starts an instant from the present. Second, uncertainty can never disappear because all of science, medical science or any other, is about generalities. By contrast, every patient is a particular individual and, therefore, necessarily different in some respect from the general. Thus, clinical judgments are always uncertain and medical knowledge necessarily involves uncertainties. By the nature of the uncertainties of clinical medicine, the more important the knowledge required by the decision, the less tolerable is the uncertainty. Since physicians commonly make decisions that have profound implications for the lives of others, uncertainty is a constantly disturbing factor in medical practice. Patients have the same or greater uncertainties as physicians. They commonly solve them by trust—trust in the physician—that increases the burden of the doctor’s uncertainties. Physicians follow the same path, as they trust their consultants and their technology. There are numerous strategies to reduce uncertainty, and technology can play a part in each of them. The ¤rst strategy is to reduce the clinical problem until it concerns not a particular sick person but an organ. For example, a patient complains of unusual chest pains. The pain is not related to exercise, position, or food, yet it is persistent, raising the question of coronary heart disease. One test for heart disease after another is normal. The patient is reassured that the pain does not represent heart disease because the tests are negative. This may be correct, but the pain, not the presence or absence of heart disease, is the problem. Positive or negative probabilities of

The Special Problem of Technology

[71]

each test provide an answer to the rede¤ned problem and reduce the physician’s uncertainty. Perhaps the patient will also be reassured but perhaps not. The physician’s statement, that “Your problem is not . . . ,” is not nearly as good as a positive answer to the question raised by the symptom: “Your chest pain is a result of . . .” Further, the question of chest pain has been changed to the question of coronary heart disease and then to the question of coronary artery disease. As I have noted, changing the question is a result of available technology. Rede¤nition of the problem in terms of a technological answer is often employed in the case of back pain. Here the question of the cause and treatment of the patient’s pain becomes a question about the pathological anatomy of the spine, which can be answered by an MRI scan. As we know, the cost of this transmutation has risen steadily over the last decade as the initial simple X-rays of the spine have been superseded by CT and, most recently, by MRI. The image on the ¤lm—with its implication of objective certainty—comes to stand for the patient’s back pain, to the point at which greater weight is given to the image than to the pain. Problems such as chest or back pain were previously addressed by taking the patient’s history and doing a physical examination. These diagnostic methods were fraught with uncertainty and were particularly dependent on the skill of the individual physician. These techniques had the further disadvantage of forcing the physician to confront an intractable source of uncertainty: the individuality of the patient. Technological methods move the evidence employed in diagnosis away from the patient and reduce the impact of the patient’s particularity on the physician. Thus, physicians mistakenly believe they can reduce uncertainty by changing the patient’s problem to one for which there is a technological answer. They then reduce the problem from that of the patient to that of an organ or part for which a technology exists, and they distance themselves from the patient by employing that technology. On the therapeutic side, technologies may reduce uncertainty by providing treatments that, although of unquestionable value in some situations, are also employed in situations where they have no utility. Technology would not produce problems in relation to uncertainty if it did not frequently reduce uncertainty, sometimes dramatically. Probably because of the change produced by effective technologies, Renee Fox’s description, previously cited, of medical

[72]

doctoring

students speci¤cally trained in the management of uncertainty no longer applies. Consequently, physicians generally use any diagnostic or therapeutic technique that promises to reduce uncertainty. All the wonder, dislike of ambiguity, and fear of uncertainty that af¡ict physicians are present among patients. The stakes are highest for them. In the current medical world of the United States, patients have a signi¤cant voice in the choice of diagnostic strategies and treatment. Patients are now generally knowledgeable to an unprecedented degree. Not surprisingly, their knowledge is greatest about new technologies and treatments, details of which ¤ll the pages of newspapers, magazines, and health promotion newsletters. It is fair to say that many patients believe that it is the test rather than the physician that makes the diagnosis and the drug rather than the physician that effects the cure. (If a CT scan shows a lung tumor, did it make the diagnosis? A physician chose a CT scan rather than, say, a plain ¤lm of the chest and decided to image the chest rather than, say, the abdomen. The CT scan was employed in the diagnosis.) Consequently, patients have been an active force in the increasing deployment and dominance of technology. The restraint of technology is, ultimately, also the restraint of patients’ demands for it.

The Self-Perpetuating Nature of Technology Employing one technology frequently leads to the use of another. This is most easily demonstrated by the function of computers in neonatal intensive care units. Commonly, each “bed” in a neonatal unit has its own computer to analyze and display the physiological state of the infant. The requirement of computers for digital information encourages the proliferation of instrumentation that produces such data. Another example is the effect of automated blood chemistry machines in making redundant the manual skills of technicians. Other automated laboratory examinations become necessary because technicians no longer do the tests by hand. The results of using one technology frequently raise questions that can apparently be answered only by other technologies. CT images of the central nervous system may introduce doubt that only MRI scans can resolve. General expectations have been created among physicians about levels of accuracy, certainty, and lack of ambiguity that can be met only by other technologies—even if such

The Special Problem of Technology

[73]

accuracy, certainty, and lack of ambiguity are not important in a particular instance. Doctors who have mastered a technology tend to use it as often as possible—not necessarily for pro¤t, but because they love their skills and technology. As noted earlier, problems are rede¤ned so that a technology becomes appropriate when it might otherwise not be. Physicians trained exclusively in this high-technology era may simply not know that patients often get better on their own. Not all (perhaps, not most) patients with angina need an angiogram or an angioplasty. Technologies also extend their application after introduction. Endoscopic gallbladder surgery is a case in point. Endoscopic cholecystectomy promised to reduce the cost of gallbladder surgery in the United States primarily by shortening the length of hospitalization. That expectation was not ful¤lled because, within a few years, the total number of cholecystectomies almost doubled.

The Desire for Power The ¤nal reason for the inappropriate use of technology is the power it confers on physicians and their institutions. While the meaning of power seems self-evident, some further explication is required. The power to act is basic to human existence and is employed to control or in¡uence events. In its absence we feel powerless, which is a self-destructive state. We exist for ourselves and for others in our actions; when we act, we simultaneously create ourselves and our world. The scope and effectiveness of our actions in both selfcreation and in¡uence on the world are determined by the degree of our power. Since we are social beings, all of our actions take place in a world of others, and our power is relative to the power of others. Thus, my ability to act among others is partly dependent upon permission to exercise my power by those who are more powerful or my desire to exercise my power in relation to those who are less powerful. Frequently, the word hierarchy is used to refer to social ranking according to power. Power resides in us not only as individuals but also by virtue of our acknowledged place in society— our social status. Thus, hierarchy may be role dependent rather than the result of self-generated power. Power relationships—which also exist among and between animal groups—are dynamic. It is dif¤cult to exaggerate the importance of the exercise and experience of power.

[74]

doctoring

Even in sophisticated societies, the ability to do things better than others confers power. Possessions confer power because they bestow status—material wealth is the most obvious example—but so does access to objects of superior ef¤cacy. Changes in one’s access to things that contain superior ef¤cacy in themselves may alter one’s status. A well-known ethnographic example involved a culture in which the only available axes, made of stone, were in the possession of the tribal elders. These axes were not only used to cut wood, but were also measures of status and a factor in barter among tribes. Western missionaries came to the tribe and offered more effective steel axes as incentives for conversion to their religious beliefs. Wide distribution of these powerful objects—both totemic and effective— dramatically altered hierarchical and status relations within the tribe and its associations with its neighbors. In this discussion, technology, as noted earlier, refers to modalities and instrumentalities that greatly extend the power of human action, sensation, or thought and have this ef¤cacy independently of the particular user. There is little doubt that one attraction of technology is its ability to confer status and rank on individuals. Medical power is shown when a plastic surgeon makes someone look younger or when infection is treated, blood pressure lowered, or pain relieved. Every therapeutic and diagnostic act is a demonstration of ef¤cacy and thus of power. The therapeutic effectiveness of the relationship between patient and doctor depends, in part, on a belief in the physician’s individual and institutional power over the forces of nature. In previous epochs, the physician’s power came not only from his or her shared knowledge of the body and disease, but also from the personal development of knowledge about the sick and sickness, and from demonstrated effectiveness in the diagnosis and treatment of patients. Personal power of this sort takes many years to develop and is inevitably a result of the ripening of the medical self. Technology confers power on individual doctors with much less personal involvement and less effort. Why struggle to learn to use a stethoscope or respect someone who has learned? Order an echocardiogram. Why develop expertise in performing a neurological examination? Get a CT or an MRI scan. Why know a person rather than order a test? The modern tendency toward specialization encourages the use of this more easily gotten power because it reduces the knowledge necessary to exercise it. Technology also gathers to itself personnel and space that exhibit power and are self-perpetuating. Intensive care units and transplant

The Special Problem of Technology

[75]

units are excellent examples. In medicine as elsewhere, technology engenders special training, which furthers the world view of technology, which further increases political power. Employing or having access to technology also garners social power or status from laypersons, the press, the university, or hospital trustees. Similarly, technology confers status on hospitals and other medical institutions. Technology would not confer power on doctors and the profession of medicine if it were not seen by the larger society as having power. It erroneously appears to free the patient from the necessity of depending on the individuality and individual skills of the physician. (This is re¡ected in uniform fee schedules that pay for a particular physician’s act—of¤ce visit, surgery, and so on—as though doctors dispense a uniform technology rather than a personal, individual service.) It is not surprising, in view of this widespread public belief in the independent power of technology, that physicians—who are in¡uenced by the public they serve—depend increasingly on technology whether or not its use is appropriate.

Knowledge at a Distance In our daily lives, we are accustomed to confronting much of our world in its representation rather than in itself—in photographs, recordings, radio, movies, and television. This has produced the widened perspective and scope of knowledge about things distant and close to us with which we are all familiar, but especially for doctors, it has created problems. Technology represents a kind of knowledge. It epitomizes the twentieth century ideal of knowledge—scienti¤c, objective, and seemingly separate from humankind. For medicine, the scienti¤c knowledge and subsequent technology developed in response to the challenge posed by sickness and suffering has assumed an actuality more convincing than the reality of sick persons themselves. Consider this common situation. A patient has severe pain in the hip, and the doctors can ¤nd no evidence of disease. With each negative test, increasing doubt is raised about whether the patient is truly in pain. Then a radionuclide bone scan is done showing cancer in the hip bone. The patient will now be believed. Why is the celluloid rectangle with fuzzy black dots more believable than the patient’s pain? The usual answer, that the pain is

[76]

doctoring

subjective, does not hold. The pain may be subjective, but the report of pain is a thing that can be evaluated. Further, we are of a piece; we cannot have severe pain without its re¡ection in other aspects of our physical, social, and psychological selves. A person with severe pain moves, acts, thinks, feels, displays emotion, and relates to others differently than the same person who is pain free. All of these features are apparent to others or can be evoked; they are objective. Objectivity alone isn’t the issue. The facts that tell us that the patient really has pain do not meet the ideal of medical scienti¤c knowledge developed over the last 150 years. His report of symptoms and the way he looks and acts—all objective—cannot be measured or seen apart from the patient or doctor. Scienti¤c knowledge, surely not the only way to know things, has come to be accepted as more actual than patients or their pain or suffering. Medical technology also produces representations of patients’ original reality that are another reality in themselves. For example, EKGs, X-ray machines, monitors, CT scanners, MRI machines, and PET scanners are all imaging devices that distance physicians from the sick person. Their focus of interest inevitably shifts from the patient to the part or the disease—from the whole patient and the patient’s lived world.6 Realizing this, physicians, commentators on, and critics of medicine have largely depended on moral injunctions to return medicine’s focus to the sick person. This is an uphill struggle because the problem is based, in part, on the nature of medical knowledge itself and is shared by twentieth-century humankind.

Conclusion Technology holds sway over medicine and the public because of its capacity to induce wonder, root us in the immediate, remove ambiguity, increase certainty, perpetuate itself, and enhance power. Since this is not well known, it is hardly surprising that technology, by itself inert and useless (although beckoning for attention through its inherent purposes), should be blamed for the troubles it brings rather than the doctors who use it, the public that loves it, or the narrow knowledge on which it is based. Medical technology’s form and character arise from medicine’s focus on disease and pathophysiology as the arena in which the origins of and solutions to human sickness are to be found. The values on which it is based come

The Special Problem of Technology

[77]

primarily from the pathophysiological and anatomical criteria for disease and normalcy, now largely de¤ned and perpetuated by the technology. Our task, it seems to me, is to stop blaming, regulating, and complaining about technology—without which modern medicine is unthinkable—and start working toward a solution based on understanding, as we have done with so many other problems in medicine. Primary care medicine will achieve its medical, social, and economic goals only to the extent that it solves the problem of rampant technology.* Toward this end, we must learn how to teach primary care physicians, who are themselves the primary instruments of diagnosis and treatment, to tolerate uncertainty, accept ambiguity, deal with the complex, and turn away from mere fascination. Accepting these assignments and redirected goals and following them as far as they lead will be a suf¤cient task for decades, but this process can begin now.

Technology’s Counterspell Paradoxically, each of the characteristics that make technology so captivating to physicians—its capacity to fascinate and induce wonder, root its user in the immediate, increase certainty, remove ambiguity, be self-perpetuating, and enhance power—ends in a contradiction and leads the physician away from the patient. This contradiction is the fundamental ¡aw in the application of technology in medicine, since its development and use is meant to serve patients. The most entrancing aspect of technology is the fascination and sense of awe that it elicits. Fascination is a transient emotion, however, and must constantly be fed if it is to continue. It is a surface phenomenon, unlike awe or wonder in the face of mystery, where * This is not, of course, a problem for primary care medicine alone. All of medicine and society in general are fascinated with technology. One aspect of the advance of society is the control of those things that threaten it. Some group demonstrates that control of technology is possible, or that ambiguity and uncertainty are not necessarily bad (seeing that they are always present), and the idea spreads. The intense desire for certainty that is evident in current society has not always been present. It is part of the heritage of Descartes and Newton and the seventeenth century. Physicians have been in the forefront of many societal advances throughout history. The control of technology by primary care medicine may be another instance. (See Stephen Toulmin, Cosmopolis: The Hidden Agenda of Modernity, Chicago: University of Chicago Press, 1990.)

[78]

doctoring

attention is drawn deeper and the senses and the intellect are enlisted in the search for its source. Ultimately, therefore, both fascination and wonder, like infatuation, lead to boredom as interest ¡ags. Medical students often express disbelief when they are told that boredom is a problem in emergency medicine. How can that be in a ¤eld where so much is happening and where there is so much opportunity to do interesting procedures? Because, no matter how fascinating a technical procedure is, over time the challenge fades and the interest dies with it. In emergency medicine patients come and go quickly, there is virtually no time for the emergency department physician to know them, and there is little follow-up. The epitome of immediacy in medicine is in the intensive care unit. Here, patients are monitored for possible events. The whole apparatus of the intensive care unit—staff, equipment, and architecture— is structured to respond to such problems. The monitors and other measurement technology hold observers’ attention in the here and now. Procedures such as placing a Swann-Ganz catheter, bedside echocardiography, or doing a venous cutdown focus the patients and staff in the moment. Time—clock and calendar time—loses meaning. Patients’ lives prior to and after the hospitalization become abstractions. This approach to medical care is becoming common throughout medicine. Unfortunately, the immediate in medicine is also inherently contradictory in both the temporal and the intellectual sense. The immediate problem of the patient has a history and is part of a process. Whatever problem occurs now does so because of what happened previously, and the present will always result in a future that will be affected by what is done now. In addition, remaining in the immediate present distances the doctor from the patient, who is the enduring subject of past, present, and future. The immediacy of technology contradicts itself and leads to the perils of precipitous and unre¡ective action. In the intellectual sense, immediate, unre¡ective action contradicts the long history of medicine’s intellectual achievements and the training of individual physicians. Medical care in the immediate present is the antithesis of the longitudinality, comprehensiveness, coordination, and emphasis on prevention associated with the ideal of primary care. It is probably correct to say that the greater the degree of certainty in any test or medical activity, the greater the abstraction of the measure of certainty from the reality of the patient’s life. This situation is akin to the old joke about specialists being people who know

The Special Problem of Technology

[79]

more and more about less and less until they know everything about nothing. The technological pursuit of certainty started as an effort to serve the best interests of the patient. The paradox is that increasing certainty is achieved only by getting further and further from the interests of the whole patient, substituting instead the best interest of one of the patient’s parts—for example, the coronary arteries. When something is unambiguous, its values are clear. Technology usually comes into being after categories have been created that clearly de¤ne the good and bad or the healthy and diseased. The categories may or may not be good representations of the disease being considered, but they are necessarily abstractions. You can experience the same phenomena by watching television newscasts. No matter how complex a particular problem is, the oversimpli¤cation that necessarily follows from portraying it in pictures for a few minutes fools the mind. The ambiguous has seemingly been made clear. It appears that you know the good guys from the bad guys. This apparent clarity is misleading, however, and for that reason the real meaning is further obscured. To mistakenly believe that you understand something increases its inherent ambiguity. Issues of value in medicine are often ambiguous, because the human condition is complicated. Much of medical technology’s pursuit of the unambiguous, and of certainty, has the same problem as the television news presentation: the distinction between what something is immediately and what it is essentially. The fundamental, essential whole that gives meaning to everything in clinical medicine is the patient—the sick or well person. I may fool myself into believing that what is good or bad for the part or the disease is good or bad for the person, but often this is not true. The power conveyed by technology also leads to a contradiction, but one that is more obscure. Historically, the power of the profession of medicine has been the power exercised by individual physicians over sickness, disability, and death. By that light, these are strange times. Despite the enormous power of modern medical technology, acknowledged by patients and physicians alike, the power of the medical profession is at a low point. Doctors had greater public power in 1950, when technological power was modest. The command of technology by individual doctors—cardiac surgeons are a case in point—still impresses other doctors and the public. Institutional technological capacity stirs the public (which has come to expect it). Unfortunately, the public knows the truth: doctors are

[80]

doctoring

subordinate to their technology—they have ceded their power. It may seem that they have maintained control over the technical aspect of their work, but that control also has been diluted as the public has become more knowledgeable. It could not be otherwise. If it is the knowledge of science and technology that does the work in medicine—makes the diagnosis and treats the disease—then whoever has the knowledge has access to the power. The claim of twentieth-century medicine has been that medical science is the source of medicine’s power. The importance of the individual physician has become subsidiary, as though any doctor with the knowledge and command of the technology is as good as any other doctor. (Doctors may still have power in their institutions—hospitals are a case in point—but that too has diminished with increasing regulatory pressure, the use of professional medical administrators, and the dominance of economic issues over all others.) One should not come away from this discussion thinking that I believe that technology is bad: that is an untenable and maybe even a silly stance. I know, however, that there will not be any primary care medicine worth its name until physicians are trained and experienced in subordinating modern medical technology to the practice of medicine. Training provides skills in the use of technology and in the avoidance of its pitfalls. Speci¤c training in the management of uncertainty and the other problems raised by technology is one of the new areas of medical education created by the progress of medical knowledge. It takes time for physicians to gain enough experience to realize the importance of the skills and to see them produce a better and more rewarding medical practice.

The Clinical Method

[81]










4 The Clinical Method








T

he clinical method is the means by which physicians discover facts about the sick or well patient and enter them into the diagnostic and therapeutic process in equal partnership with information about disease, pathophysiology, and technology. Medical education takes its basic form and content from two ideas: that the diagnosis and treatment of disease are based on medical science and that physicians learn medicine best by direct experience with patients. It is expected that the scienti¤c knowledge of medicine will continually become outdated. In the course of their education, physicians not only gain scienti¤c knowledge, they also learn the method of medical science and how to think in scienti¤c terms. This effectively provides the basis for their continued growth in this sphere. It is less obvious that students are also taught a method for learning from patients the more formal aspects of which are how to take the history of an illness and do a physical examination. Prior to the current generation of diagnostic technology, what physicians discovered from patients by their personal efforts was the most important source of information on which diagnosis and treatment were based. With continuing experience, their expertise in interviewing and examining patients grew and continued to provide the basis for their patient care. These skills [81]

[82]

doctoring

have lapsed in this era, and physicians are now distanced from their patients by their technology. When a method of learning is good, it should allow physicians to depend on it for the rest of their lives and support the adventure of self-instruction, whether about medical science or about patients, as far as they want to go. This appears to be true of scienti¤c thinking but distinctly less so of the method physicians are taught to learn from patients. The inadequacy of the clinical method as currently taught and practiced creates problems for physicians, patients, and the health care system—especially primary care medicine. There are three problems. First, knowledge of persons, generally absent from the method, is just as important as knowledge of disease as a foundation of primary care. Second, much of the information required for diagnostic and therapeutic decisions in primary care must still be obtained from communicating with and examining patients. Finally, the clinical judgment of individual physicians is also a component of the clinical method that is insuf¤ciently stressed, and yet the health and well-being of patients are directly dependent on it. It is necessary to examine these assumptions and to explore the components of the clinical method.

The Importance of Knowledge of Persons Knowledge of persons is particularly important today because what most clearly distinguishes chronic disease from acute disease is that it takes place over a long enough period of time so that the nature of the person has an undeniable in¡uence on the unfolding narrative of the disease and the disease in¡uences the further development of the person. This principle is as applicable to children and teenagers as to pregnant women and other adults. Two separate meanings of this interaction can be distinguished. First, the diseases themselves, their pathoanatomy and pathophysiology, are so often in¡uenced by the actions of the patients that even if the same disease began in two patients in the same manner, its manifestation would quickly become divergent because of the differences in the persons. Second, the nature of the illness, how the disease expresses itself in all the dimensions of the person, varies from patient to patient. The consequences of these facts for the care of patients and the training of primary care physicians are considerable. At any particu-

The Clinical Method

[83]

lar moment in a chronic disease, the doctor must be concerned with the state of the disease, its effects on the person, the person’s impact on the disease, and with discovering in these factors the clinical problem to be solved or the goal to be achieved. A brief example may help. A married sixty-six-year-old television director was found to have an elevated prostate speci¤c antibody level (8.4) during a routine physical examination two years ago. He considered himself healthy, exercised regularly, and followed a very careful diet. Biopsies revealed cancer of the prostate. Instead of taking the tests to be sure that the cancer had not spread outside of the gland and undergoing the radical prostatectomy suggested by the urologist, he and his wife jointly decided that he would follow a rigorous schedule of diet and large doses of vitamins suggested by an alternative treatment physician in their city. A prostate speci¤c antibody level last month as part of a company-sponsored physical examination was 24. He and his wife have come to the doctor to discuss whether he should start “ablative hormone therapy,” which was recommended by a friend who is said to be informed about prostate cancer.

This patient may have adversely in¡uenced the outcome of his disease by his actions of the past two years. He acknowledged that his treatment decision did not succeed in eliminating his cancer. He has also demonstrated a set of beliefs and attitudes about treatment (about which he is articulate) and about the patient’s place in decision making. Whether his disease has spread beyond the prostatic capsule is not yet known. If it has, but does not threaten neurological involvement, many urologists believe that no treatment is necessary at this time and that cure is impossible. If it has not, then both radical prostatectomy and radiation therapy are possible. Both are used to cure the disease, but have different side effects. The case may be discussed from perspectives as different as the literatures about pathology, diagnosis, treatment, and outcome in carcinoma of the prostate, the job of physicians, the doctor–patient relationship, in¡uencing patient choice, ethical issues in patient decision making, and the goals of treatment. Each point of view should be taken by the patient’s primary care physician in deciding what is to be done. No consultation with a specialist will solve the primary care physician’s problem. There is no way to disentangle the disease from the patient. One facet of the case may be dominant at one moment, others at other times. Whatever actions are taken at this point in the

[84]

doctoring

case, the unfolding story will not stop at this visit. This man has a chronic disease, and everything done at this juncture will continue to in¡uence the many dimensions of his care in the future. If he dies of his disease, what is taking place at this visit will in¡uence the ability of the primary care physician to ensure an optimal trajectory to death.* Two other patients were discussed in Chapter 1 whose stories bear on the question of what needs to be taught to primary care doctors. The ¤rst was a woman with long-standing insulin-dependent diabetes who is virtually blind and, after months of vomiting, clearly suffering. The other was a patient with chronic back pain whose negative attitude toward treatments of virtually any type complicates her care. In both instances, knowledge of their diseases is vital to taking care of them, but this in itself is not suf¤cient to make a substantial difference in their future. The biology of diabetes is well known; understanding of common back disorders is primitive. Both patients have chronic conditions whose treatment must be considered not only with the present moment in mind, or even the immediate future, but also with the understanding that the doctor is trying to change what will happen months or years from now. In the case of the suffering diabetic woman who has received good care by current standards, her present problems are due in part to a failure to think about issues other than what is best for her diabetes. We have to ask ourselves whether the clinical method by which medical students and resident physicians are currently taught to know the patient is robust enough to meet the science of medicine on an equal footing. Is it a way of knowing that will last them a lifetime and take them as far as they can ever want to go? The answer to both questions is clearly no. I do not believe it is suf¤cient to say that we must reemphasize the importance of taking a good history or doing a careful physical examination, important as they may be. What is being asked of the clinical method for primary care medicine goes beyond these skills. The issue is clear. We are in search of a clinical method strong enough to live in equal partnership with medical science, adequate to the knowledge of persons necessary to primary care medicine, * This case should also help distinguish what I am discussing from at least one view of what is often called patient-centered care. Of course, this patient should help determine his medical care; in fact, he already has, but not only in the manner he intended. Freedom of choice is only one aspect of a person that must be considered under the heading of respect for persons.

The Clinical Method

[85]

and rich enough to last clinicians for a lifetime. The avenue to such training and even clinical wisdom is open. It leads, in part, through an understanding of the nature of the information on which it is based and the method employed to obtain those facts. Primary care requires that physicians be able to think equally well about both diseases and persons. It is essential that primary care physicians be well trained in scienti¤c medicine, able to think in its terms, and able to command its technology. Chapter 2 showed, however, that thinking about patients from the perspective of medical science may con¡ict with thinking about them as persons. It is necessary to understand that the reductionist, atomistic, context-free manner of thinking about diseases that is the heritage of almost 200 years of medical science is not appropriate for thinking about persons. In fact, in thinking about the disease process, the idea that diseases are objects that have somehow entered the body and made the person sick has outlived its usefulness. The change in structure or function that characterizes the disease process is best considered by clinicians as something altering the life history of the person with diagnostic, therapeutic, prognostic, and personal implications. In thinking about the molecular biology of the disease, on the other hand, the classical canons of scienti¤c thinking should apply. They are effective and appropriate modes of thought for this purpose. How are these two apparently con¡icting kinds of thinking to be resolved within the clinical method employed by clinicians? The most obvious and simple answer depends on the natural consequences of having a consistent, logical focus of thought. Whatever the logical focus of thought, everything else within the same universe of discourse will be seen to be related to it. This is the basic problem with the oversimpli¤cations of even the most sophisticated, cost-effective analyses or of any singleminded perspective. By contrast, when the person is the logical locus of concern, the scienti¤c information about disease and technology become subservient to ideas about the interests of the person. This is the exact corollary (but the opposite) of current situations in which sick persons’ concerns are made secondary to the body or body part. To achieve the goals of primary care and to promote clinical wisdom, the clinical method must place the sick or well person at the center of the physician’s thoughts but without impairing the physician’s ability to think and act scienti¤cally. The medical idea of objective (measurable) knowledge is in con¡ict with the often subjective (personal) knowledge that is neces-

[86]

doctoring

sary for the physician to know the sick or well person. Similarly, the mediated knowledge of the disease acquired through tests, measurements, and imaging seems to be in con¡ict with, and clearly superior to, the unmediated knowledge of the patient acquired directly through the subjectivity of the physician. The problem is not better or worse information, however, but what is required to resolve the clinical problem. In knowing persons, no modality is as effective as physicians acting as their own instruments. There is no reason why doctors who have a clear focus of concern and knowledge cannot employ different kinds of thinking. Architects, for example, are able to think in objective, measurable, engineering terms at the same time that they think about aesthetics.

A Plain and Simple Method Making the patient the logical focus of medicine, resolving the dif¤cult problems of thought, clarifying what is objective and subjective, becoming one’s own instrument so that subjectivity can meet objectivity on level ground, de¤ning the problems of chronic disease in this new fashion, understanding the special nature and problems of chronic disease, understanding that knowledge of the mechanisms of disease are only one of many requisite kinds of knowledge, and developing clinical wisdom—how is this all to come about? Physicians need a method, a way of going about things that will clarify both the nature of the disease and the nature of the process in the person, as well as allow them to integrate the two in a manner that ¤ts the way chronic diseases evolve. Sir William Osler said, “Medicine is the only world-wide profession, following everywhere the same methods, activated by the same ambitions, and pursuing the same ends.”1 (Sadly, his faith in this statement would not have survived into our era.) Let me start by reviewing the current method and some of its history. The patient presents a symptom—a statement about a perceived dysfunction. The doctor presumes that pathology—a disease—is the source of the symptom and asks questions that reduce the number of possible diseases that might cause the symptom. Then the patient is examined for changes in the body that support or deny the expectation of the doctor. Finally, tests are done that further point to one rather than other diagnoses. Treatment is proposed based on scien-

The Clinical Method

[87]

ti¤cally validated alternatives. Since the 1930s, the patient has also been asked about personal habits and social factors that might bear on the diagnosis or its treatment. Since the 1950s, psychological factors have been included for consideration, and since the 1970s the patient has been an active participant in the choice of therapeutic actions. On follow-up visits, basically the same method is pursued. The patient, prompted by questions, describes the evolution or resolution of the dysfunction, and the doctor looks for signs of changes in the pathology or pathophysiology aided, if considered necessary, by further tests. If no evidence of pathology is discovered, it is presumed that the patient does not have a disease, is not sick, or has other, nonmedical, sources for the discomfort. If the pathology resolves, the patient is believed to be well; continued dysfunction or disability would be considered psychological in origin.* This method, based fundamentally on discovery of the cause and treatment of the disease through careful observation of the patient, is the legacy of Hippocrates. The alternative method is to base diagnosis and treatment on some theory of the causation of disease such as the four humors of Galen or the mechanical principles of bodily action proposed by Borelli. Here the physician attempts to ¤nd evidence based on the dictates of the theory. This method, based on one theory or another, ruled medicine for almost two millennia! In the mid-seventeenth century, Thomas Sydenham brought the principles of Hippocrates back to medicine. His clearly stated idea was to carefully and fully observe the actual phenomena in the patient without the bias introduced by hypothesis. John Locke, a student of Sydenham, summed up Sydenham’s empirical method in a sentence * It would be untrue to imply that this method has not been criticized in many quarters and over a long period. The famous paper by Francis Weld Peabody, “The Care of the Patient,” is an important instance because of its in¡uence ( Journal of the American Medical Association, 88:877–82, 1927). A more recent example based on the long experience of a wonderful clinician and teacher is that of Eugene Stead. The patient’s complaint may arise because (a) a well body is equipped with a sensitive nervous system that detects changes induced by the environment that are not detected by less sensitive systems; or (b) a well body is being used in a way that will make most bodies complain; or (c) the patient’s nondiseased but nevertheless highly individualized body is being used in a way that would not make most well bodies complain, but does make his well—but somewhat different— body complain; or (d) illness is raising its ugly head. (A Way of Thinking: A Primer on the Art of Being a Doctor, edited by Barton F. Haynes, Durham, NC: Carolina Academic Press, 1995, p. 117.)

[88]

doctoring

in a letter to William Molvneux: “You cannot imagine how far a little observation carefully made by a man not tied up to the four humors [Galen], or sal, sulphur, and mercury [Paracelsus], or to acid and alcali (sic) [Sylvius and Willis] which has of late prevailed, will carry a man in the curing of diseases though very stubborn and dangerous; and that with very little and common things, and almost no medicine at all.”2 Medicine already has the basis for the clinical method that our contemporary problems call for. We have only to revisit it in its plain and simple form, acting for the good of the patient out of reasoned and deliberate judgments based on careful direct observation of the patient, unbiased by preconceived notions about diseases or the nature of persons.* This means a return to careful history taking, listening to the patient’s story, artfully enhanced by skilled questioning at every point in an illness. Persons are more than their words, so this process must be supplemented by discerning scrutiny of patients’ presentation to the world, behavior, mood and feelings, environment, and context. The physical examination has continuity with the history and remains a central diagnostic technique; nothing can replace the hands and the senses. Furthermore, in questioning and examining a patient, the diagnostic ideas generated by the one are tested by the other. Both history taking and physical examination have sensitivity and speci¤city as do any other source of facts. These methods can be supplemented but not replaced by the mediated investigations offered by modern tests and imaging. For clinicians, all of the information gathered by these observations is given its meaning by the focus on the patient; it is not permitted to have a life of its own or, conversely, to de¤ne the patient. On the other hand, contrary to the beliefs of the era of Locke and Sydenham it is now known that no one can see anything * What are the limits of medicine? If the logical focus of medicine is the patient, does that mean that anything and everything amiss discovered by observation belongs in the domain of medicine? I believe Engelhardt was correct when he said that the domain of medicine is de¤ned more by the vexations and complaints brought by patients than by pathoanatomists, pathophysiologists, and pathopsychologists. On the other hand, investigation of the patient’s vexation may reveal, for example, its spiritual nature, the treatment of which is not manifestly in medicine’s realm. (This is not to say that physicians are not concerned with the transcendent dimensions of humankind; they are. But when the patient’s torments are primarily spiritual, others may be better suited for their relief.) Medicine is historically rooted in the body and its af¡ictions, and I do not believe that this situation is changing or will change.

The Clinical Method

[89]

as the thing it is without concepts, and meaning is automatically attached to everything observed.

Clinical Information The clinician’s task of observation and the educational problem posed by teaching observational skills are complicated by the twentieth-century understandings that all information has meaning as the mind receives it; that, as received, this information is part of a whole; and that it carries its symbolism—its meanings—with it as it is thought about. Restated simply, in everyday life information is not divorced from thinking about it. A case makes the point. Mary Franklin was a seventy-year-old married woman with rheumatoid arthritis. She has visited one doctor or another almost weekly for years. She developed bronchitis, its cough persisting despite the use of antibiotics and bronchodilators. The cough was spasmodic and nonproductive. After several weeks of coughing, she developed pain in the right side of the chest. Examination revealed local tenderness. The pain was attributed to the cough. The cough ¤nally subsided, but the pain continued, causing her to call the doctor several times. She decided, and the doctor concurred, that it was her arthritis. She took codeine, and when that was inadequate, she switched to oxycodone, taking it several times a day. Tenderness over the painful rib persisted. X-rays of her chest and ribs were negative and a bone scan showed extensive arthritis, but the painful rib did not light up. Her escalating use of analgesics continued, as did her endless telephone calls. She decided that her doctor no longer cared about her and found a new physician. This doctor also became quickly aware of her behavior and analgesic usage. The tender rib was noted and the X-rays and bone scan reviewed. A protein electrophoresis was performed on the ¤rst visit. Multiple myeloma was discovered.

For the ¤rst doctor, the observed facts of pain of long duration even after the cough disappeared, considerable analgesic requirement, and tender rib were not separated from his picture of Mary as a complainer. The doctor’s interpretation, reinforced by the negative X-ray and bone scan, was that she had found something else to complain about and that she liked her Percodan. No one practices medicine very long without making a similar error in judgment. This is another way of saying that in the absence

[90]

doctoring

of special training, it takes most physicians many years to learn to separate the facts of the case from their preconceptions—to separate their observations from their interpretations. This is the basic lesson of observation. On the other hand, if nothing can be observed apart from its meanings and symbolism or the whole of which it is a part, how can any pure fact be discovered? The answer is that facts are not discovered, in clinical medicine, as much as disengaged from their “invisibility”—the almost automatic assignment of meaning, interpretation, and value that occurs with any perception. For example, in looking at skin color, we are able to gauge the color for what it is because of our knowledge of skin and its colors. No skin is uniform in hue like a color chip in a paint store. If the skin is pale, that is not a color, but a change from normal expectations. A bad sunburn on a black-skinned person is different than on a whiteskinned person, yet the interpretation “bad sunburn” is derived from skin color. Pale black-skinned individuals look different from pale white-skinned individuals, yet both are pale—unlike the expected. The skin of white or black children looks and feels different from the skin of adults of the same color. Whites and blacks are neither white nor black, and Asians are not yellow. Skin colors, as everyone knows, are not merely hues but also have considerable individual and social meaning. To truly observe the skin is to disengage the sensations of sight and touch from all these other meanings, most of which are below the level of awareness until they are made conscious. To cite another example, the feel of the prostate to the examining ¤nger in the rectum must be disengaged from the other associations of the rectum and the interference of the glove. Initially, during training, this requires conscious effort, and ultimately it occurs automatically, but at no time are these other meanings nonexistent. In these examples—skin color and the feel of the prostate through the wall of the rectum—the observation is made with a purpose. Is the person anemic or is the prostate normal? In both instances, the observation has been given a meaning as part of a medical whole. The observer is not merely feeling the anterior wall of the rectum but attempting to determine the state of the posterior prostate. In order to do this, some standard must exist within the observer against which the observation is being compared. This standard gives the determination not only a meaning, but also a place in a larger system—prostatic disease—and a value—healthy,

The Clinical Method

[91]

enlarged, bad, cancer—all of which terms have values attached to them. On the other hand, the more the larger system—for example, prostatic disease—determines the observation, the further removed the observer is from a disengaged fact—what a prostate actually feels like through the rectal mucosa—and from ideally developed skills of observation and the method of clinical empiricism. The primary standard could have been, should have been, other prostates and other rectal mucosae felt by the examiner. There are always interpretations; it cannot be otherwise. It is the level of interpretation to which the observer leaps that makes the difference. The greater the complexity of meanings, the further the observer is from the original interpretation. The closer the observation is to the phenomenon observed, the more readily it can be used to formulate new knowledge. If I have my hand on the painful back and remain aware of what groups of muscles are in spasm, over time I begin to understand the relation of spasm to pain and move away from more global interpretations of disk disease or similar diagnoses. Listening to a patient narrate events is also observing. Here, what the patient actually said and the exact manner—words, pausing, pitch, emphasis, and so on—in which it was said is the observation that must be separated from what the listener thinks the patient meant. I emphasize this point because ¤ndings from research on clinical thinking have been employed to support a different conclusion. Studies of clinical reasoning have shown that diagnostic hypotheses are generated very early in the data-gathering process.3 The clinician then goes back to gather more data to con¤rm or expand the hypothesis. By the time the physical examination is done, it is often completely focused by the hypotheses raised in the give-and-take of the history. In real life, the doctor comes away from the process not only with one or more strong diagnostic possibilities, but also with a ¤rm idea of what kind of person the patient is. Alas, the facts themselves are too often lost in the hunt for good diagnostic hypotheses. Observations by investigators of clinical reasoning that showed how quickly clinicians come to conclusions about the disease process seem to have created the belief that it is desirable to start generating hypotheses early on. (See the series on clinical problem solving appearing monthly in The New England Journal of Medicine. The clinician is given a fragment of history and then is asked to state what diagnostic possibilities come to mind. Then comes a tumble of

[92]

doctoring

disease possibilities, and the next phase of the history and physical examination is presented.) Two things happen in these didactic exercises that mimic investigators’ beliefs about ideal clinical reasoning: the logical focus is on the disease rather than the sick person, and the particulars are lost in the jump to the generalization. From the point of view of teaching primary care medicine, both are important errors. As clinicians, we are not interested in what disease exists, but rather in what stage of what disease process in which person, and why, and why now. Our goal is not merely to name the disease, but also to make the sick person better as a result of our actions (even if the patient is about to die). To do that, the particulars, the facts of history taking, clinical observation of the patient, physical examination, and tests and measurements must remain in dynamic balance with the diagnostic judgments about the disease process and the sick person. The idea of merely making a diagnosis is, quite simply, passé. Hypotheses cannot be proven correct. An important function of a diagnostic hypothesis, therefore, is to be the target of attempts to disprove it. Another function is to be the postulated end of a story that causes all the preceding parts to make sense. All illnesses are stories that involve a person and the person’s body, take place in a time and a place, and are unique to the characters involved (including the physician).4 Clinicians attempt to improve how the story ends or how it will continue after their intervention. Any intervention to accomplish these purposes is legitimate if it is effective or at least based on valid evidence of effectiveness, is done with the patient’s participation or consent, promises appropriate bene¤t in proportion to risk, and is as economical of resources as possible. The story is based on the history, the physical examination, and the results of diagnostic studies. Learning to separate the observation from the interpretation is to no avail if the observation was not made in the ¤rst place. You cannot interpret what you have not seen, heard, felt, smelled, tasted, or intuited. (The meaning of the tricky word intuition used in this context will be discussed below.) The ¤rst goal of the special education of primary care physicians is to teach the empirical clinical method that the goals of primary care require—rich, deep, sophisticated clinical observation of patients. There are two other areas where classical observational skills are important in primary care. The ¤rst is the observation of impairments and dysfunctions for

The Clinical Method

[93]

which disease designations have proven inadequate. The prime example is musculoskeletal disorders of the back or limbs. As noted in Chapter 2, diagnosis and treatment of back disorders as though they were the result of, for example, abnormalities of vertebrae or intervertebral disks has led to disappointing results. This is because, whatever the origin, the problem concerns the functional whole that is the back. Another example is how disorders in the way the whole leg functions—for example, gait abnormalities—may express themselves as joint pain. Here diagnosis requires an ability to observe the body or the part in motion to begin to understand how the dysfunction came about and how it is to be treated. Observation in these circumstances rests on knowledge of anatomy and on established methods of physical diagnosis, both of which are sometimes disvalued. This points up the fact that training in primary care medicine requires reestablishing the importance of observation, which in the past was considered foundational in clinical medicine but which has languished in medical school curricula. Ultimately, however, observation is just that—seeing the phenomenon that the patient reports. A sixty-five-year-old woman had had back pain for six years. She had been to many physicians, most of whom had not helped her. More recently, a physiatrist had treated her and much of her pain was gone. Still, however, dysfunction because of pain in her left low back and intermittent sciatic radiation into the calf interfered with her life. When her latest physician took her history and asked about her health in childhood, she revealed that she had had polio at age thirteen, which affected her right leg. She had not previously offered this information. When the physician watched her walk, the shift of weight to her left leg was obvious. The right leg was slightly weaker and the muscle mass of the right buttock had less volume.

Why had her previous physicians not been attentive to her history of polio? In part, they failed to appreciate that all chronic illness has a story rooted in the past and that the active hand of the past always intrudes into the present. But a more basic reason was a theoretical construction of the causes of back pain, in which the function of the physical examination (observation) is to ¤nd evidence that the doctor’s theory is true. This is the opposite of what Sydenham meant by clinical observation; for him, observation came ¤rst and theory second. Without contradiction, one might say that the failure in this case was that an adequate history and physical examination were not done. Whether because of premature closure

[94]

doctoring

of the reasoning process or lack of attention to detail, the outcome was the same. Physicians must also learn to observe the unfolding history of chronic illness. The past emphasis on acute illness or acute events in chronic disease has placed a premium on observations of a short period of time. Watching events unfold over a long period requires a developed memory of observations and is distinctly different from thinking about an acute event. Also distinct is the experience of keeping in mind that long period when evaluating a present episode. It is unfortunate that in the present climate of managed care, with its emphasis on short-term goals and possible changes in health care plans every few years, long-term observation is not feasible.

Where Subjectivity Meets Objectivity on Level Ground* Up to this point, the observational skills I have described do not differ fundamentally from what Thomas Sydenham and John Locke would have asked of physicians. However, knowing patients suf¤ciently well to keep them in the center of the clinical process and the current broader ideas of health are different goals that do not rest on the precedent of centuries. Since here the subjectivity of both patients and physicians is of paramount importance, it is necessary to clarify the difference between objective and subjective knowledge. When we ask the patient to tell the story of the illness, the events since the last visit, or the experiences of the past that are important, the information is of many kinds. The patient’s narrative of events includes symptoms, the actions of self and others (including physicians), perhaps stories about what happened years ago, the results of previous examinations and tests, and the patient’s ideas about the illness or events, worries, concerns, desires, and interests. Some of this information is part of shared reality and is objective by any measure. Most of it is what in medicine is called subjective—not subject to measurement. I pointed * Subjectivity here refers to the introduction into judgments about both the meaning of information and the proper direction of clinical action elements—for example, meanings or values—arising within the subject that is the physician. Thus, judgments and actions are what they are because they are made by this doctor rather than another. Objectivity here refers to judgments about information and the choice of suitable clinical actions that are based solely on external evidence, untinged by the individual physician making the judgments.

The Clinical Method

[95]

out earlier that such a de¤nition of subjective and by extension, objective, is unique to medicine. Others mean by subjective that which arises within the subject and can be known only to the subject. The de¤nition of objective employed in medicine implies that, no matter who looks at them, there is a set of facts that are real, solid, and incontrovertible—things you can hang your hat on—of which scienti¤c facts about the body and disease are a part. Everything else is subjective—maybe not as real, like pain or the patient’s worries; or tainted, like memories of the past; or questionable, like interpretations of sensory information such as the color of urine. This understanding of objective is a heritage from the empiricism of the seventeenth century, which has been espoused by the positivist philosophers of science of this century and happily embraced by medical scientists. Of course, there are facts in the real world that meet all the criteria noted above. They change, however, the moment anybody—scientist, doctor, or layperson—starts to use them. It is when they are used that things begin to have symbolic value, stand for things other than themselves, and have meanings imputed by the person who is thinking about them. Some seem universal in that virtually everybody who acknowledges the fact will react the same way. In the intensive care unit is a woman who came in three weeks ago with massive pneumonia. She has been on a respirator since admission. After the pneumonia began to resolve, she developed adult respiratory distress syndrome, which has progressively worsened. She has just had her third trial off the respirator. She is a beautiful twenty-sevenyear-old married woman with a nice husband, an attentive, lovely family, and two young children. As the visiting doctors make rounds on the unit, the intensivist hands them the report of her blood gases taken when the respirator was disconnected. Her arterial blood oxygen level is 35—too low to sustain life off the respirator and an indication that she will die.

The patient’s arterial blood oxygen level is an incontrovertible objective fact, but as it tells observers that she can not exist off the respirator, it acquires unhappy meanings from which it cannot be dissociated. This is a dramatic example, but nothing happens, comes into our mind, or is acknowledged by us without being given meaning. The moment such meanings are attached, the atomistic facts of science no longer meet the ideal of the medical scientist. The ¤rst step in resolving the subjective-versus-objective dilemma is to realize that all facts in the practice of medicine, objective or

[96]

doctoring

subjective, are information on which actions are based. Information is true—useful, in this context—to the extent that it approximates the reality that it describes. This suggests that truth value is a probabilistic statement or that one can say of something that it is believable within certain con¤dence limits. Knowing this, one wants to increase the probability or narrow the con¤dence limits about some particular fact. This statement is as true of objective information as of subjective information. For this reason, we don’t necessarily believe the blood test numbers on the printout (which are unquestionably objective) unless we can be sure that it is our patient’s blood that was tested, that there is a track record of test accuracy that can be depended on, and that the result makes sense. Similarly, we don’t necessarily believe the patient’s report of dyspnea unless further questioning and evidence show the report of the symptom to be coherent with other facts, unless we have no reason to question the accuracy of the patient, and unless the report of dyspnea makes sense. When we have gone through these steps with the report of shortness of breath, we have acted to increase our con¤dence in the symptom, which is no longer subjective because it has become objective. The patient’s dyspnea, as it is experienced, is subjective. The patient’s report of the symptom is objective. The report of the symptom is a thing that can be examined as much as any other thing—for example, the result of a blood test. The patient’s desires, worries, concerns, feelings, and experience of the world are also subjective, but they too, once reported by the patient, can be examined, thought about, evaluated in the light of other information, and then imported into clinical judgments. Part of the clinical reasoning process involves assigning weight to information from any source, including these facts reported by patients. Facts not only have truth value, they have importance. A trivial piece of information does not become important because it is accurate or the result of measurement. Great importance is often given to abnormalities shown, for example, by various imaging methods because of their objective nature. The recent ¤nding that many individuals without back pain have abnormalities of their intervertebral disks demonstrated by MRI is a case in point.5 Important information, on the other hand, becomes less important in planning action if it is questionable. The patient’s history of the illness or report of symptoms, behaviors, and feelings is extremely important because it is usually the most essential element in making diag-

The Clinical Method

[97]

noses and decisions. It is necessary, therefore, that its reliability and accuracy be increased. What has been said about information arising from the subjectivity of the patient is true also of facts that arise from the subjectivity of the doctor. Ultimately, it is the report of experience that is being discussed. Earlier, it was the experience of skin color, and the experience of the sensation under the examining ¤nger in the rectum that provided the facts from observation in the instances given. We could have added examples such as hearing heart murmurs, feeling abdominal masses, or smelling the drainage from a wound; and in each case, it would have been the doctor’s experience of exterior sensory information that was under discussion. It is well known that sensory experience is ¤lled with uncertainties. One must have had experience for experience to yield its maximum value. Novices have dif¤culty hearing heart murmurs, feeling for abdominal masses, and allowing themselves to smell the wound without being repelled. Experience does three things that make this kind of sensory information more reliable. First, it provides con¤dence in the reliability of the raw, unprocessed sensory information: if I think I feel something, then I do feel something. Next, it builds con¤dence that the information is what it appears to be: I really do feel this pulsating mass in the middle of this patient’s abdomen; I have felt things like this before. Finally, seasoned physicians have learned to disengage the process of assigning value to the sensory experience so that they can accept the observation for what it is. For example, a young woman whose carcinoma of the lung was irradiated some months previously comes in for a follow-up examination. There is a new pattern of enlarged veins barely visible on her upper anterior chest wall. I don’t want to know what they indicate, so for the ¤rst few minutes I don’t acknowledge their existence. Ultimately, I grant their presence and think about the unhappy probability that the patient is developing superior vena caval syndrome because her cancer has recurred. One of the most common reasons that doctors fail to detect important ¤ndings (once they look) is that they do not want to know what their senses are telling them. For my experience to be useful, I must have concepts into which it ¤ts. Remember, all percepts are given meaning spontaneously. If the abdominal mass had no immediate meaning for me, I would ¤nd a meaning anyway. However, physicians usually do not have to grope for meanings. They know what is in the abdomen, what

[98]

doctoring

makes masses, where this one is found and where that one is never felt. They know that pulsations are arterial in origin, that the abdominal aorta is subject to aneurysms, that aortic aneurysms are more common in the elderly, that they most commonly arise below the origin of the renal arteries, and that their size is important because it is a determinant of surgery. The more one knows about abdominal aneurysms, the more information is obtained by feeling them, and the more one feels them, the more knowledge one gains about them. This reciprocal process gives rise to the saying that percepts without concepts are blind and concepts without percepts are empty.* In addition to needing previous experience and knowledge for experience to be most useful, there is a ¤nal requirement: a language to describe the experience to oneself and others. As they learn medicine, doctors acquire a rich descriptive language for the things they come to know. There are jokes about the fact that doctors use the words millet-sized to describe the X-ray appearance of certain lung lesions when they have never seen a millet seed. They may not have seen one, but they know its size because of the X-rays they have seen. Millet does not, in this case, refer back to a seed but to a size. Every specialty has a language to describe its distinctive pathology. Very few non-eye doctors speak ophthalmology, and the same thing is true of dermatology. This language use effectively builds a fence around those specialties. It also points out that one of the reasons most physicians know so little about dermatology is that they do not possess the vocabulary. After saying the words macule, papule, and scaly they generally run out of terms. Without a language they cannot describe lesions to themselves, much less to others, and thus cannot look things up in books and cannot search the literature. Clearly, concepts and descriptive language are related, but they are not the same. Language clari¤es experience and helps direct it to the appropriate conceptual nexus. * This paragraph may seem to contradict the earlier discussion about the importance of separating observation from interpretation. Although interpretation is being made, the good observer dwells on the sensory information without leaping several levels up during interpretation. Rather than saying, “I think this is an aneurysm that is larger than ¤ve centimeters; I’d better get a sonogram and make arrangements for a consultation,” all of which may true, the good observer continues to feel the pulsating mass to get as much information as possible and then continues to examine the remainder of the abdomen carefully. After that is done, the doctor considers the best way to proceed.

The Clinical Method

[99]

The rich, varied language for pain used by physicians helps them clarify the patient’s experience so that they can arrive at a probable reason for the discomfort. The professionalization of experience that training and clinical seasoning provide extends to the evaluation of the experience of others. Doctors do the same thing for their own experience that they do when they question their patients. This is why interviewing requires skill; it causes the patient to provide subjective information for reexamination by the physician. To elicit this information, doctors must disentangle the patient’s interpretation of events from the original observations, which generally remain available for reinterpretation.6

Clinical Judgment In all of the cases cited, the patients’ present and future were in¡uenced by the actions of their physicians based on the physicians’ judgment. Interestingly, the basis for the physician’s actions are usually called decisions rather than judgments. This is strange, because part of examining patients for their cognitive intactness involves testing their judgment, not their decision-making ability. I have checked several current books on clinical epidemiology and decision making and the word judgment was not indexed. The dictionary states that a decision is the act of forming an opinion. Judgment, similarly, is forming an opinion or evaluation by discerning and comparing. The act of judgment has been the focus of much dispute among philosophers of logic and thought, particularly in the ¤rst half of this century, in part because of attempts to get rid of the aspect of the personal that seemed to many a contaminant of the process. Thus, in current philosophical terminology, what is usually called ‘a judgment’ would be named ‘a proposition.’ That will not work in medicine, as this next case illustrates. Rachel Graber is a sixty-two-year-old woman with end-stage carcinoma of the ovary for which she refused treatment. When it became dif¤cult for her to get around, she took an overdose of sleeping pills. She was discovered still breathing the next morning, and a note at her bedside told what she had done and why. The medication was not identi¤ed. She was brought to a local hospital, intubated, and placed on a respirator. She has no family, but her attorney was identi¤ed as her spokesperson. He had her recently written living will which

[100]

doctoring

speci¤cally refused resuscitation for any reason, but in the absence of a health care proxy statement, the hospital would not honor it. Physical examination revealed her abdomen to be distended with ascites; the liver was enlarged and hard, and the lower quadrants were occupied by ¤rm masses. Initially thought to be brain dead because of a ¡at electroen cephalogram, she recovered consciousness within seventy-two hours. A pneumonic in¤ltrate was seen on X-ray, and intravenous antibiotics were started. On the ¤fth hospital day she requested that treatment be stopped, but her request was not honored and she was placed in restraints. On the ninth hospital day, at her request, she was transferred to the inpatient hospice unit of another hospital prior to discharge to outpatient hospice care.

Whatever one believes to be the proper course of action in the care of this patient, the decisions are dif¤cult and cannot be made solely on the basis of medical science. If you think she should have been resuscitated, or treated for pneumonia, or restrained, then it follows that you believe her wishes should not be honored. If you believe that she should not have been put on a respirator, restrained, or given antibiotics, then you believe that her personal objectives override the pathophysiological demands of her state. Either way, science and technical ability alone will not help here because moral questions, issues of value, are present that cannot be resolved scienti¤cally. The practice of medicine, in general, is a moral-technical endeavor precisely because doctors’ actions are directed at the good of the person, not just the person’s disease. This is important in acute illness but fundamental in chronic illness. It is what makes clinical practice different from medical science. Learning how to act in the best short- and long-term interests of the patient—developing clinical wisdom—is what distinguishes clinicians. This is true of all of medicine, but of course, particularly of primary care, where the person is both the object and the subject of practice. Twenty-¤ve hundred years ago, Aristotle, who was the son of a physician, called the capacity to make these kinds of decisions practical wisdom: bringing general knowledge to bear on a particular human problem through reasoned deliberation in order to act for the good of someone. He speci¤cally distinguished the ability to do this from either possessing knowledge or technical ability.7 The issue of judgment or clinical wisdom seems to represent a problem for medicine because the implication is that these are personal, subjective actions of individual physicians that are idiosyncratic in nature. It is just this prob-

The Clinical Method

[101]

lem of clinical medicine that objective medical science was meant to solve. To some, the choice is straightforward: either a permanent framework of science and objective fact on which to base medical practice or the chaos of a medicine based on personal opinions or self-proclaimed authority.* The recent spate of guidelines for clinical action covering many diseases and states, and the new call for evidence-based medicine (including the founding of a new journal of the same name), are current examples of attempts to get around the necessity for the judgment of individual physicians.8 While guidelines and the clinical epidemiology from which evidence-based medicine is derived may represent interesting educational and research initiatives, differences in patients, physicians, and context make individual decisions as essential as ever. We must face up to the personal side of medicine (in both patient and physician) and call the process what it is, clinical judgment, just as Alvan Feinstein did thirty years ago when he introduced systematic thinking into the subject.9 More important than impossible attempts to eliminate clinical wisdom are attempts to understand the process as a basis for better training and inclusion in the clinical method. One problem with considering judgment is that the idea shares with aesthetic taste and moral sensibility the notion that it is an intellectual virtue that one either has or does not have; that judgments do not obey reason but are more like sentiment. It is true that the logical basis of judgment—subsuming a particular under a universal, recognizing something as an example of a rule—cannot be demonstrated. For this reason, judgment cannot be taught in the abstract, but only taught and practiced from case to case. We can see why the belief has arisen that judgments are not only private but * This is not a problem for medicine alone. The end of the nineteenth century and the beginning of the twentieth century saw the development of successful science and a belief in the power of empirical knowledge. In recent decades an increasingly strong case has been made for the fact that thc concepts of truth, rationality, reality, and others must be understood as relative to speci¤c contexts, cultures, societies, and conceptual schemes. The claim that science, and science alone, is the measure of reality, knowledge, and truth is in con¡ict with the understanding that there are other legitimate sources of knowledge and experience found in the classical humanistic disciplines and the tradition of practical wisdom. As the twentieth century draws to a close, a more balanced view seems to be emerging that cannot help but in¡uence medicine. (Richard J. Berstein, Beyond Objectivism and Relativism: Science, Hermeneutics, and Praxis, Philadelphia: University of Pennsylvania Press, 1988.)

[102]

doctoring

idiosyncratic. Doctors, even in the privacy of their minds, however, are not alone with their thoughts. They are members of the community of medicine, as well as members of their larger communities. They are never free of the socialization provided by the training that shapes the horizon not only of their clinical judgments, but of their concepts and perceptions as well. Their thinking, my thinking, and your thinking share in the sense of the community and its culture.10 It is an error, quickly corrected by re¡ection, to believe that taste is not a matter of dispute. Taste can be re¤ned, however, by experience and training; so can judgment. The clinical wisdom and judgments of physicians, while clearly personal, are not merely private and idiosyncratic. As Hannah Arendt says: The power of judgment rests on a potential agreement with others, and the thinking process which is active in judging something is not, like the thought process of pure reasoning, a dialogue between me and myself, but ¤nds itself always and primarily, even if I am quite alone in making up my mind, in an anticipated communication with others with whom I know I must come to some agreement. From this potential agreement, judgment derives its speci¤c validity. This means, on the one hand, that such judgment must liberate itself from the “subjective private conditions,” that is, from the idiosyncrasies which naturally determine the outlook of each individual in his privacy and are legitimate as long as they are only privately held opinions, but which are not ¤t to enter the marketplace, and lack all validity in the public realm. And this enlarged way of thinking, which as judgment knows how to transcend its own individual limitations, on the other hand, cannot function in strict isolation or solitude; it needs the presence of others “in whose place” it must think, whose perspectives it must take into consideration, and without whom it never has the opportunity to operate at all.11

As surgery takes place in a public amphitheater, physicians have the presence of at least two important groups whose perspectives they must take into consideration in making their judgments: their patients and their colleagues, to say nothing of insurance companies, managed care directors, utilization committees, and all the other people who now know what happens between patients and doctors. Doctors begin in their training to learn that their judgments are not merely private matters. In my experience, however, explicit training in judgment is uncommon. Instead, house of¤cers are dif¤dent about expressing an opinion on a chart, of writing a note that states

The Clinical Method

[103]

their judgment and the reasons on which it is based. In primary care, and in all medicine, the time has come to confront the issue of training judgment. The training and experience of doctors have altered their relationship to the world of patients, disease, and illness. They do not merely know different things; they themselves are different fromnonphysicians in relation to their professional universe. If that world changes, they must not merely learn new things, but also change within themselves. As we start to discuss what it means to know sick or well persons as the logical center of the medical endeavor in the next chapter, it is necessary to understand that in the course of their training, primary care physicians are being changed as doctor-persons differently than their colleagues in other specialities. To answer the question of how this change will occur, let me start by referring to the somewhat misleading phrase used earlier: “where subjectivity meets objectivity on level ground.” The implication, referring to medicine’s well-known bias about subjective information, is that subjectivity and subjective information will be of¤cially introduced into medicine in relation to knowing persons, and that it will be as effective as objective information. As the preceding pages should have made clear, the phrase is misleading because subjective information has been as essential to clinicians as objective information from the beginning of time, and the subjectivity of the doctor as an experiencing, valuing, judging subject cannot be divorced from medical practice. Of course, it is wonderful to see a measure of forced expiratory ¡ow rate rather than estimating it indirectly from the patient’s statements, the movement of the chest during respiration, and the sounds heard through the stethoscope. In doing research on the treatment of asthma, nothing is better than the objective measurement of expiratory ¡ow rate for comparison between patients. Its measurement narrows the con¤dence limits and increases the probability that the physician knows how much air the patient can move in a unit of time. For the treatment of this patient, however, the measurement of expiratory ¡ow rate must take its place alongside many other factors in the judgment of a particular subject, the doctor. Consider this case. A forty-two-year-old derelict lying on a stretcher in the Bellevue emergency ward said that an hour earlier he had been walking on the street when he suddenly became severely short of breath and could not walk. He leaned against a building for a while and gradu-

[104]

doctoring

ally felt better. His extreme anxiety did not subside, however, so he came to the hospital. There had been no other similar episodes, and he considered himself “a healthy drunk.” He had not previously come to the emergency room and had last seen a physician years earlier. His blood pressure was 220/130. There were no other ¤ndings of note except for his general ¤lthiness, tachycardia, and cut and bruised legs. An hour later the resident asked the intern why nothing had been done. Where were the chest X-ray, ECG, blood gases? Only blood had been drawn and an IV started by the nurse. The intern explained that the man was a bum and had come into the emergency room because it was cold outside; and besides, there was nothing to ¤nd on examination. Spurred on by a very angry resident who had taken his own history and warned how sick the patient was, no matter what he looked like or who he was, the tests were completed just before the patient went back into pulmonary edema.

There is no possibility of clinical medicine without the judgments of individual physicians. In primary care as in the rest of medicine, disciplined judgment is part of the clinical method. We can either actively train medical students in clinical judgment, toward the goal of clinical wisdom, or trust native ability and fortunate experience. Judgment requires information gained from examining patients, regardless of the dif¤culties caused by the subjectivity versus objectivity. Subjectivity meets objectivity on level ground within physicians when they are adequately trained in the clinical method. Then they will be able to ensure that the information on which they base their decisions and actions—whatever its source—has maximum truth value and narrow con¤dence limits, and that their judgments are as free from personal bias as possible.*

* How is a bias different from subjectivity in general? A bias is a prejudice that generally prevents the person from seeing objects, persons, relationships, or situations in any manner other than that suggested by the preconception. All perception and subjectivity implies preexisting concepts or ideas, but these can be changed to allow the observer to see the percept in another fashion or from another perspective. Biased perceptions or thinking are altered only with dif¤culty. An important part of training teaches physicians to learn to acknowledge their biases and let go of them so that their clinical judgments are undamaged. If they want to retain their biases in their personal lives, that is a personal matter.

The Clinical Method and the Patient

[105]










5 The Clinical Method and the Patient








T

he question to be answered is this: how can the clinical method and its techniques of observation that have served medicine so well be extended to the person of the patient? The beginning of the answer is another question. What do clinicians need to know about their patients? The answer involves more questions. What is it about this patient—socioeconomic status, education, occupational factors, home environment, behavioral patterns, and so on—that in¡uences the disease and the illness? What is the pattern of this patient’s life with which medical interventions should best ¤t? What must the doctor know about this patient in order to make a therapeutic alliance? What do clinicians need to know about themselves? Generally clinicians are not being asked to make an assessment of personality or character, approve or disapprove of behaviors, or like or dislike their patients. Let me begin with two examples that demonstrate this point. Vivian Progoff is a seventy-four-year-old woman with sarcoidosis of the lungs (a disease similar to tuberculosis but not infectious) diagnosed two years ago. Her doctor, who has been taking care of her for more than twenty years, believed her to have two overwhelming passions: her daughter and her appearance. Prior to the onset of sarcoidosis, she had never had an important illness. She

[105]

[106]

doctoring

characterizes herself as a worrier and a hypochondriac, but she has visited her physician infrequently; she says that she is afraid of doctors. When she does have an illness, however mild, she is articulate about her impending doom. The diagnosis of sarcoidosis was made after a persistent cough forced her to see the physician. The X-ray was suggestive, so she had a computed tomography (CT) scan of the chest, saw two consultants, was asked to return frequently, and had a con¤rmatory biopsy by mediastinoscopy. During her workup she developed Bell’s palsy, which cleared very slowly but almost completely. Her disease remained stable and required no treatment. Adult-onset diabetes was diagnosed about the same time as the sarcoid. Throughout this period she made endless telephone calls to the doctor to calm her fears, positive that she would never live through the experience (even her daughter could not reassure her). After about ¤ve months, her cough cleared and the calls stopped. On her most recent routine visit, she characterized the last ten months (which had been without symptoms, unlike the previous illness-ridden times) as the worst of her whole life. Why? The very slight residual droop of her right eyebrow occasionally interferes with her vision, and she has required several unpleasant dental procedures.

This is the kind of patient doctors generally do not like. She is demanding and vain. She gets as upset by trivia as by things that really threaten her. She consumes time and requires endless patience. But, despite all the Sturm und Drang, fears of impending doom because of her chest disease, the undoubted blow to her vanity of the Bell’s palsy, and the shock of her diabetes, Mrs. Progoff did everything necessary because of her diseases and in a timely manner. Why? For ¤ve months she required more telephone calls than most patients, her fears required considerable attention, and she needed the reassurance that the doctor would not abandon her—nothing more. A therapeutic alliance was formed. Walter Harnick is a ¤fty-one-year-old executive with a ¤ve-year history of diabetes. He runs a large corporation and sits on the boards of many businesses and civic organizations. His opinions about business and politics are valued by many leaders. He does not take care of his diabetes, however, does not follow a diet, exercise, or test his urine or blood, and does not keep doctor’s appointments. Pushed into seeing his physician because of a syncopal episode, his glycohemoglobin was 13.8 (very high!). He adamantly refuses to take insulin but says that he now realizes he must be

The Clinical Method and the Patient

[107]

more careful. However, he did not follow up on the advice given or keep his next appointment.

Admirable, handsome, likable, and seemingly much easier to deal with than Mrs. Progoff, he is heading for real trouble. His care, if success is possible, will require endless effort, but a real therapeutic alliance of mutual trust is unlikely. Yet, most physicians would rather take care of Walter Harnick. These two cases highlight a common error. Witness how the letters from consultants commonly end— for example, “Thank you for permitting me to take part in the care of this lovely lady.” What has “lovely lady” to do with the consultation or the patient’s care? How different are relationships in everyday life! There we are concerned about personality and character and everyday behaviors. This is how we choose our friends and associates, make work decisions, and steer our associations with others. These features of persons are the essence of the value judgments we make all the time in order to get through life. We can see the difference between physicians’ relationships to their patients and to persons in everyday life by using as an example the relationship to the body in everyday life. For most people, the body is their body and they have a personal relationship to it. In the same way, we have a personal relationship with other people, whatever place they have in our lives. Physicians, on the other hand, in the course of medical school and socialization, learn about the body, an impersonal object of professional interest apart from their own personal bodies. In the same manner, primary care physicians (preferably, all physicians) must learn about the person, an impersonal “object” of professional interest apart from persons in everyday life. There are real differences, of course, between persons and bodies that make the task of knowing persons as objects very different from knowing about bodies. The ¤rst and most important difference is that if persons are treated as impersonal objects, they cannot be known in the way required to optimize their care. The second difference is that only a person can know another person, and if medical knowledge of persons is to enter medicine differently from the everyday knowledge of persons, the doctor persona—doctor as doctor, Eric Cassell as doctor—must be developed and distinct from the doctor’s everyday person—Eric Cassell as husband, friend, or neighbor. For years, I took exception when people asked me what I, “as a doctor,” thought about something. I took pains to explain that while I might be a physician, there was only one me. Wrong. There is one

[108]

doctoring

person, perhaps, but more than one persona, one of which is a doctor. Physicianhood is a role—a set of performances, duties, obligations, entitlements, and limitations connected to a function or status. The role has its social counterpart in the expectations of the role that patients and other people have that make the performance of its duties possible and impose its obligations and limitations. It is a special role that makes possible the special, almost magical connection that constitutes the doctor–patient relationship.1 The role has historical and ethical dimensions that have led one author to liken it to a covenant in which doctors embrace the norms, knowledge, and historical expectations of the profession in return for its powers, its status, and the protection of their personal health and happiness that a professional identity provides.2 Early in training, medical students suffer the emotional pains that come from taking the sicknesses and deaths of their patients personally. As their training continues, they develop a professional self-image, as part of which they become increasingly immune to the emotional distress of their patients; unfortunately, they do it in a self-defeating way.3 As with so many other aspects of doctoring that are not explicitly taught, the doctor’s identity as a physician—the physician persona thus acquired—frequently has ¡aws. Probably the most common ¡aw is the doctor’s relationship to the emotions of patients. The world of sickness is ¤lled with emotions—for example, fear, sadness, grief, relief, dread, happiness.* The variety and intensity of emotions in the medical setting are demanding. They can be upsetting, or even overwhelming for inexperienced students or physicians. The usual reaction to this stress, explicitly reinforced by many teachers at all levels, as well as by the * The words feelings and emotions are often used as synonyms. I believe it is important to understand their differences in the context of training primary care physicians. Emotion and affect can be considered synonymous. I might say of someone, on the other hand, that “I have a good feeling about him,” but not mean that I feel happy or some other emotion. “I have an uneasy feeling about this case,” but as I say that, I am unaware of any particular emotion. It is the impression someone gets who is observing or experiencing. The word feeling is sometimes employed in aesthetics to represent the idea the viewer gets from a painting. We use feeling, as above, to describe the effect on us of a situation or an event. In medicine, feelings are experienced as qualities of our patients or clinical situations that are not vague or inde¤nite at all but have a concrete and particular character. They do, however, resist conceptual, systematic, or logical treatment. It is with this category of feeling that we will be dealing; separately but in addition to mood and emotion. (Susanne K. Langer, Feeling and Form, New York: Charles Scribner’s Sons, 1953.)

The Clinical Method and the Patient

[109]

medical care setting itself, is to suppress the emotions and distance themselves from the patient. Before long the doctor is able to deal with the worst of human misery, emotionally distanced from the patients and with all negative affect repressed.* In the absence of awareness of patients’ emotions or of their own affect, doctors lose an important tool for knowing patients for which no other can substitute. The ability to know patients’ emotions is a learned skill. It depends primarily, however, on the doctor’s ability to be consciously aware of his or her own affect and emotional state. If this means that doctors will again be subject to all the unhappy consequences of the sea of emotion in which they work, what is to protect them? Here is where the distinction between the doctor as a separate self or persona and the everyday person is crucial and must be emphasized in training. It is sad when one of my patients dies but not more than that. The patient was not a member of my family or even a friend, even if he or she has been a patient for many years. A patient as a patient is not a friend even if the patient is a friend. A friend who is a patient is a patient while a patient and a friend in other settings. (This may be a dif¤cult distinction to maintain on occasion. When I see my friends who are also patients in personal settings, I literally cannot remember intimate medical details about them.) When a patient lies to me, it is not Eric Cassell, the person, being lied to, it is the doctor. Lying has a different meaning in this setting than in everyday life. It tells something about the patient in relation to sickness and medical care. Seductiveness directed at me is also about the reaction of the patient to sickness or to relationships with doctors. Sexual feelings are prevalent in medical settings, but they are much more often related to the intimate connection established in the care of the sick than about sexuality per se. Knowing these things increases a doctor’s diagnostic and therapeutic effectiveness: it makes no sense to merely hope that a doctor will * A high personal price is paid for this solution to the problem of emotions. As physicians repress the unpleasant emotions of their patient, they also repress the good affects, including the patients’ emotional support of their physicians. They don’t get the pleasure of gratitude or enjoy the emotions of successful care of patients. They may also be so effective in repressing their affect arising in the clinical setting that they are unable to enjoy the emotions of family life or be aware of the negative affects in personal life—for example, the deadening of their reaction to their spouses’ emotions—that warn of the necessity for psychological repair of oneself or one’s intimate relationships.

[110]

doctoring

¤nally learn them, if at all, after twenty years of practice. At a hospice meeting I attended, a physician in his forties got up to say something extremely important to him that he had never been able to talk about. He felt it necessary to confess that he had once had very strong sexual feelings for a dying young woman in his care. The intensely close connection necessary for effective care of the dying is often interpreted as and feels like sexual arousal. All perceptions, even of feelings, are given meaning, and meaning is expressed in all dimensions of personhood, including physically. Most people have no experience of such closeness except sexually, so that is how they feel it in the clinical situation, often to their embarrassment. All that anguish occurred because no one had ever taught this physician how common sexual feelings were in the care of the sick. In general, the hidden (and sometimes explicit) sexuality present in the medical setting must be mastered by clinicians if they are to be maximally effective and comfortable in their work. Conversations with sexual meanings, handling and examining bodies, breasts, and genitalia, as well as the attractiveness of some patients to doctors and doctors to patients, brings sex into the thoughts of many physicians on frequent occasions. If physicians-in-training are not actively taught to deal with this problem, many end up ducking discussions of their patients’ sexual problems and avoiding examinations of sexual organs. Yet I have rarely encountered doctors who were taught about the problem of sex in medical practice. There are some things one should not have to learn about by oneself. It is impossible to teach physicians how to be their own instrument while avoiding the issue of sex. Training programs can teach physicians to be their own instrument if they understand the task and its importance. Teaching the concept and the speci¤c skills requires a supportive atmosphere because of the undeniable dif¤culties such a goal poses for most medical students and recent graduates; the previous paragraph about sex provides an example. The problems faced are highlighted by the discussion in Chapter 3 of the effect on physicians of technology. There it was pointed out that technology is overused, in part, because of the certainty that it seems to offer in an otherwise uncertain world; by the what-you-see-iswhat-you-get quality of immediacy that attends its use; by the unambiguous values it represents; by how fascinated doctors become by its new, shiny, and exciting features; and by how it confers power on its

The Clinical Method and the Patient

[111]

users. Some of these dif¤culties are best solved by focusing on the patient as the logical center of medicine and by speci¤c training in the use of different modalities. On the other hand, managing uncertainty is a personal challenge to even the best physicians. There are techniques of reasoning that help reduce its impact, but ultimately it is dealt with within oneself. Managing uncertainty, for example, requires a kind of training different from that of teaching technical medicine. Trainees are being supported while they personally change and while they learn to depend on kinds of information arising from within themselves that were often deprecated in medical school. The highly competitive milieu of many programs is not conducive to this type of training, particularly as the physicians are learning to utilize their emotions professionally in a manner that might have been deemed weakness in their previous training.

Knowledge from Patients There are four distinct kinds of information, apart from brute facts, that doctors acquire from their patients—information that tells them about the patient as this individual patient—meanings, emotions, aesthetics, and intuitions. meanings Meanings are at once the most accessible and most complex type of information. There are at least two senses of the word meaning that are important to physicians. The ¤rst sense is signi¤cance: numbness in the toes of the diabetic may signify diabetic neuropathy; anterior pressing chest pain with effort may mean angina. The alarm clock ringing means that it is time to get up. The second sense is importance. Diabetic neuropathy means that my diabetes is ¤nally getting me. Angina with so little effort means that I can hardly do anything anymore. The alarm already? I don’t want to get up. The importance of something is always personal because it is always important to someone (even though the thing may be important in the same way to many others). Importance is another way of saying value; all meanings have an element of value. When we know what an event means to people, we know both their beliefs about the thing and how they value it on the

[112]

doctoring

scale of good and bad or right and wrong. If one knows a lot about the beliefs and values of someone, one knows a lot about the person. When people perceive something about their bodies or functioning which seems wrong—perhaps a strange-looking thing, or an odd sensation or discomfort, or a change in breathing—the ¤rst interpretation is that something is different. Usually, an assignment of meaning quickly follows—for example, the strange-looking thing is cancer, the discomfort is from the kidneys, the changed breathing is asthma. People call or visit doctors because they interpret as illness, or as something wrong, alterations in their customary sense of themselves or of their function. How something so perceived becomes a symptom has a complex relation to the person’s ideas of normal being or function, what illness is, and what it is proper to present to doctors. Physicians’ understandings of symptoms are quite different. Doctors act as though a symptom is the direct effect of the disease on the person, as though a symptom is the voice of the disease through the patient’s mouth. These differences are a source of frustration to both patients and doctors. How symptoms are formed, and the differences in their meaning to doctors and patients, is discussed at length in Chapter 7 of The Nature of Suffering. It is common for patients not to tell physicians directly about their observations—for example, the cough or tightness in the chest—but to go directly to the interpretation, “I think I have bronchitis.” Most often the interpretation is familiar, a diagnosis or a disease name, but sometimes it is so strange that the doctor may disregard the original observation. This case is an example. Joseph Heelon is a thirty-one-year-old single carpenter whose union provided a physical examination. He said that he had been to a lot of doctors with the same problem: in the last two or three years he’s noted a connection between ejaculation and phlegm. Phlegm rises into his sinuses and into the scars on his forehead and makes them puffy. It also ¤lls up his nasal passages and his chest. When he doesn’t ejaculate, he has much less phlegm. Because of the effect of orgasms he was thinking of giving up his girlfriend, but the problem occurred even when he masturbated. He had become increasingly concerned about the symptom and the effect of sex on his body. The doctors always told him that he was perfectly healthy and that there was no such thing as what he was reporting. Further questioning revealed that he had a perpetually “congested head”

The Clinical Method and the Patient

[113]

and a stuffed nose that was worse at work (where wood dust was always in the air), apparently associated with a full feeling in his head when he had an orgasm. Examination revealed well-healed 6-cm scars above each eyebrow (the result of an auto accident years earlier). The nasal mucous membranes were reddened, glairy and boggy, with a thin, watery mucus discharge. The appearance was that of allergic rhinitis. The remainder of the examination was negative. The patient was advised to be more scrupulous using dust masks at work and treated with inhaled nasal corticosteroids. The pathophysiology was explained to him. Over the course of several months, the symptoms disappeared and the nasal mucous membranes became normal.

Joseph’s previous doctors had made the common error of accepting the patient’s strange interpretation as the symptom itself and treating him accordingly. As we all know, when people start assigning meaning to events, they rarely stop at a simple naming of the phenomenon. Instead, one interpretation follows another, usually ending with the most dire one that the imagination can conceive. Whatever meaning is assigned, the original perception is usually available in memory for reexamination, allowing the physician to give it a meaning in medical terms. The meanings related by patients, as with everyone, are not only connected to symptoms, but are presented in relation to every object, event, or relationship that enters a conversation. Meaning, especially the evaluative component, is also conveyed by gestures, body language, and other nonverbal communications. In virtually every doctor–patient interaction, whether full¡edged history taking or the briefest telephone conversation, these two systems of meaning coexist. The interconnected set of the patient’s meanings has been called the life world, as opposed to the doctor’s world of medical meanings.* In order to know the patient, the life world cannot be disregarded. In times past, doctors’ questions and observations extracted only the kernel of disease, diagnosis, and treatment-related facts and discarded the rich layers of personal meanings in which those facts were wrapped. Primary care physicians cannot continue this * Many authors have discussed the importance of the life world of the patient and the opportunities for optimal medical care that are lost when this information is disregarded. One of the best such discussions is in Kay Toomb’s book, The Meaning of Illness (Dordrecht, the Netherlands: Reidel, 1992).

[114]

doctoring

practice. It is dif¤cult for physicians to make therapeutic alliances with patients whose meanings are disregarded or discarded. The physician listens to the patient and hears the meanings attached to the manifestations of disease and the everyday life events in which they are embedded. The doctor’s questions and explanations should be framed with the assumption that the patient’s world is the world. You, the doctor, place yourself within the patient’s system of beliefs to the extent possible. This does not suggest that you agree with or endorse the patient’s beliefs. You are not there at all; the doctor is. In fact, the patient’s ideas may be abhorrent from the lifeworld perspective of the everyday person who is the doctor. What does matter is the doctor who is asking questions and offering explanations. Medical explanations in technical terms or in jargon fail not only because the patient usually does not understand them, but also because they show that the doctor does not understand the patient. Otherwise, why would the doctor remain in a world external to the patient? Why should patients trust the advice of someone who doesn’t understand them? Doctors often believe that patients listen to them because they know medicine. Patients take that for granted. The patient’s actions, behaviors, and purposes are another way in which meaning is displayed. There are many levels of meaning within everyone, and only some of them can be brought to awareness. Sometimes people assign two or more con¡icting sets of meanings to the same thing. Some determinants of action are unconscious, hidden because of the inner con¡icts in which they are ensnared. Others are long forgotten, yet the behaviors based on them continue as part of everyday life. Some actions are taken without a thought, but on re¡ection the person can tell you why he or she behaved in this fashion. Despite the fact that behaviors may be based on con¡icting meanings, all normal actions, behaviors, and purposes are logical in the same sense that all normal speech is logical.4 Logic is used here in its most basic sense—as a system of related premises that lead to a conclusion, or a way of relating ideas so that they lead to another idea. Patient A says: “It is logical that I take my antibiotics because the doctor said it will help my infection.” Patient B says: “It is logical that I don’t take my antibiotics even though the doctor said they will help my infection. I want to treat my infection, but antibiotics hurt your immune system. Doctors don’t believe that antibiotics hurt the immune system, but they are

The Clinical Method and the Patient

[115]

wrong. I am more afraid of injuring my immune system than of not treating my infection.” The behavior, taking or not taking antibiotics, is the logical conclusion. (The example also makes the point that because something is logical does not mean that it is correct.) Inexplicable as some actions or behaviors may seem, when the meanings on which they are based are known, they make sense. From the clinician’s point of view, this implies that meanings can be discovered by ¤guring out what would make the behavior reasonable. Since future behaviors are likely to be similar to current behaviors, the meanings on which actions are based must be relatively stable. Put another way, if one wants to change the patient’s behaviors, the meanings on which they are based must be altered. The values component of meaning can also be read from actions. The patient who, in the face of an infection, doesn’t take antibiotics must feel strongly about the importance of protecting the immune system to risk the untreated infection or not believe the infection to be serious. The attempt to change the behavior must take into account not only the incorrect meaning, but also the importance to the patient of the values. These behaviors can be changed, but only from within the patient’s system of beliefs. Using meanings as information requires not only careful observation and careful avoidance of premature interpretation, as with all observation, but also the use of conscious reasoning about the logical structure that relates one meaning to another and makes sense of observed behaviors. Both speech and the actions of patients ¡ow by so fast, and their meanings so often seem obvious, that it takes systematically trained effort to reconstruct the belief system of the patient. This is especially important in chronic disease because it is this nexus of beliefs (given the disease and other circumstances) that determines the nature of the illness and the degree of suffering. On the other hand, the physician’s knowledge of the patient’s structure of belief is limited, even in the best of circumstances. What is necessary is solely the focused knowledge that makes optimal care of the patient possible. This is like knowing that the route of travel through a region does not imply grasping everything about the region. It is the same way that the patient knows the physician—as a physician, not as a person.* * A more complete description of how one comes to know who the patient is, and how their beliefs, values, and aesthetic dimension describe persons, is given in my book The Nature of Suffering, chapter 11.

[116]

doctoring

emotions Walk into the consulting room of a primary care physician and look for the box of facial tissues. If they are present, patients know that tears are acceptable; if not, it may be because the doctor is averse to displays of emotion. The same physicians who are uncomfortable in the presence of tears have learned to tolerate cries of pain; otherwise, they cannot do a necessarily painful procedure. Physicians who do not solve the problem of patients’ strong and painful affects cannot acquire the information about the person of the patient necessary for their work. Emotions, affects, passions, and sentiments are spontaneous responses to circumstances occurring before there is time for thought.5 They may be almost instantaneous, as with the rise and disappearance of a ¡ash of anger. They may become the person’s state of being—for example, an angry or happy person. They may break into action, as when someone acts angrily. Some people may be unaware of their own emotion even though it is obvious to an observer; in others it dances around the person’s surface, evident to all. Despite these variations, emotions are a re¡exive commentary on events that are often as true to situations and relationships (or more so) than words. We are often of two (or more) minds about things. Each “mind” has a set of beliefs coherent with the emotions based on them. For example, patients may want the doctor to know how brave they are and how much faith they have in the doctor (to keep the doctor on their side); on the other hand, they may be ¤lled with dread. It is the dread and its underlying beliefs that must be addressed by the physician, as well as acknowledging bravery and reassuring the patient that abandonment is unthinkable. Fear may radiate from someone who seemingly has nothing to fear. If there was nothing to fear, then fear would not exist. The doctor, sensing the fear, must ¤nd its source and, if possible, lay it to rest. Even if the fearful event cannot be dismissed, the loneliness of unexpressed fears can be managed by strengthening the relationship. None of this can be done if the emotion is not known. Patients’ emotions may be evident from their facial expression, posture, or gesture or may not be apparent except as they are sensed as the emotion of the doctor. Therein lies the problem of using emotion as information. When psychiatrists tell medical students to “use your feelings,” it is this phenomenon to which they refer. The students are often puzzled on two counts. They do not know how to

The Clinical Method and the Patient

[117]

be aware, moment by moment, of their own affect, and they do not know that emotions can be transmitted from one person to another. Primary care physicians may be similarly ignorant. Psychiatrists-intraining soon learn both facts because emotional states are their stock in trade. If they can master these states, so can primary care doctors. How can the doctor be sure that the emotion is what it seems and that it originated in the patient? Certainty is not possible. However, the perception of an emotion must meet the tests of any other piece of information. There are four criteria. First, it must have external logical consistency; it ¤ts the other facts about the patient and the circumstances known to the doctor. Next, it must have internal logical consistency; it doesn’t ¤t the doctor’s inner state. (“Why do I feel hopeless? I have no reason to feel hopeless. Maybe I’ve caught the patient’s feeling of hopelessness. I’d better ¤nd if, and why, the patient feels hopeless.”) Third, it must conform to experience; it is an affect seen in similar situations. There are ways in which it can be con¤rmed. Finally, the risks of any action based on it are appropriate to its certainty. There is a further problem about using information such as the transmitted feelings of the patient that feels so “soft.” The law of facts is that hard facts drive soft facts from sight, and soft facts drive softer facts into hiding. Doctors must gain con¤dence in their ability to observe accurately and employ personal information. Otherwise, emotions as data will always be defeated in a contest with numerical data. Patients’ meanings, values, and emotions, doctors’ emotions, and even the sensations under the physician’s examining hand can be easily overwhelmed. The following case makes the point: Abby Corcoran is a sixty-one-year-old secretary who called herself nervous. Three weeks ago, she had the ¤rst of three similar episodes that caused her to go to the emergency room of her local hospital. She became acutely short of breath and panicky; and the dyspnea lasted for several hours. It subsided while she was in the emergency room. Except for tachycardia and ¡orid varicosities of her lower extremities, the examination was unrevealing. Laboratory studies, electro cardiogram (EKG), and chest X-ray were negative. Blood gases were not tested. She remained shaken by the experience and frightened but otherwise well until four days later, when it happened again. In the emergency room the same scenario followed, and she was sent home. One week later it happened again and, ¤lled with

[118]

doctoring

dread, she went to the hospital. On this occasion, the chart also recorded negative ¤ndings and the impression that she was having panic attacks. While waiting to be discharged from the emergency room, she became severely dyspneic and cyanotic. She was effectively treated for what was shown to be a massive pulmonary embolus (with evidence of several previous emboli).

This was a preventable near death resulting from failure to give the patient’s dread and dyspnea more weight than the negative studies. The difference between the physicians’ lifesaving expertise in the face of the emergency and their inept response to her initial symptoms makes a striking statement about their training. Panic attacks in a sixty-two-year-old woman with no previous history and no precipitating episodes are, I believe, less common than pulmonary emboli. Among the other errors was the failure of the physicians who made this judgment to ask themselves the crucial question, “What if I’m wrong?” Errors such as this are prevented by training doctors how to observe, how to think about the personal information they gather, and how to give it as much weight as hard data. Every event has a history, no fact exists in isolation, nothing is observed outside of a setting, and there is always an observer doing something with the information. When the physician remains focused on the patient, on what the information means to and for the patient, on what must be done for the patient, then it is easier to integrate hard and soft facts, deal with con¡icting values, and maintain mastery over both patients’ and doctors’ feelings. Perhaps the best method of teaching these issues and strengthening the development of a doctor persona is in the discussion of the emotional burdens of patient care in groups such as those described by Michael Balint.6 Balint was less interested in the problem of integrating the information obtained through patients’ meanings and emotions, but these concerns may also be taught in groups. The group not only helps the trainee deal with his or her own emotions and those of the patient, but also validates the basic concept that it is necessary to be aware of these emotions as clinical data. Role playing or video or audio critiques may also be used in these groups. The training program must mandate attendance at these groups or they will fail, because the house staff does not think them as important as things emphasized by the specialty boards.

The Clinical Method and the Patient

[119]

aesthetics In addition to knowing people through their meanings and emotions, we also know them as aesthetic objects—wholes with parts in speci¤c relationships to each other and to the whole.7 Look at your hand. There is no other way to know that this speci¤c hand belongs on your wrist besides the visual appraisal of a good “¤t.” You could make the (almost) absolute determination that it was your hand through DNA analysis, but that wouldn’t tell you whether it belonged on the right or the left arm. Much of the knowledge physicians utilize in clinical medicine is of this type. There has been much debate about the nature of aesthetic thought,* but we will not get into trouble if we simply accept that there is an aesthetic intuition.8 In this act of thinking, objects and their relationships in a percept are grasped as a whole, and an idea is formed based on this information. This is why, when looking at persons, we say that they are old, young, strange, or beautiful—making almost instantaneous judgments that we are often hard pressed to spell out in detail. We hear the patient’s story of events and it doesn’t make sense—aesthetic sense. The patient’s distress at the dis¤gurement resulting from disease is at the aesthetic insult of the disease. Our unhappiness with some surgical procedures can refer to the unaesthetic result. But things that do not ¤t the life pattern of the person can be similarly unaesthetic, so it becomes necessary to develop an aesthetic sense of how patients with chronic diseases have lived their lives in order to better ¤t medical interventions into that pattern more effectively, or to treat patients so that they resume an existence more nearly like their normal lives. It is here that telling or writing the story of the patient’s illness or treatment * Aesthetics is a problematic word. Aesthetic knowledge is an often untaught grasp of order versus disorder, pattern versus chaos, that which in a sensory perception ¤ts and is appropriate versus that which is disjunctive and inappropriate. Aesthetics is essential in so many medical phenomena, from the appearance of wounds to the presentation of persons to the world, that it is curious that its importance is virtually unheralded in medicine. Suzanne Langer argues persuasively that intuition is the form of cognition of aesthetic phenomenon (Feeling and Form, New York: Charles Scribner’s Sons, 1953, and Mind: An Essay on Human Feeling, Volume I, Baltimore: Johns Hopkins University Press, 1967). In which case, clearly separating aesthetics and intuition is probably incorrect—the latter being how the former is known.

[120]

doctoring

can be very effective in understanding the aesthetic dimension of persons. (See the discussion of the description below.) Just as the medical world and the life world differ in their meanings, differences exist as well. Ultimately, aesthetics deals in relationships, and what it teaches is the relationships of parts to wholes, of wholes to groups of wholes, and of groups to communities. It keeps the physician focused not on this patient to the exclusion of everything else, but on this patient as a person who must always exist in relation to others and to a community. It is possible to see acute disease apart from these relationships, but it is not possible to understand chronically ill patients in isolation from others because that isolation would be part of their illness. Patients’ lives also have a rhythm that endures over the years. Sometimes it is a languor that impatient physicians characterize as slowness, or the staccato way in which the patient acts, or a digni¤ed, waltzlike motion in which everything is done. Working within the patient’s rhythm, where possible, strengthens the therapeutic bond. To ¤ght with it is a waste of energy. This is another aesthetic fact that should be appreciated. These two cases make the point. Janet Freeman has been my patient for thirty years. In that time she has had idiopathic pleuropericarditis, a hysterectomy, and a fractured humerus, as well as assorted upper respiratory infections, urinary tract infections, and other af¡ictions and injuries. The salient fact about her care is that she gets better very, very slowly. I think she should be out of bed and back at work; she can’t. Her cough should be gone; it isn’t. Her pain should be better; not yet. This happens not in one, or two, or even three instances, but in every illness, large or small. It is the rhythm of her life and it cannot be speeded up; it can only be respected—although it took me a while to accept that fact. Sophie Robertson married a New York artist and bore three children in the ¤rst three years of her marriage. Her husband was successful, and they lived in the intellectual artists’ world of New York City. Except for her children, Sophie was chronically discontented. She was a weaver and a banjo player, but much more the weaver. She developed carcinoma of the breast in her forties, eight years after her divorce. She tolerated the mastectomy and the chemotherapy, as she had so many other things by just doing what she had to do. Then she discovered Kentucky bluegrass banjo music, and it transformed her life. Years have passed since then, and she has remained the happy, engaged woman she became as soon as she discovered her rhythm,

The Clinical Method and the Patient

[121]

Kentucky bluegrass. She moved to Nashville, where she teaches bluegrass banjo and performs. There were no important new relationships or signi¤cant others, no change in her children or her health; she simply found the proper rhythm.

intuitions As discussed in the footnote below, there is another method of acquiring information from the environment, often called a feeling, as in “I have a feeling about . . .”* Here, in observing or experiencing the patient, ideas come to us; we form impressions or intuitions that we cannot spell out in detail. In fact, the act of spelling out in detail is called discursive thought, separate from the intuition, and it is the kind of thinking to which ordinary logic applies. These intuitions, or feelings, sometimes called nondiscursive representational thought, are not logical in the conventional sense, but they provide the raw material on which logic operates. There certainly is no antagonism—as though one is good and the other bad—between nondiscursive intuitions and discursive, logical thought. To the contrary. “Carpenter’s knees,” we say, looking at Joseph Heelon’s legs, where rough calluses are seen below both patellae. Then we move on to assumptions that logically follow about how carpenters work. Or, seeing how sick the patient looks (the nondiscursive intuition), we brush aside as not making sense (the discursive, logical follow-through) her contention that she has merely had the “¡u” for a month. The skill of Sherlock Holmes lay not only in the careful deductions he made, but also in the perceptive intuitions on which the logic was based. * Intuition is a dif¤cult word. On the one hand, it is used to signify a kind of knowing that is akin to instinct: the alleged possession of information without any demonstrable source. As such, it has been (justly) disparaged in medicine as a kind of magical knowing that is the opposite of the logical marshaling of information acquired by observation. Its more important interpretation is the process by which the mind directly grasps forms of things, recognizes the similarity of some percept to a known form, relationships, and meaning. As such, it is a basic form of thinking akin to thc meaning of feelings as impressions, described earlier. John Locke said of intuition: “Such kind of truths the mind perceives at the ¤rst sight of the ideas together, by bare intuition, without the intervention of any other idea; and this kind of knowledge is the clearest and most certain that human frailty is capable of” (An Essay Concerning Human Understanding, Book IV, chapter ii, Amherst, NY: Prometheus Books, 1995, p. 433).

[122]

doctoring

Think how wrong intuitions can be. You think a patient looks ¤ne, but it turns out to be clever makeup. A limp is a fake, grimaces of pain are not pain but grimaces, and so on. “You see that thing there?” asked my professor, pointing to a shadow on the chest-ray. “That’s cancer. How do I know? I just know, like you know I’m male without looking at my parts.” It was tuberculosis. In the era of objective scienti¤c data, this kind of evidence has been discredited as old-fashioned, ¡awed, and hopelessly subjective. Guilty on all three counts. However, how often is an aunt mistaken for an uncle, or a hand for a foot, or a basal cell carcinoma for a wart? What else can be counted on to make the crucial determination that “this patient is sick”? Reliable intuitions are distinguished from undependable ones by experience and training. Over the years you have been trained to distinguish instantly your left shoe from your right, just as physicians are trained to distinguish the sound of muscle tension crackles on the chest wall through a stethoscope from rales in the lung. Reliability is increased by learning how to verbalize the information (even if only to oneself) so that it can be evaluated and then organized into an anecdote. Anecdotes are knowledge, and carefully constructed anecdotes—accurate about the events and balanced in the reporting, so that the story speaks for itself—retold at the appropriate time are effective knowledge. Telling an anecdote is a crucial clinical skill. It is not as if, in this era of scienti¤c and technological medicine, nondiscursive intuitions have been banished from medicine. That is impossible; they are part of normal everyday thought. Instead, their skilled use has atrophied, leaving behind inexpert (and therefore frequently incorrect) intuitions to provide a type of clinical information available nowhere else, even when inadequate in the particular instance. The trained use of intuitions, impressions, or feelings about something is marked not only by an understanding of their importance, but also by the ability to ¤t them in with and examine them in the light of other facts. Skilled clinicians use the same method to make more subtle assumptions about patients’ thoughts, behaviors, and past history, and then quietly test the ideas against other facts before basing any action on them—even the next question. Adequate training will provide the basis for the development of these skills. Many of the examples I have given are based on visual observation, and that is probably the easiest area in which to train doctors. Teaching physicians how to be attentive observers is designed

The Clinical Method and the Patient

[123]

to complement the information from meanings and emotions that they have or will obtain. They need to learn to see who the patient is—to make the patient real as a person—at the same time as the details of the body and disease are observed: to provide an integrating center for all the information (and therapeutic action) that bears on the purposes of the patient. This is why presenting cases on rounds is so important. There are a number of components to attentive observation. First, attentive observers are always observing. Next is the importance of separating the observation from the interpretation previously noted. The third facet is the appreciation of the relationships between the percept, its parts, and the surrounding world. Fourth is attention to change. It is not an isolated event that is observed, but an event in an unfolding process. Fifth, an attentive observer does not leave a perceptual ¤eld prematurely. Finally, observations are in the service of goals that anchor them to other observations.

Description The ¤nal step in the process of knowing the patient, and an essential feature of the clinical method, is description. As discussed earlier in relation to ¤ndings on the physical examination, describing what has been observed is essential to the observation. The written histories of patients currently admitted to the hospital, which are meant to serve this purpose, generally bear little relationship to an actual person. They are often de¤cient as narratives of the unfolding process of the illness or chronic disease itself. They do not provide the model for the description of both the patient and the evolution of the illness that should be possible given the information from brute facts, meanings, emotions, aesthetics, and intuitions obtained by an appropriately trained primary care physician. Teaching this kind of descriptive narration is an essentially new ¤eld in medicine created only in the last twenty years with the advent of courses on literature in medicine.9 Understanding the interaction between the unfolding life of the patient and the evolution of the disease is best accomplished by narration. A narrative approach to questions raised by human illness ¤nds company in the recent work of scholars in other ¤elds of the humanities and the social sciences. There is now a rich bibliography and many trained faculty in the ¤eld. The characterization of a patient and the narration of the story of the

[124]

doctoring

illness are not only descriptive acts, but experimental as well. For example, the physician can see how changing an element in the description changes his or her idea of the patient, and consequently how subjective knowledge can be easily biased or protected from bias. Such learning further strengthens the separation of the doctor’s professional judgments of people from mere opinions easily voiced by nonprofessionals. Similarly, changing certain features of a narrative and then seeing what that does to possible futures or pasts allows physicians to understand how this patient’s story can be changed by different actions of the doctor or the patient or the context. The ultimate functions of description and narration are to make the subjective objective and open to examination. This case is an example. Magda Ofkowski is a very attractive Polish-born, married, ¤ftytwo-year-old woman who has been a violinist with a major orchestra for twenty-seven years. As on one previous occasion, she has had to stop playing because of severe pain in her neck radiating to her right arm. She found a new physical therapist whom she likes, but after the second treatment she developed severe pain in the right buttock radiating to the inner thigh and down to the foot. On examination she was obviously in pain, and straight-leg raising was restricted, but there were no other neurological ¤ndings or muscle weakness. The pain prevented her from making the anticipated ¤rst visit to a psychotherapist. With adequate pain medication and a benzodiazapine, she was sent for bed rest. The pain in the leg got worse, as did the neck pain. A magnetic resonance imaging scan of the lumbosacral spine showed a signi¤cant disc bulge at L3–4 with impingement on the foramina. A few days later, while she was gargling with her head thrown back, she developed very severe pain all the way down her right side into the foot. An MRI scan of the neck showed degenerative disc disease and a few areas of minor disc bulges. Questioning revealed that she has not had sustained bed rest for even a half-day. She had had a successful mitral commissurotomy for rheumatic mitral stenosis at the age of thirty-one. A left mastectomy was done at age forty-two. She had left home at age twenty-one following one of her mother’s routine beatings to make her practice the violin. She got married immediately, but it ended within a few years. Her present marriage of twenty years, childless, has been excellent. This episode of neck pain preceded not only psychotherapy, but also the start of rehearsals by a new chamber music group of which she had

The Clinical Method and the Patient

[125]

become a part. The previous occasion of disabling neck pain ended an attempt to create a career as a soloist.

One can see the temptation to turn this case into an example of either psychogenic back pain or intervertebral disc disease requiring intervention. Change the size and extent of the disc intrusion, and an impatient generalist could have her in a surgeon’s of¤ce. Change the era in which it took place, and she would soon be in an operating room and perhaps on the way to having an overoperated, failed back. Forget the history and the obvious pain (which happens not infrequently), and she would soon be dismissed and wandering from one alternative therapy to another. It would not take much to transform her into one stereotype or another. Disastrous permanent disability waits only for error. The case cries for more information about her and her physical status, all of it available. It also makes clear how little changes in the story of the person or the body would change the case and, therefore, how important are the stories of both the person and the body. Recently, the new editor of the Annals of Internal Medicine, Frank Davidoff, wrote about his ideas as he took over the journal. He spoke about the importance of science to internal medicine: “Science is cognitive, involving accurate observation and clear description, hypothesis generation, data gathering and interpretation, and the creation of theory. But science is also a state of mind: skeptical, open, balanced, respectful of evidence, thorough, always on the alert for bias.”10 Everything here except perhaps the creation of theory applies to the clinical method. In your mind’s eye, consider the past and you will see a tradition that goes back to Eugene Stead and William Osler, to John Locke and Thomas Sydenham, and then to Hippocrates. Why, over the millennia, have they seen the clinician’s task in such similar terms? Because they all knew that the fundamental truth of medicine resides in knowing the patient. Over the ages the meaning of the word patient has expanded its dimensions, so that it now includes the whole person. Physicians themselves are the means, the relationship with the patient the vehicle, and the clinical method the tool by which this knowledge is gained. We have only to know how to teach this.

[126]

doctoring










6 Where Should Primary Care Be Taught—and by Whom?








T

he experience of this century has made it abundantly clear that medicine is best taught while doing it. The origiconcept on which this teaching model is based is that doctors should learn about diseases from experience in caring for patients and should see ¤rsthand the contributions of science and technology. In this era we have added the necessity of knowing the patient—the sick (or well) person. This new requirement demands a kind of information that can only be obtained by a return to the fundamental clinical method, attentive clinical observation in which physicians are their own instruments. It becomes necessary, therefore, to teach physicians not only what they need to know, but also what they need to be.

The Importance of Experience Where primary care should be taught will become evident, I believe, by looking more closely at both requirements together; clinical knowledge and clinical presence. These are both learned through experience—but experience is a complicated word. Experience can mean the method by which one con¤rms an opinion or an under[126]

Where Should Primary Care Be Taught—and by Whom?

[127]

standing of something by recourse to the thing itself. This is what is meant by saying that someone has seen for himself. This is often what is currently meant by clinical experience. Physicians have an understanding of a disease or the use of a technology, and by experience with appropriate patients or the performance of procedures, they see whether their beliefs are correct or not. To go further with this sense of experience, it is also what is meant by learning from experience when expertise is gained by doing something repeatedly. Thus, we want trainees to do a number of spinal taps or thoracenteses so that they gain experience (pro¤ciency) in the act. Endless examples could be given to reinforce the utility of this kind of experience in everything from listening to the heart to telling patients bad news. Experience of this type is a kind of experimental method, trial and error, learning the pitfalls, and so forth. There is no substitute for this kind of experience when training physicians. There is a second meaning of experience that I believe must also be part of the training of primary care physicians. In this, one undergoes an experience in the sense of letting the matter itself demonstrate itself in its truth.1 Having experience with sick patients in this sense means allowing their experience of their illness to become part of the experience of the physician. It is a kind of powerful empathy at its best. For example, I was once asked by a patient who was almost paralyzed by her fears to walk across the street with her. She left her home only with dif¤culty; shopping in crowds was impossible; crossing the street was an agony. As we started across the street with her ¤ngers digging into my arm, I began to experience her terror. My angle of vision narrowed so that I could only see straight ahead—like tunnel vision. Everything seemed to appear suddenly and come right at us from nowhere—people, cars, even a dog. It was awful. I understood what she had been saying in a new way. In caring for patients with chronic disease, this process involves experiencing the pace of the disease or its impact on the patient from the patient’s point of view. This means that as the physician receives information about the patients’ meanings, emotions, and intuitions; this information forms part of the doctor’s experience of the patient’s experience. Patients’ experiences of being sick vary enormously, not only because of individual differences, but also because the manifestations and severity of disease and illness cover an immense range. One patient, bleeding from extensive sudden trauma, may be terri¤ed. Another, with late-stage

[128]

doctoring

terminal illness, may be totally withdrawn from the everyday world into the domain of illness. Still another is in a state of constant anger at the indignities of helplessness. Other patients may be no different than when healthy, except that they have a disease that requires a doctor’s attention. The doctor’s experience should encompass all of these experiences of patienthood. In this kind of experience, the distinction between person and body, between patient and illness, is blurred or lost, allowing the doctor to return, in the best tradition of the clinical method, to the thing itself, the phenomenon of human illness. We expect doctors to learn from experience in both of these senses. The experimental or con¤rmatory type of experience that is necessary to reinforce existing knowledge and sharpen skills is an active behavior. The other, a richer, deeper kind of empathic experience, is gained in two ways: in the passive sense of being acted on by the experience of the patients and in the active sense of holding oneself within the experience even when it is unpleasant, contradictory, surprising, or challenging. There is a third meaning of experience that bears on the training of primary care physicians. The experience of being a physician in the course of acquiring the other two kinds of experience becomes important if doctors are to be their own instruments. Physicians’ emotions, concerns, desires, fears, and sources of pain, previously considered a private matter, now become important to their teachers; persons as physicians are being taught, not merely persons who have the knowledge of physicians. Whatever diminishes or enhances the instrument is a concern of the training program. Knowing that we want all three kinds of experience for our trainees tells us where they must learn primary care. Much of the change in education for primary care suggested in the literature has focused on the place of training. It seems clear that the traditional method of training physicians solely on the wards of teaching hospitals is inadequate. Many consider it essential that primary care training should take place in an ambulatory setting such as an outpatient clinic, of¤ce practice, or in the community.2 Unquestionably, the problems presented by patients outside of the hospital are different than those of inpatients, and different but complementary skills are necessary for their care. Furthermore, many patients who previously required a hospital for their care or surgery are now commonly treated outside the hospital. When the

Where Should Primary Care Be Taught—and by Whom?

[129]

community itself is the locus of intervention, or when the physician has community and public health responsibilities, experience must be gained in acting in such roles. There can be no transformation in the direction of medicine without a concurrent change in the method of training doctors so that their education matches their actual tasks in patients’ care. This goes far beyond merely changing their place of training.3 The changes are designed to extend experience in the ¤rst sense of the word—exposing trainees to the different manifestations of chronic diseases, the various conditions of ambulatory patients, and the tasks of widened community responsibility. In one sense, the differences between inpatient and outpatient training are arti¤cial because patients can be learned from in every site. Trainees must also know what being a physician means to the chronically ill: the importance of the role to their patients; of their constancy to the sick, for whom virtually nothing else is constant; of being trustworthy to those for whom trust in another is problematic; their command of events as agents for patients otherwise powerless; and the comparative certainty of their predictions for the ill who live in awful uncertainty. As they learn to be these things, trainees experience the patient’s world of illness and their own presence as physicians. They experience the doctor-patient relationship, and for this to be maximally instructive, it must be actively taught by the faculty or by mentors. In matters of the relationship between them and the patient and the patient’s experience of illness, ambulatory care alone is not suf¤cient because trainees must experience the experience of their patients throughout the entire spectrum of human illness, including the care of the dying.4 The very sick are generally in the hospital, so there is no substitute for hospital experience. There is an advantage to training on the wards of the big-city general hospitals—for example, Bellevue in New York, Charity in New Orleans, and Cook County in Chicago. Doctors learn from patients who are terribly sick in the manner experienced only by the very poor, disadvantaged, and down and out. Not only are their diseases awful, but the patients often have never had the experience of anyone who cared; thus their doctors become not only their only hope for survival, but also their personal champions (especially given the distressing state of most such hospitals). In such settings, doctors learn to be (or appear to be) calm and assured in dealing even with the most severely ill and

[130]

doctoring

panicky patients. This kind of training—which teaches one of the classical roles of the physician—seems to be disappearing with the rise of tertiary hospitals and the emphasis on advanced technology, although it may return as a consequence of severe budgetary restraint in American cities. If this training is unavailable elsewhere, service in refugee camps overseas or in desperately poor countries like Bangladesh would probably serve the same purpose. At the very least, this should be arranged for the most promising trainees.5 Recently, a new specialty, the “hospitalist,” has been proposed. These physicians would provide inpatient care and be responsible for teaching students and house of¤cers in the hospital. Primary care physicians, including general internists, would not follow their patients when they are admitted to the hospital; this would fall to the “hospitalist.” The argument for the development of the specialty is based on the increased complexity of hospital medicine and the evidence of increased survival of certain groups of sick patients when taken care of by specialists. The dif¤culty in caring for hospitalized patients argues for better training of primary care physicians and better integration of the work of specialists and generalists rather than the creation of a new specialty. Survival, while simple to measure, is a poor indication of the quality of care. If you doubt this, walk through any nursing home. It calls to mind A. N. Whitehead’s comment about thinking about evolution primarily in terms of survival value. By this criterion, rocks win, hands down. Measuring the quality of medical care predominately by heartbeats and body heat is one of the reasons modern medicine got into its current dif¤culties—focused more on diseased organs and technology than on the goals of sick persons. Patients do not simply want to survive, they want to survive in order to live a life in which they can recognize themselves and in which their values are preserved. There are reasons for broadening rather than narrowing the expertise employed in the care of the sickest patients. It is too common for patients with, for example, pulmonary failure to receive the best care possible for their lungs while their muscles waste away, and their other functions wither for lack of use and attention. If they survive to discharge it takes many months, if ever, for them to return to their previous lives. They need teams consisting of physical therapists, an occupational therapist, a psychiatrist, and perhaps others, specially trained and dedicated to this—not further treatment by someone knowing only the increasingly arti¤cial world of the hospital. It is

Where Should Primary Care Be Taught—and by Whom?

[131]

understandable that the call for “hospitalists” might arise in response to demands for greater ef¤ciency and lower cost, but will it make for better medicine, now or in the future? It is bad enough to be a lemming, but is it necessary to help ¤nd a faster route to the sea? At the other extreme, but just as important, is the fact that most illness is cared for in the home. Physicians must be able to provide medical care in this setting. It is often stated that house calls are inef¤cient and uneconomical, but these are not necessary truths. One learns things about a patient by a home visit that can be discovered in no other manner. However, care in the of¤ce or clinic should make up the bulk of doctor–patient interactions, and the instruction needed here is speci¤c to the site. Training generalists, then, should take place everywhere that patients are found. Family physicians, general pediatricians, and general internists require training that suits them not only for those roles and skills that are common to all generalists, but also for the tasks unique to each. It is not merely skills and technical knowledge that are being taught, but also a structure of values. There is no such thing as not teaching values; they are always present. One should give as much attention to the values being transmitted as to everything else in the educational setting.6 Not enough attention has been paid to the fact that the upheaval and largely unplanned reorganization of medical care currently taking place has largely destroyed the way in which the values of medicine have been transmitted to trainees during this century. It has been during their hospital training that doctors learned, for example, that the patient always comes ¤rst, that concern for the patient’s well-being takes precedence over the doctor’s personal values such as money, sleep, status, or even family. The manner in which these things were taught generally drew on the history of the hospital and of medicine in general. Other values, ranging from the central importance of the knowledge of medicine to the sanctity of normal tissue were also imparted during hospital training. The importance of these ideas were sometimes transmitted by teachers directly and on other occasions in anecdotes that act as parables. Committees of the Hospital Medical Board—for example, the tissue committee, medical records committee, or the operating room committee—served an oversight function to maintain standards. This has all changed drastically in the last few years. Patients generally remain in the hospital such a short time that students and residents have little opportunity to form the bonds on which training about

[132]

doctoring

responsibility can be based. The technology has moved outside the hospital to such a degree that its use cannot be effectively monitored. In the same manner, outpatient surgery offers much less opportunity for the teaching of the values that constrain its inappropriate use. The task of manifestly teaching values is rarely given much thought in medicine (and elsewhere). It is simply part of the background. Thus, when teachings such as the traditional value structure of medicine are disrupted, as it has been recently, little thought may be given to its loss until the profession becomes unhappy with the values of its new members. Unfortunately, by the time that happens and solutions are found, a decade can pass. It is not all clear what is to take the place of the hospital in this task. Ambulatory care settings are not as effective as hospitals because of the dif¤culty of establishing the group identity that fosters new values. There is a further issue of values. Medical education—graduate and postgraduate—has valued the independent, individualistic, competitive, rewarded oneupmanship, honored esoteric knowledge above the virtues of patient care, attacked errors out of proportion to the error itself, and promoted the politically savvy above the medically sound. Doing this creates defensive physicians who constantly fear criticism and go out of their way to protect themselves. The personal price young physicians pay for these values is unfortunate, as is the interference of defensiveness in the relationships with patients. Worse, this model of physician is simply out of date. Competitive individualism does not function well in an era when physicians must work closely with each other and other professionals in order to use complex diagnostic and therapeutic tools. Competitively pretending to have a comprehensive knowledge is unnecessary or worse at a time when a few minutes at a computer keyboard will bring to the screen the resources of the National Library of Medicine. We must give active thought to what kind of physicians we want and take care to ensure that these are the values transmitted by our training programs. Long after medical knowledge has become outdated, the values gained during training endure. The goals of postgraduate training of generalists cannot be adequately met only by experience with patients, no matter what the setting. Systematic instruction in classrooms, seminars, and organized groups is required to meet the goals of generalism discussed in this book. I am well aware that this idea is both outré and repugnant to most medical educators, but I believe we must reexamine this

Where Should Primary Care Be Taught—and by Whom?

[133]

question carefully. The present method of training, as noted previously, was developed in the service of an idea. It is a new idea that the clinical method and the skills of doctoring—training the physician as instrument—must be explicitly taught to meet the goals of modern primary care. New methods of teaching must be developed to meet the idea. Just as the subspecialist goes to the laboratory to learn science, so the generalist must return to the classroom to learn the fundamentals of doctoring. Ophthalmologists have been taught in classrooms for years because the basic sciences of the eye are not taught in medical school. Orthopedists are increasingly being taught the use of new technologies in this fashion. Physicians going into public health return to the classroom, as do modern physician-administrators. Primary care physicians should learn their basics in didactic classroom exercises, groups, and seminars. It is ¤ne, for example, to use critiques of video or audiotapes of residents’ interactions with patients to reinforce what has been taught about the doctor–patient relationship or about communication with patients. The fundamentals of these essential elements of doctoring, however, require didactic instruction based on a well-developed curriculum covering the clinical method. Just as prescribing medication in practice is no longer based on trial and error, these therapeutic (and diagnostic) skills cannot afford to depend on native talent or inconstant instruction. The clinical method is the basis of primary care medicine; it is too important to be left to the chance of a talented preceptor. Program directors may wonder when in their residents’ busy day such classroom instruction would be possible. One hour a day, every weekday, will meet the teaching requirement if it is clear, from day one, that this is the way this service teaches, that it is non-negotiable, essential to success in the program, interesting, and makes patient care more effective and easier. Family physicians, general pediatricians, and internists differ in their relationship to hospital training. General internists, I believe, should spend their ¤rst year of postgraduate training on the hospital ¡oors, as is done at present. In their second year, they should return to the classroom, as well as doing ambulatory (and, perhaps, home) care. In their third postgraduate year, general internists should again return to both the acute medicine of hospitals and the ambulatory setting. One advantage of this kind of training in primary care is that on their return to the ¡oors, the generalist and the subspecialty

[134]

doctoring

residents could teach each other their unique knowledge bases as they care for patients side by side. There is more than enough advanced material to support a fourth postgraduate year or academic fellowships in primary care, but this will not be necessary when medical schools ful¤ll their obligations in training for primary care medicine. Family physicians could divide their time between hospital and ambulatory care throughout their training, as is the custom in most family medicine programs at this time. There is a further problem to be solved. We like to believe that doctors gather evidence and reason from the facts and their knowledge of medical science to reach a decision that is best for a particular patient. To the contrary. The modes of thought in which physicians are trained and that govern much of their behavior are based on virtually automatic skills and rules that are preprogrammed and learned early in their training.7 These skills and rules concern disease and technology, not persons. Doctors have not been trained to acquire similar rules and skills about persons. Until they are, their best intentions for a sick person may be overridden by an automatic rule. Postgraduate instruction must teach doctors to be their own instrument with such con¤dence in their command of the clinical method and the discipline of their subjectivity that their ¤ndings can compete with possibly con¡icting images on ¤lms or the numbers on a printout or with the siren call of a sweet technology. New rules and skills must be acquired that are related to persons. Hands-on postgraduate training is no longer adequate to these tasks. I have been practicing medicine for years, and I have been writing about and studying this problem for more than twenty years. I still have dif¤culty integrating objective data, moral imperatives, aesthetic information, that which is value laden, and that which is affective. I do not think the dif¤culty is unique to me. It is not merely principles or attitudes that must be taught, but also dif¤cult skills. Learning to write narratives about patients and narrative histories of illness should also be a classroom activity that can move back and forth between patient care areas and the seminar. Teaching narrative writing should be done by experts from the humanities because this is a special skill in written English. A large literature exists on the subject, and faculty are becoming increasingly available. These aspects of the clinical method should probably be taught by a multidisciplinary team that also includes behavioral scientists.

Where Should Primary Care Be Taught—and by Whom?

[135]

It is crucial to remember, however, that it is medicine that is being taught, not behavioral science or English composition. Earlier, I discussed the use of groups modeled after Balint but brought up-to-date to teach physicians how to utilize the information from their patients’ and their own emotions. Small-group teaching is, in general, an excellent forum for teaching the components of the clinical method.8 The master class method is also very useful in teaching physicians how to integrate the information they have, whatever its source, into judgments about what is best for a patient. In master classes, all the participants are assumed to be knowledgeable; what the student hopes to learn is clinical expertise.9 In medicine, master classes consist in discussing a patient who poses a problem to the presenting doctor from the most personal to the most technical. The other participants, guided by the more experienced clinician (the “master”), discuss the issues raised by the case, their approaches to the problem, alternative constructions of the question, and so forth. These sessions are not tests of factual knowledge, which all participants are presumed to know, or debates about issues that computer search programs or other sources can quickly settle. The master class is a give-and-take setting where points are not scored. Students are asked to leave their defensiveness and competitiveness outside the room (this takes some doing, but it is possible), as no one is assumed to know the answer; judgment is being learned. I have held these classes with accomplished physicians, and it is both a wonderful teaching method and a lot of fun. One point of the educational process for generalists must be to reduce the competitiveness that marks medical education from premed college years through medical school. Emphasis on competition promotes a feeling of inadequacy no matter what degree of excellence is achieved, and inadequacy promotes defensiveness. It is dif¤cult for doctors to build therapeutic alliances with patients, be aware of their patients’ and their own emotions and meanings, and be optimal instruments if they are constantly trying to defend themselves against criticism. The almost pathological competitiveness of medical students in this era is a barrier to the development of team approaches to the care of patients. I am envisioning general internists and pediatricians acting as consultants to family physicians in the care of patients with dif¤cult multisystemic illnesses and family physicians integrating the care of these patients back into the family and community. This will require a change in the relationship between physicians that will not come about if it must depend only on friendship

[136]

doctoring

between individual physicians. Other specialties can continue to be competitive if they think it is a useful thing, but it defeats good training in primary care. In some areas of the United States where large multispecialty groups have been the mode of practice for many years, the relationships between physicians and different specialties are already well established. In other parts of the country, they are yet to be organized. I have written only about postgraduate training. Why have I not discussed changes in the training of medical students? First and foremost, so many con¡icting interests are involved that changing the curricula of medical schools is very dif¤cult. Next, the task of medical schools at this time is to train students in the basic sciences and their application to the care of patients. They do this extremely well, and while changes may be desirable in many schools, the basic mission is essential. Medical students leave to go into many different specialties for which medical science is foundational. If necessary, all primary care can be taught in the postgraduate years; not all medical students need its special understanding and skills. Practically speaking, most primary care faculty have more control over postgraduate training programs than they have general in¡uence in their schools. If the medical schools immediately instituted the training that I am suggesting, there would not be suf¤cient faculty and the effort would fail. It is better to wait and change the medical schools later. Does this mean that I think that the fundamental ideas of primary care included in this book or a primary care track should not be part of medical education from the beginning? Not at all. In fact, I believe that it is the responsibility of medical schools to start teaching the fundamentals of the clinical method—for example, observational skills, the basics of conversation necessary to effective communication, and the construction of the narrative history—in the ¤rst year. These are hard skills and would pro¤tably replace all the “introduction to medicine” courses and “introduction to a patient” exercises presently employed to keep the students interested. The concentration on the clinical method and all of its components would provide a focus on doctoring throughout the four years appropriate to physicians with diverse career goals. It would also reduce the need to teach residents what they should have learned in medical school. Ultimately, that is what will happen, but for the present I am patient . . . and realistic. One further consideration in choosing where to begin the new training is that postgraduate programs can create small-group

Where Should Primary Care Be Taught—and by Whom?

[137]

psychology where the belief structure of primary care medicine can be reinforced by pride in belonging and where, as a consequence, the participants will learn more, and more rapidly, than they would in medical school. Psychiatry has been successful in this regard, as have other narrow specialties. Creating a group identity is especially important when the ideas of the specialty are new.

Who Will Be the Teachers? The kinds of knowledge and skills included in the role of the primary care physician described here are different from those usually found among general internists, family practitioners, and pediatricians. Who will be their teachers? In the past there were role models, physicians devoted to clinical medicine who cared for patients on the ¡oors of medical centers. Such doctors are mostly gone from the teaching hospital, over age, or forced away by the full-time/part-time faculty split aggravated by current ¤nancial competition between the two groups. Even the best clinicians may have dif¤culty teaching knowledge and skills they have learned by experience but that are not valued in the world of scienti¤c medicine. In family medicine programs, the faculty is often practicing family physicians teaching what they do and what they are committed to. Despite this area where role models exist, it is unlikely that a large group of teachers can be found in a short time or that whole classes of medical students can soon be taught the knowledge bases and personal skills of experienced clinicians. The model of teaching endoscopic abdominal surgery will not serve here. Surgeons learning how to operate endoscopically already have the basic surgical knowledge, ability to use instruments, and diagnostic and therapeutic purposes into which the new technology ¤ts. In primary care, students and house of¤cers, if they are interested at all, have an unlettered moral imperative to treat “the whole patient,” surrounded by strong countervailing forces. They, and perhaps much of their faculty, are unaware of how the present teaching of science and technology, essential as it may be, distances them from patients and nulli¤es their best intentions, as described in Chapter 2. I believe we must accept the realities that despite some excellent programs there are no adequately developed complete curricula, nor

[138]

doctoring

is there a suf¤cient cadre of teachers for training the type of primary care physician described here. At the same time, my direct experience suggests that throughout the United States there are interested faculty and clinicians, as well as well-developed programs in several subjects that must be taught. Signi¤cant resources are available for program development. There is an important role that medical students and physicians should play in choosing and de¤ning their own agendas and ¤nding teachers. They must be encouraged to understand that throughout their lives they should be ¤nding teachers in different aspects of medicine. Sometimes it is as simple as ¤nding an obstetrician/gynecologist colleague to teach them how to do a thorough pelvic examination. At other times, they might ¤nd older clinicians with whom they can discuss cases. If you ¤nd your students or residents doing this, you will know that you have been successful in teaching them to overcome the defensiveness that plagues most physicians. They will have learned to accept their own inadequacies and the need for lifetime learning. The opposite side of this is to encourage faculty to be mentors, not just teachers. Mentors and faculty can be drawn from all three generalist ¤elds—family practice, internal medicine, and pediatrics.10 A diffusion model best ¤ts the present situation. First, the teachers of the teachers—directors of primary care programs and occupants of chairs in medicine, pediatrics, and family practice— must become aware of the need for advanced program development. Just a few of them; a few leaders will be suf¤cient. Through their leadership, they must allow young faculty to be taught and to begin to teach other teachers. They must give their interested faculty the time to read and study, to travel and meet other likeminded teachers. Courses have to be developed, tried out, evaluated, revamped, and brought back on line. The house of¤cers taught in these departments must go on to be the teachers in other programs.11 Communication between faculty, trainees, and program administrators should be actively encouraged.12 Since I started writing this book, I have become aware of an unhealthy competition between some general internal medicine and family practice faculty. This suggests, perhaps too optimistically, that certain programs have already achieved whatever goal is being spoken about—a self-defeating attitude. The race to discover (the equivalent of) the double helix is never won by claiming achievements that do not exist. Healthy competition consists in trying

Where Should Primary Care Be Taught—and by Whom?

[139]

harder to reach an impossible goal: the perfectly trained primary care physician. Some departments of general internal medicine believe that the medical interview is the curriculum; and others accord this honor to medical decision making or clinical epidemiology. These are necessary subjects, but they are not suf¤ciently broad or inclusive. There is a body of knowledge that graduating medical students do not command. Postgraduate curricula (and the eventual primary care track in medical school) must cover the use of technology, therapeutics, knowledge of persons, knowledge of function and process, preventive medicine, and personal skills (observation, communication, clinical judgment, clinical epidemiology, the doctor–patient relationship, and understanding the role of the attending physician) discussed in the next chapter. The best methods of teaching these subjects are not yet known. Unfortunately, we have left the bedside in the sense that Sir William Osler intended in order to gather around the viewbox where the X-rays and other images are displayed. Technology prevents students and teachers alike from learning directly from the patient. On the other hand, without adequate skills it is futile to try to learn from patients. (Ludwig Eichna has discussed the dangers and the self-indulgence of letting students work with patients before they know enough to make a contribution to care.13) Accepting a diffusion model means learning patience. It will be years before there are enough teachers to go around and suf¤cient, fully developed course materials. This period is also necessary for the development of a common language, uni¤ed purposes, and feedback from early teaching experiments. There is always the danger that new programs will speak about, for example, communication skills, the importance of psychosocial elements, and the place of the community but not teach them adequately for lack of adequate ideas or teachers. We live in a world of hype and all of us know the importance of saying the correct thing in a request for funds. We must not, however, fool ourselves. It will be sad if the inadequacies of the generalists trained in inadequately conceived programs are taken as evidence of the failure of the underlying concept. This is a hazard to which we must be alert. The model of growth for the advanced generalism advocated here should be the spread of scienti¤c medicine in the United States after the turn of this century. William Osler and his colleagues at Johns

[140]

doctoring

Hopkins taught a way, unique in this country at the time, of seeing disease as a biological entity that was as subject to scienti¤c inquiry as any other object of nature. They were teaching an ideal of scienti¤c medicine well before medical science was suf¤ciently advanced to achieve what they promised for it. What we would now call clinical laboratories were established just off each ward, but think how little these laboratories could do given our current expectations. Little matter. They represented a fresh concept of medicine. The new breed of physicians were excited by what they were doing and were, by all accounts, exciting teachers; they improved the care of patients, and the physicians of the time knew none better. A wonderful book, Doctor Dock, gives an interesting picture of how these innovators at the beginning of this century taught medicine to students.14 The success of Oslerian medicine, which probably reached its peak in the 1970s, was won against entrenched political and economic interests in the medical colleges and hospitals of its time. There is a somewhat different array of forces now, but otherwise the problem is essentially the same. Primary care will inevitably grow and ¡ourish, but its progress will be hastened if money for its support is available. (Johns Hopkins supplied the money to support William Osler’s idea.) Nothing so focuses academic interest as program funds. The current restructuring of the health care delivery system, based on primary care in the absence of educational support, will slow the diffusion process because of the absence of support for research and training in generalist medicine. On the other hand, if the corporate managed care of today leads to multispecialty group practices tomorrow under the leadership of physicians, and with capitated payment for patient care, we will have a much better environment for the care of patients and for the support of advanced ideas of primary care. The kind of training discussed here is fundamentally different from present medical education and requires speci¤c attention. The Robert Wood Johnson Foundation recently devoted program attention to facilitating the training of generalists. They have required successful applicants (fourteen schools are now participating) to show success by the increased percentage of their students who become generalists— hoping to attain 50 percent of the class. The participating medical schools are focusing on the applicants for generalist tracks, their characteristics, and successful recruitment techniques. They are trying to provide greater exposure to generalist role models, as well

Where Should Primary Care Be Taught—and by Whom?

[141]

as modifying the content of the ¤rst two years in ways that might promote generalism. The program asks the schools to decentralize the training in the clinical years away from the specialist hospital settings. Committed teachers are expected to emerge from these experiments.

Who Will Teach Before There Are Teachers? New purposes will be pursued with old knowledge and skills. It cannot be otherwise. Initially, primary care physicians will not be optimally trained, but if new programs emerge, change will be in the air. The literature will re¡ect these developments, and physicians will begin to learn new skills. Continuing education courses covering the same materials will also proliferate. The forces behind the changing face of health care delivery have a crucial role and responsibility in this evolution. If physicians are treated as the solution rather than the problem, rising morale will help spread new ideas. If primary care is forced to remain primarily an organizational concept, development will be slowed. Ultimately, however, primary care medicine will not begin to ¡ourish for a decade or more, and there will be insuf¤cient numbers of new training programs. In a nurturing environment of academic, political, and social interest, new teachers will spring up. In a medical world of struggle between the old and the new—which is the most likely scenario—fundamental interest in the care of the patient and the success of new ideas in promoting better patient outcomes will gradually force change, cultivate teachers, and bring increasing numbers of students and physicians into the ranks of modern primary care, as in the Oslerian model.

Student’s Choice of Career Path There is an emerging interest in curriculum development aimed at providing primary care tracks for interested students.15 These programs have as one objective increasing the interest of students in a career in primary care. It is not clear, however, why a student should enter such a program. Students de¤ne their career interests based, for example, on their own interests (some students were anointed in their cradle to be hand surgeons) and in recent years

[142]

doctoring

many more entering students have expressed an interest in generalism. Others join their ultimate ¤eld because of the basic orientation of their medical schools; the so-called elite schools concentrate on the training of subspecialists and research scientists, while many state schools expect their graduates to embrace some form of primary care and enter practice in their own geographic area. Money plays a role in career choice among recent graduates as they perceive that some specialties promise a higher income than others. Attempts to change this situation have concentrated on raising the income of generalists relative to those of other specialists. As a result, there has been a slight increase in the income of generalists. In the last few years the number of applicants to primary care programs has risen sharply, suggesting that students are aware of the shift toward primary care in the United States and the problematic nature of many of the subspecialties formerly so attractive. Recent proposals to shift postgraduate training positions toward more generalists and fewer specialists would also create powerful incentives. The so-called 50/50–110 proposal would greatly reduce the number of specialist postgraduate residency slots, resulting in parity between generalists and specialists-in-training. Additionally, in this proposal, the total number of residency positions would be decreased.16 Directors of primary care or generalist training programs must ask themselves if they have provided adequate grounds for entering the ¤eld. If they have not, if there is no fundamental excitement in their view of primary care—something the student can become passionate about—the next few decades of medicine in the United States will be discouraging. If the students are not excited, how interested will the faculty be? Conversely, if the faculty are not excited about the challenges of their chosen ¤eld of scholarship, why should the students be? The concept of primary care or generalism that these students and perhaps too many of their faculty have is that generalism simply is not specialty medicine. This is not the perspective on primary care that is the basis of this book. Instead, what is being described is what John Stoeckle of Harvard’s primary care program conceives to be more exciting, more intellectually demanding, more central to the medical challenges of our time. If teachers of generalism do not yet feel this way about their ¤elds, they have yet to embrace the program. For teachers, the time has come to look around, see what others are doing, visit their programs, try things. And read not just what

Where Should Primary Care Be Taught—and by Whom?

[143]

is written about primary care (though that, too), but also the history of medicine to see how past changes in basic ideas have come about. They should also read biographies of earlier physicians to understand the ethos of the medicine of previous times. Particularly interesting and relevant to the changing medicine of today is Sharon Kaufman’s The Healer’s Tale. The author describes the lives (mostly in their own words) of seven well-known physicians whose careers started in the 1920s and 1930s and developed in the era of scienti¤c medicine, bridging two medical worlds.17 The primary care medicine now coming into being is the medicine of the future. It is not a depleted medicine of the past, a defeat for the growth of science and scienti¤c medicine. Rather, it is the inevitable continued growth of the fundamental idea of medicine that has been our heritage for two and a half millennia—physicians caring for the sick.

[144]

doctoring










7 What Should Be Taught?








Questions and Answers

I

n medicine, the teaching of skills is usually based on both concepts and basic understanding. Teaching physical diagnosis, for example, depends on the student’s knowledge of anatomy, as well as concepts of disease into which the physical ¤ndings ¤t. So it is the skills of the generalist with which this chapter is concerned. In many ways basic understandings are different from the student’s or resident’s previous experience of medicine, and they are certainly different in emphasis and intensity from current subspecialty medicine. This means that trainees will be asked both to learn new skills and to change their mindset. The faculty will also probably be trained in a different point of view, and so will have the problem of both teaching themselves as well as their trainees. Transformation of the clinical method is not easy to teach or maintain. Technologies, as discussed in Chapter 2, forever pull physicians toward themselves or the things with which they are concerned. The computerized record, decision analysis, the Internet, and online conferencing for patients and physicians, while providing important new sources of information, are some current examples of things that vie for the personal attention physicians must give to their [144]

What Should Be Taught?

[145]

patients. The skills—for example, taking a history or doing a physical examination—considered as things in themselves, can also become diversions unless viewed in terms of patient care. Why? Because, fascinating as technologies and skills may be, patients are complex and the questions they pose are often enigmatic. Take this case, for example. Jack Darnell is a sixty-nine-year-old retired business executive whose prostate cancer was discovered three years ago, after it had already spread beyond the gland. He required no treatment for a year and then was started on ¡utamide and luprone, with a good response. Recently, he developed extremely severe pain in his back and was discovered to have a metastatic tumor of the spine at T9 (the ninth thoracic verteba) which was believed to be the source of his pain. The distribution of the pain does not seem correct if the metastatic lesion is the only source of pain. Mr. Darnell has just ¤nished radiation therapy to the area. His pain, initially relieved by morphine in a patientcontrolled intravenous pump, has become much worse. This morning, as the house staff makes rounds, he is visibly frightened by the awful pain of a few hours ago. He wonders whether yesterday’s vigorous physical therapy set off today’s pain. His anxiety for his future occupies his utterances. His wife is putting cold compresses on his lower back, and as he lies on his side, his feet are in constant motion. The intern, resident, and medical student have just left his room, their platitudes of reassurance sounded empty, even to them. Their discussion centers on how much morphine each activation of the pump should now provide.

Why is he in so much pain? Is it yesterday’s physical therapy? Is edema from the radiation therapy making the pain worse? Is there another metastatic lesion that is now the source of the pain? Should his discharge be delayed? Will he and his wife be able to cope when he is discharged? Is he going to a setting where good pain control will be possible? The house staff’s questions did not address these issues; they were taken aback by his obvious misery when they entered the room. Outside the room, they do what doctors so commonly do when a quandary presents itself—they concentrate on a technical issue. During the day they will probably recover and try and unravel the issues, but the sudden appearance of the quandary unsettled them, and their skills are no match for the situation. The diagnostic and therapeutic skills of the clinical method are designed to provide answers. Answers, however, are always correlative

[146]

doctoring

to questions. Question and answer go together. Answers are propositions, statements such as “He has carcinoma of the prostate.” A proposition cannot be the right answer to any question that might have been answered otherwise. “Why did Jack Darnell have a sudden resurgence of pain this morning?” He has cancer of the prostate. He has a metastatic tumor of the spine. He just had radiation treatment. His pain pattern suggests a new lesion. He is not getting enough morphine by activating the pump. All of these propositions, correct in themselves, cannot be the right answer to the question. Each is the right answer to some question raised by Jack Darnell’s illness. Each of the different questions is important in itself.* Each answer cannot be evaluated separately from the question.1 If this is the case, then the clinical method must be based not on ¤nding answers, but on ¤nding the question-and-answer pair that is central at a particular time. This simple case also makes the point that question–answer pairs nest within each other—a sequence of questions and answers, each following naturally from the preceding one. For example, the fact (proposition) that the concentration of morphine in Mr. Darnell’s pump was not suf¤ciently high is not an answer to the question of why his pain suddenly surged this morning, but it does explain why he did not get adequate relief when he used the pump. This question–answer pair leads to a therapeutic action. Which is, in fact, the point of each question–answer pair; the development of a therapeutic act. Medical actions are speci¤c things that are meant to be done to or with a patient. Since they follow from the answer to a question, the answer itself must be detailed and particularized, which implies that, properly asked, the question must be equally detailed and particularized. The clinical method is based on formulating detailed, particularized question–answer pairs. The true unit of clinical thought is a question and an answer in an inner (or outer) dialogue. Considering, for example, acute appendicitis, you may not accept this statement. The diagnosis is not highly detailed and particularized; yet it is the correct answer and leads to a speci¤c, detailed, and particularized action—an operation on a particular * What is the right answer? The physician will have to go back to the patient, question him (and his wife) carefully, and then examine him with care. When this is done, I believe he will be found to have new tumor growth much lower in his spine requiring further radiation therapy. However, his increased pain will have to be relieved, and his loss of con¤dence in the future must be addressed.

What Should Be Taught?

[147]

patient by a particular surgeon using a speci¤c technique. The question it seems to be answering is general: what is the cause of the patient’s right lower quadrant abdominal pain? But a moment’s re¡ection will show you that neither the question—“why does this patient have right lower quadrant pain?”—nor the answer—“acute appendicitis”—is vague or general. As one asks questions of the patient and performs the physical examination that leads to the diagnosis, a particular patient marked by myriad individual details is being considered. Similarly, the decision to operate on a particular patient implies more than just the diagnosis “acute appendicitis.” The patient’s suitability for surgery, the state of the appendix at this time, other diseases, allergies, and many other features have been considered in the decision to act. Thus the words acute appendicitis are a shorthand for the whole long diagnostic statement in the clinician’s head that is the answer to the complex and detailed question about this patient with right lower quadrant pain. The question–answer method is applicable to every clinical setting and every step in the clinical process. Virtually every patient contact with a physician—admission to the hospital, subsequent hospital visits, of¤ce visits, home visits, telephone calls—raises diagnostic questions for which answers may be required. The formal exercise called the history and physical examination, to which most attention is given in clinical teaching, was designed for admission to the hospital of patients with acute illness or an acute event in a chronic illness. It is a necessary ¤rst step in medical care, but it is only one step. Clinical skills are grounded in the concept that in every presentation of a patient a detailed, particularized question(s) is posed for which there is a detailed particularized answer(s). Another way to look at this process is to see the question as exposing a knowledge gap that has to be researched, either with the patient, on physical examination, or in the literature. The clinician’s task is to discover the question(s) and attempt to formulate answer(s). It would be nice if the answers were true, but that is not of primary importance: What is necessary is that the answer be right enough to allow the question-and-answering process to continue. When it is realized that the questions raised by patients’ problems are always historical and situational—the question “Why is this person sick?” is a very different question in 1996 than in 1976 and in New York City than in Boise, Idaho—the meaning of detailed and particularized becomes clearer.

[148]

doctoring

The Clinical Interview The ¤rst and most basic clinical skill is the clinical interview. All medical students have been taught how to take a history. The advanced skills of the primary care generalist will necessarily be built on this foundation. Generally, however, when students take a history they are trying to make a diagnosis—to discover what disease will account for the symptoms they have elicited from the patient. This is a minimalist view of the history. As just discussed, it is the opportunity to begin to understand the detailed questions posed by the patient’s visit. It is also the physician’s chance to start developing the therapeutic relationship on which all further medical actions will depend; it provides the ¤rst occasion to observe the patient in every detail of his or her presentation to the world, linguistic and otherwise. It is the patient’s ¤rst opportunity to observe and judge the physician. (This is something physicians have to actively think about as they take a history. Patients are trying to discover whether physicians are taking them seriously, are on the right track, or are empathetic. When things are going well, they feel understood.) Because of its richness, the history is one of the central moments in clinical medicine.2 Many good studies have been done and excellent works written on the interview process. Recent studies have gone beyond the initial history to discuss the dynamics of communication between patient and doctor in general.3 (It is sometimes forgotten that communication is not an end in itself.) These publications provide an excellent basis for the development of teaching materials and methods in patient–doctor communication. Most medical students have never been observed doing a history and physical from beginning to end, so they graduate with variable degrees of pro¤ciency. The abilities to interview, examine, and observe a patient constitute the methods on which the entire structure of general medicine rests. There is no substitute for teaching these skills with great care, no matter how labor intensive the process may be. In recent years, critiquing trainees’ videotaped interviews has been employed successfully, as has the use of simulated patients. Properly trained simulated patients have the advantage of providing immediate feedback to their interviewer. Simulated patients are not only very effective but are also an inexpensive method of teaching. (Dennis Novak, personal communication. Professor Novak is at the Medical College of Pennsylvania.)

What Should Be Taught?

[149]

Videotaping, although widely employed, has the disadvantage of being expensive and intrusive. The camera does not see exactly what the physician or patient sees without the use of costly and distracting equipment. I believe that audiotapes are an excellent, unobtrusive, and inexpensive way to “attend” in the clinical interview and other interactions as well. Tapes can be critiqued by an instructor with attention to the actual language and paralanguage, without the distraction of body language. Nonverbal communication is undoubtedly important, but its interpretation lacks precision at this time and, as a consequence, is open to mere opinion. The spoken words, however, are what they are. Good-quality recording is important so that the listener is aware of the conversation rather than of the recording per se. For this purpose, good microphones (preferably transmitter microphones) and stereo recording are best. The quality of the tape recorder is less consequential. It is desirable to get trainees into the habit of listening to their own interactions so that they can continue to do it as a lifelong learning technique. The tapes can be listened to at any time, even in a car, and learning from them can become a habit. Elsewhere I have discussed at length the importance of teaching students and residents the workings of the spoken language on which all verbal interaction is based.4 This foundational knowledge is as important to interviewing as knowledge of anatomy is to the physical examination. The key to understanding meaning and the symbolic basis of a patient’s behavior is language. In fact, understanding language is as important to understanding persons as understanding anatomy is to understanding bodies. Perhaps it is dif¤cult to convince physicians of this because it seems that if one speaks a language then one must know how it works. Or because understanding language does not seem to be medical knowledge. As I noted in the Introduction, the person is in many ways a new object of medical attention and requires new methods of observation. Language is one of them. Language is also a therapeutic tool. Virtually every word that doctors say to patients has an effect. Forty-three years ago, when I was a medical student working as a subintern in an emergency room, a young man with an inguinal hernia came in. As was the custom, we called the surgical resident to con¤rm the diagnosis and decide what to do. She knelt in front of the patient to check the

[150]

doctoring

hernia and then looked up at us and said, “His penis is awfully small, isn’t it?” I mention this because I’m sure that despite the many intervening years, that man has never forgotten what she said. It is important for physicians to think about the impact of their utterances. Rhetoric is an important aspect of language usage that should be part of the clinical method; communication goes both ways. Listening to recordings of yourself interacting with patients provides an opportunity to know yourself as object as well as subject. Under most circumstances, we are only subjects of our experiences. Through our senses we experience the world interpreted by our meanings. Objects, events, and relationships are what they are and mean what they mean to us because they are evaluated by criteria developed from infancy. Others, obviously, see things differently. One function of professional training is to develop meanings common to the profession; blood, for example, means different things to doctors than to others. Professionalization, the development of a doctor persona, follows from the ability to interpret objects, events, and relationships as a doctor would rather than as the person of the doctor might. Blood and gore may be simply handled by a trainee, but sexual feelings arising from an encounter with a patient are not so easy to address professionally. This is where coming to know yourself as an object is important. While listening to audiotapes or watching videotapes of your interactions, it is possible to evaluate them employing the criteria you might use for another person. You are an object of your own knowledge. From this point of view, for example, the physicians may realize that those sexual feelings are not about having sex, but about the patient’s feeling of closeness. With enough practice and perseverance, you can routinely be both the subject of your experience and an object of experience. When this happens, you can hear your own words and see your own actions, as well as subjectively experiencing your impact on the patient. This adds an important dimension to the concept of physicians as their own instruments.

Physical Examination The videotape shows the ¤rst-year resident taking the history and examining a patient who complained of having a congested head,

What Should Be Taught?

[151]

postnasal discharge, and cough. When she gets to the examination of the chest, the resident lifts the patient’s sweater and listens brie¡y with her stethoscope to both posterior bases. In front, she puts the stethoscope down through the top of the sweater and listens to one place, presumably over the heart. The discussion of the videotape is about her history and presentation to the patient; the inadequate examination is not mentioned. If the resident’s physical examination is so poor in the eighth month of her ¤rst year as a physician, what will it be like in eight years? The physical examination of patients is done for four reasons. The ¤rst is to gather information about abnormalities or disease to help answer the question(s) raised in the interview. The second reason is to see the patient as a whole; a functioning organism-in-motion. The third reason is to provide the physician with a spectrum of what is normal against which all future patients’ ¤ndings can be compared. A patient has a terrible headache that has been worsening for two hours. The doctor wants to examine the ocular fundi for signs of increased intraocular pressure. If the physician doesn’t use the ophthalmoscope all the time, he or she might as well not bother because it requires practiced knowledge of what is normal for the examination to be useful. In the same manner, it is almost pointless to do pelvic examinations on women with lower abdominal pain unless physicians do them all the time. What excuse can doctors offer themselves (to say nothing of their patients) for having to send patients to other doctors just for a pelvic examination? The fourth reason for doing physical examinations is the need to touch the patient. For this reason, it is necessary that physicians learn how patients judge the sensation of the doctor’s hands. The physical examination, like the history, is part of the treatment. There are many good textbooks on the physical examination, as well as secondary materials— videotapes of special aspects of the physical examination, heartsound recordings, and the like. Thus, a wide choice exists to ¤t almost any teaching setting. The emphasis of this book on the physician as instrument, the return to the primacy of clinical observation, and view of clinical thought as a dialogue between question and answer should guide the use of these teaching materials. When medical students learn to perform a physical examination, they are too often taught as though the object was to learn, for example, to see the eyegrounds, hear a certain heart murmur, or feel the spleen as things in themselves, with no emphasis on their impor-

[152]

doctoring

tance to this patient. This is natural enough in the beginning because students may feel like intruders or may be embarrassed at their ineptitude, and so may have trouble putting their ¤ndings in context. The physical examination is so crucial an element in clinical observation that as trainees conduct examinations of patients, they must be observed at least once. And once is not enough. Don’t we want to see evidence that pro¤ciency is increasing? We observe the observers not only to check technical pro¤ciency, but also to see if they understand and are using their examination as part of their reconstruction of the patient as a narrative object of medical concern. How did this person get to this point in the illness or life (and what will happen next)? This process is a variation on formulating particularized questions and answers. This case demonstrates the point. Lilly Ho¡etter is a forty-year-old woman who had mild asthma with severe allergy to pollens, animals, and dust. One day she went to an antique sale at a very dusty loft building. When she returned home after an hour, she felt a suffocating feeling (without wheezing) and went to a neighbor’s house, where she collapsed. She had respiratory arrest in the ambulance and was intubated and hospitalized for one week. She was all right for about a month and then hospitalized again with an exacerbation of asthma. She said the hospitalization resulted, in part, from a feeling of panic. Her medication is adequate. She has come to a new physician (reluctantly) because of a friend’s recommendation. This is the tenth consultation she has gotten in three months. She measures breathing function daily (or more) with a peak ¡ow meter. It varies as much as 50 percent from day to day, but nobody has explained why. Measles encephalitis at age six was followed by grand mal epilepsy, for which she still takes medication. She has moderate scoliosis, for which a back brace offered at age twelve was refused. She was initially hostile and not forthcoming with information. She sat in the chair in almost constant motion, slumped over, chin on chest, moving this way and that with peculiar mannerisms. She walked like a Raggedy Ann doll at odds with itself.

Why did she have a respiratory arrest? Why does she have such variable lung function? Why does she move so strangely, and why does she have such an unusual relationship to her body? Why has she seen so many doctors? What doubt of hers has remained unresolved? With the possible exception of the ¤rst, none of these crucial questions can be answered without a thorough interview, careful

What Should Be Taught?

[153]

observation of her, and examination of her body. A routine chest examination will certainly not be suf¤cient. Seen this way, this is a fascinating problem. Seen this way, so is almost every patient. Seen this way, almost no physical examination is routine.

The Description of the Patient In Chapter 5 I described the role that a careful written description plays in the clinical method. After the clinical interview and examination of the patient, the ¤ndings must be written down. The description harvests the yield from observation. Training programs must give as much attention to this step as to the previous two. Faculty from English departments can be enlisted to help. I dictate my notes at the end of of¤ce hours, and I struggle to describe a patient in more than monosyllables because I was never trained to describe patients. (Of course, I can describe patients and write a narrative while sitting at my computer thinking only about the description and given enough time—but that, alas, is not what happens in medical practice.) Teach the physicians in your program and they will do it properly the rest of their lives and probably enjoy it! We’re not looking for Mark Twain, but good clinical descriptions can form the basis for signi¤cant clinical research.

The Doctor–Patient Relationship Everything described thus far, indeed, every step of medical care, requires the cooperation of the patient. People, sick or well, do not do things just because they should or because a physician tells them to. They must trust that the physician is knowledgeable, correct in the instance, and caring about them. The world of sickness is ¤lled with far more uncertainties for patients than for physicians; as the threat of a disease increases, each decision seems more crucial and the room for error narrows. As the patient’s uncertainties increase, ¤rm resolution becomes less possible. Patients solve this awful dilemma, in large part, by trusting their doctors. They say, in essence, “I may not know what to do but my doctor does.” The vehicle that makes it possible for you to trust someone you may have known for only minutes (in emergencies) is the doctor–patient

[154]

doctoring

relationship. Doctors and patients may have ¤nancial relationships, political relationships, social (in the everyday sense of the word) relationships, personal (in the sense of two persons) relationships, or even very private relationships. When I speak of the doctor– patient relationship, I am speaking of the basic and central relationship that exists solely because one person is a physician and the other is a patient. They are connected by the fundamentally benevolent and common goal of improving the health of the patient. I have argued elsewhere that this is a relationship based in the inherent social aspects of being human.5 It is bilateral, as crucial to the doctor as to the patient. When it functions best, it makes it possible for patients to change at every level of the human condition and for doctors in this benevolent relationship to extend themselves at every level of the human condition. These levels are social, emotional, personal (persons’ relations with themselves), spiritual, and physical. Just as it may do immense good when properly employed, it may do harm in the hands of physicians who are ignorant, careless—or worse. How it works is, in these rational times, seemingly mysterious. That it does all that I have said, and much more, is unarguable. Its careful, systematic, and disciplined development and use is essential to good doctoring. This is especially the case in primary care medicine, where the doctor has such broad functions. This is one of the functions of critiquing audio and videotapes and of simulated patients. Trainees can be shown how their actions promoted or discouraged the relationship. They should be taught what they are attempting to achieve, as well as the limits of the doctor– patient relationship. One measure of the training of primary care physicians, then, is their ability to develop and maintain effective therapeutic relationships. In one setting, it can be actively taught and encouraged; in another, discouraged and disparaged. A test of generalist training programs, therefore, is whether they heighten the possibility for patients and physicians to develop a good relationship. The revitalization of primary care medicine through a return to the primacy of the patient can be accomplished only in an atmosphere that recognizes the value of the doctor–patient relationship and enhances it. What kind of person is the patient in the modern relationship? In the past, the science of medicine seemed to consider the patient as the container of pathology—passive to the disease and the physician. The disease came ¤rst in understanding the sick patient: same disease,

What Should Be Taught?

[155]

different patient, same problem. The patient in the contemporary relationship is known to be inextricably entwined with pathophysiology and pathoanatomy, so that the same disease in different patients may have a different onset, course, diagnostic complexity, treatment, and outcome. This is true even for many (although not all) infectious diseases, but it is absolutely and unavoidably true for all chronic diseases. Since we now know that knowledge of individual patients is essential—core, not peripheral—to medical care, the patient becomes conceptually much more important. The impetus for primary care has occurred along with the growth of bioethics, with its emphasis on patient autonomy. Patient-centered medicine, patient as partner, patient as director of care, and patient as consumer are phrases currently in use that re¡ect the increased power of the patient in the medical care calculus. The relationship of primary care to its public as it grows in strength is vastly different from the public climate in which specialty medicine has developed; both patients’ knowledge and their power will inevitably in¡uence the shape of generalist medicine. Knowledge of persons takes its place alongside science as a fundamental knowledge base. It is not clear what knowledge of persons means. Until recently it was considered behavioral science, a discipline of which current graduates may know little. It is not uncommon to ¤nd graduating medical students who have had no training in normal or abnormal psychology, no experience with non-drug treatment of emotional illness, and no instruction in the psychology of physical illness. Thus, at a time when the patient is becoming central, there is little of the systematic, disciplined knowledge about sick (or well) persons required by doctors for effective clinical action.* There are some Departments of Humanities in Medicine that teach subject matter about persons, but they are frequently salvationist in their goals, trying to make physicians more humane, so that they fall short of having a desired knowledge base. The same problem may be found when psychiatrists or * Why does Lilly Ho¡etter, of the case cited earlier, move around so much? Everyday experience suggests that people do this to avoid being seen. If so, what is she trying to hide, in the manner of a child? Herself? Her kyphoscoliosis? Does solid evidence exist somewhere on the subject? What is the effect on adults of a childhood illness like epilepsy? Has this been studied systematically? These are important questions for generalists, and they highlight knowledge gaps that must be addressed for effective clinical action.

[156]

doctoring

psychologists teach medical interviewing. The medical interview is different in structure and intent from the psychiatric or psychosocial interview, although they share some features. In the absence of general agreement about what knowledge of persons is, providing primary care doctors with effective communication skills, trained observational skills, and the ability to describe (narrate) in speech and writing what they know of patients gives them a basis for learning from their experience.6

Clinical Judgment Medicine is a profession requiring judgment, but judgment is a complex problem. In Chapter 4 I showed the dif¤culties that beset the understanding and teaching of judgment, as well as the fact that there is no clinical medicine in its absence. In fact, there is no way of living in the absence of judgment since every time a choice is made in which error is possible, judgment is involved. Unfortunately, as noted previously, there is no rule or principle that can be taught to guide the application of general knowledge to a particular circumstance—the de¤nition of judgment. So judgment cannot be taught in the abstract, but must be practiced from case to case.7 Given all of these dif¤culties, how are we to teach judgment to students and residents? I believe we should, in effect, retrace Alvan Feinstein’s introduction into medicine of clinical epidemiology and the mathematics that underlie it. One effect of starting with what is now called clinical epidemiology is to rid trainees of any idea they may have that simple (untrained) common sense is adequate for effective clinical medicine.8 The ¤elds of clinical epidemiology and clinical decision making are ideal for didactic instruction. These tools represent a major step in systematizing medical judgment, and they should be part of every clinician’s training. An excellent literature exists from which they can be taught, and knowledgeable physicians exist throughout the country.9 The basic concepts of clinical epidemiology and its methods for the interpretation of diagnostic information should be almost second nature by the time primary care physicians ¤nish their training. They should be able to read and interpret the medical literature. Unfortunately, it is impossible to ensure that physicians will be lifelong learners and that they will stay abreast of

What Should Be Taught?

[157]

the literature, but the task will be easier if they bring good interpretive skills to their reading. As currently described, however, clinical epidemiology includes only a portion of the spectrum of clinical judgment—almost exclusively measurable information, including clinical events of treatment or diagnosis represented by laboratory or other quanti¤able studies or endpoints. Some of the more recent publications by David Sackett and his colleagues at McMasters University discuss the speci¤city and sensitivity of physical ¤ndings, but here caution is necessary. Not all physicians have equal skills. An EKG is an EKG is an EKG, but this is not true of the hands feeling for the spleen. Samuel Levine, a wonderfully skilled cardiologist of the past, was asked about a case of endocarditis where no heart murmur was heard. He asked, “Who listened?” Clinical medicine abounds with important information that is nonquanti¤able; subjective information, values, feelings, aesthetics, and intuitions—each of which must be given its weight and, like that which is measurable, considered in terms of con¤dence limits, validity, accuracy, and precision. Yet the methods of clinical epidemiology have limited utility here. Primary care physicians should be able to use all kinds of information to examine their presuppositions, separate and examine the values at issue in each judgment, decide on goals and argue the alternatives, decide on priorities of judgment, determine if and how to intervene, and be able to evaluate the outcome. Clinical ethics, although obviously part of judgment, is a separate and very necessary aspect of training. I noted in Chapter 2 that research into clinical behavior has shown that much clinical decision making to be rule-governed, automatic “recipe” thinking. Clinical teachers may be unhappy to discover this because we like to think of medicine as a profession guided by reason. However, we shouldn’t be discouraged. In everyday life, many of our actions follow the rules. This is what makes them successful. This is because we move among situations of certain standard types, and we want to achieve standard results. Action according to rules is a very important kind of action, so the ¤rst question any intelligent physician asks is “What are the rules for acting in this circumstance?” Nevertheless, although acting by the rules is important, there are two situations in which another kind of decision is necessary. The ¤rst is a situation that is not recognized as belonging to any of the types the physician knows. In

[158]

doctoring

the second situation, the physician can place the condition in a known category and follow the rules but is not content to do so. Rules, being general entities, always involve a certain mis¤t between themselves and the situation. Often that does not matter, but sometimes it is very important. Of these two circumstances in which it is necessary to act without rules, the ¤rst arises out of ignorance and inexperience and is most common among, but not restricted to, young physicians. The second comes with great experience, and sagacity, and awareness of how great the stakes can be.10 A recent author discusses the rules, which he calls heuristics, by which physicians make decisions when there are no solid guidelines to action. He points out that such situations far outnumber those in which decisions are based on established scienti¤c facts.11 Thus, for both the inexperienced and the seasoned physician, the importance of actively teaching good clinical thinking cannot be overemphasized. Analyzing a case, deciding what questions are raised, and formulating the answers are central to the clinical method. It is not unusual to hear a case presented by a resident, with the attending physician’s questions concentrated on facts about the disease. These are important, of course, but they can be answered by a text or computer, while the resident’s thinking can only be helped by the teacher. A. N. Whitehead said that the secret of clear thinking is to stay on the subject. This seems obvious until one hears case discussions in which the subject changes two or three times in a single utterance. The master’s classes described earlier are an excellent vehicle for teaching clinical thought and judgment because they concentrate on problems that may arise in any aspect of a case. Further, trainees have the opportunity to hear clear thinking from their teachers in an informal, noncompetitive setting. This is an area where excellent curricular materials should be developed over the next several years on subjects beyond the boundaries of clinical epidemiology and decision analysis.

Information Handling One of the most exciting aspects of contemporary medicine is the marked improvement in the quality of information about disease, medicine, medical science, behavioral science, and related subjects. The ability to have easy access through the computer to this infor-

What Should Be Taught?

[159]

mation is an essential aspect of any physician’s skills. Arguments during teaching sessions about questions of fact that can be answered by the medical literature should be passé. It is easy to access Medlars or other databases through programs like Grateful Med from the National Library of Medicine. The Internet has created additional opportunities for information acquisition. Thus, training programs for generalists must ensure that physicians are able to gain access to and have an understanding of the new ¤eld, often called informatics, that is so dependent on the use of the computer.12

Technology, Diagnostics, and Therapeutics The use of diagnostic and therapeutic technologies is usually learned as part of postgraduate bedside training. Specialists in training, such as radiologists or gastroenterologists, are taught to use the technologies of their specialty, and other physicians turn to them when necessary. The problems of rampant technology, discussed in Chapter 2, demand a separate and speci¤c solution. The proper diagnostic and therapeutic utilization of technology should become a subject of formal study and training. The content of this training should be didactic, and its context should be separate as well as part of the patient-care setting. The primary care physician is being taught diagnostics—how diagnostic goals are set and how technology is employed to meet these goals. Technology admits of general principles that can be taught by the specialists in each technological ¤eld. In addition, clinical epidemiology, discussed above, has developed rational methods for deciding when a speci¤c diagnostic technique will add useful information or increase diagnostic accuracy. Generalists must know how and when and (when not) to use each diagnostic technology, even when that technology falls within the purview of another specialty. Their knowledge should be suf¤ciently generalizable so that, as new technology arises, they can master it. imaging Diagnostic radiologists are imaging experts. They know the best way to demonstrate a particular body part, make abnormal struc-

[160]

doctoring

tures evident, reveal pathophysiology, and use their technologies in an ef¤cient and economical manner. Too often, physicians request a particular study without discussing with the radiologist whether it is the best or most ef¤cient means of accomplishing the clinician’s purpose. Sometimes issues of diagnostic accuracy arise. For example, in a particular instance, is magnetic resonance imaging (MRI) or computer tomography (CT) the best method to visualize suspected intracranial disease, or is it useful, in light of the diagnostic hypothesis, to do MRI if the CT scan is negative? At other times, questions of cost effectiveness arise. For example, in this patient, is it best to do a barium enema or a colonoscopy, considering the diagnostic hypothesis, cost, and diagnostic effectiveness? It makes little sense to expect primary care physicians to learn the complexities of imaging through experience alone; the diagnostic and economic stakes are too high. Radiologists do not make up things as they go along; they work with basic concepts, scienti¤c evidence, and empirically demonstrable facts, all of which can be taught. They are heavily dependent on a knowledge of anatomy, and the best ones know anatomy best. At one time, I believed that abdominal sonography should be as accessible to the generalist as chest X-rays are—something that could be used with con¤dence to increase the probability of detecting abdominal lesions. Sonography, however, has become increasingly complex, and the equipment is expensive. It remains true that primary care physicians should be able to read sonograms, chest X-rays, echocardiograms, abdominal CT scans, and back-and-joint images with considerable con¤dence. This is not because of lack of faith in their imaging colleagues, but rather because all of these images have brought anatomy and anatomical truth back into medicine, and primary care physicians should once again know anatomy if they are to understand their patients’ problems. The physician has the advantage of turning back to (say) the knee joint after looking at the MRI image. As an example, consider Julia, seven years old, who jammed her toes into the steps during a headlong dash. The next day her foot hurt pitiably, and her mother took her for an X-ray. The radiologist wondered whether there was a traumatic separation of a growth plate. That was not where the foot was sore or tender, however, and Julia went back to school. The truth of the clinical problem posed by the foot lies in the foot, not the X-ray. It should be axiomatic that all special tests and studies should be done to

What Should Be Taught?

[161]

con¤rm or refute the diagnostic hypotheses based on the history and physical examination.* MR images have enlarged the concept—given depth and richness to the understanding—of anatomy that can now be combined with the knowledge of the patient, the patient’s symptoms, activities, function, and so on. This next case, which is common enough, shows the same idea applied to a particular patient. Naomi Larkin-Price is a seventy-eight-year-old widow who is active in the management of her real estate. She came to the doctor because of swelling and pain in her right knee, which prevented her from her usual vigorous walking and for which, in the last three months, she has taken to using a cane. She brought her MRI scan with her. She said that she had seen “eight doctors” in the last year since the knee became swollen and “nobody knows what’s the matter with my knee.” These included three orthopedists, a rheumatologist (who found ¡uid in the joint), an internist, and three acupuncturists, the last of whom made regular house calls but stopped after ¤ve weeks because he didn’t think he was helping her. She “never” went to doctors, except rarely to her gynecologist for a Pap smear, and considers herself entirely healthy and young for her age. She was decidedly against X-rays and medications. She had been trying some exercise for her knee that she saw in a magazine. She reluctantly consented to a complete evaluation, which was normal except for calci¤cations of her aortic arch, an elevated cholesterol level, and her right knee. The knee was swollen; there was a small effusion; a ¤ve centimeter popliteal (Baker’s) cyst was evident posteriorly; ¡exion was limited by pain when her heel was ¤fteen centimeters from her buttock; and there was weakness of the musculature around the joint. The left knee was normal. The MRI scan taken ¤ve months earlier showed a tear in the posterior horn * I have not included routine tests such as a complete blood count (CBC), urinalysis, chemistry panels, EKG, or chest X-ray as special tests. I consider them to be part of the modern complete physical examination. Others disagree, believing that these should be done only if considered necessary on the basis of the history or the physical examination. Otherwise, it is usually said, only speci¤c tests should be ordered. Why? I perform the whole physical exam, not just those parts suggested by the history. I think one should come away from a complete examination with as complete understanding of the patient as possible, not just of the patient’s surface. It may be that issues of cost make my suggestion impractical. If so, let’s be honest about the matter and not pretend that these tests should not be done for some other theoretical reason.

[162]

doctoring

of the medial meniscus, mild degenerative change, and the popliteal cyst. She was interested in the MRI scan and how its ¤ndings applied to her knee. She was relieved to know that, as the orthopedists had said, she would not require surgery. She started with the physical therapist, did her exercises and weight training, and was walking considerably better within several weeks.

Endoscopy is a means of directly visualizing internal structures. It has become a way of extending the physical examination. Some endoscopic procedures are complex and time-consuming and require extensive training. The modern ¡exible sigmoidoscope is a powerful tool in the hands of a generalist who has been trained in its use. Such physicians can also be taught to visualize the oropharynx. When a primary care doctor uses an endoscope, he or she should not be “doing a procedure,” in the current sense of a billable product. Rather, the physical examination is being extended because information from the history or other parts of the examination raise questions that endoscopy can resolve. Many other technologies, such as EKGs, pulmonary functions tests, cardiac procedures (treadmill exercise tests, echocardiograrns, thallium scans, angiograms, and the like), even automated blood chemistry testing, and others yet to be developed must be learned by generalists if they are to be masters, not slaves of their technologies. Learning these things also leads to more appropriate use of specialists. A caution is necessary. If skills such as endoscopy are not used regularly, then physicians are unable to identify early pathology. Or, if they do, they have little con¤dence in their ¤ndings and send patients for further studies. The essential knowledge involves not how to do these things, but knowing the indications for their use and how to evaluate the results. Others may believe that these studies should be done only by specialists who do many of them, thus increasing their accuracy and decreasing the patient’s discomfort. For them it is suf¤cient to extend their examination of the patient through the hands and eyes of the specialist. Keep in mind that in teaching about technology, it is necessary to keep coming back to the problems of uncertainty, ambiguity, wonderment, and power that have kept modern technology in the saddle. This is another area where teaching the values and virtues of physicianhood must be central aspects of a training program. As I noted previously, it is impossible not to teach both values and virtues, so they might as well be carefully chosen and deliberately taught. Here is where the faculty’s own actions count so heavily.

What Should Be Taught?

[163]

therapeutics At present, therapeutics—the use of pharmacological, biological, or other agents—rests on the pharmacology taught in the second year of medical school. However, modern therapeutics has advanced far beyond the stage that a second-year medical student can comprehend. Further, decisions about its deployment are too complex to be taught only by precept or in conferences during postgraduate training. These complex therapeutics and pharmacodynamics are thought to be the province of the subspecialist, but if primary care physicians are to carry out the role envisioned for them, they must understand the theory behind their use and be prepared for the new agents to come. The effect of inadequate training is evident not only in the use of new drugs or treatments. The problem of inappropriate use of antibiotics is well known, but the size of the threat posed by emerging multidrug-resistant organisms is now becoming clear.13 This is an example of the essentially untrained utilization of a class of therapeutic agents that has been in use for more than ¤fty years. As another example, rational use of analgesics for pain control is uncommon in even the most advanced institutions. Physicians are simply poorly trained in their use. In hospitals that have pain control units (generally directed by anesthesiologists), the relief of a patient’s pain is often turned over to the specialist, which has the effect of ensuring that other physicians will not learn how to manage patients with pain. This is typical of advanced therapeutics in general; the subspecialist learns how to use drugs (for example, chemotherapy), but the generalist remains ignorant. There are occasions when it is appropriate to call on a subspecialist, but routine abdication to the subspecialist is a result of ignorance. Here is an area where we do not lack for teachers; the clinical pharmacologists have been waiting for our call.

Knowledge of Function and Process The change of goals presented in this book requires a changing understanding of the nature of illness in which the continuum of function ¤gures more heavily than isolated disease states. This medicine is concerned less with events than with processes. Processes are things that do not begin and end but turn into one another; this is particularly easy to see in chronic illnesses. If one process turns into

[164]

doctoring

another, there is no point at which the ¤rst one stops and the second one begins. In fact, process one never stops; it merely goes on in a changed form in process two. Process two never starts; it has merely been going on in an earlier form in process one. The point is made in the case of Lilly Ho¡etter, described earlier, where the impact of her childhood illness continues on into the current illness. If one considers the illness of childhood epilepsy to be only the disease of epilepsy, this is hard to see. If, on the other hand, the illness is understood, for example, to be the disease, her reaction to it, the treatment, and her parents’ attempts to raise her “like anyone else” (which she described), then the persistence of childhood epilepsy within the asthma is obvious. Process one (the epilepsy illness) is living and encapsulated within the asthma; their “colors” combine to produce what is currently seen as asthma. This understanding challenges simplistic views of the separation of physical, psychological, and social elements in an illness and has an impact on how physicians formulate the detailed, particularized questions raised by the patient and plan their answers. In medicine, concepts and theory are rarely taught as such; physicians have poor tolerance for overt theory. Rather, certain groups of patients exemplify speci¤c problems, and it is in teaching about them that the concept is taught. The chronically ill, the elderly, and the disabled are such populations. In the Introduction and in Chapter 2 I discussed the necessity for coming to terms with the differences between chronic and acute disease. By virtue of prevalence alone, primary care is a medicine of chronic disease. It may be dif¤cult for clinical teachers, themselves raised and trained in the model of acute disease, to change their orientation in the education of their trainees. This is demanding but essential. Irene Stor, the ¤ftytwo-year-old diabetic woman described in Chapter 2, illustrates the dif¤culties that follow from actions planned in response to pathophysiological events as though the illness was not a continuum from adolescence, as though there was no yesterday and will be no tomorrow. Part of the intellectual excitement of generalism is the need to ¤nd new ways to understand and teach about such illness. Because of changing demographics, the aged are increasing in numbers and importance both medically and politically. The size of this population makes it impractical to have medical care delivered only by geriatricians. Yet, treatment suffers when traditional disease models are employed to evaluate the problems that accompany

What Should Be Taught?

[165]

advancing age. Older patients frequently have several distinct diseases and physical ¤ndings that would be a cause for alarm in younger patients. If each disease is vigorously pursued either diagnostically or therapeutically, disaster may ensue. Similarly, if the various diseases are divided among subspecialists without careful coordination (a rare happening), medication problems and other problems commonly arise. Diseases like osteoarthritis, commonly brushed aside with the question, “What do you expect at your age?”, are a major source of disability. These can often be successfully managed, keeping an old person active and functional. Of particular importance are ethical issues raised by illness at the end of life. The relentless treatment of disease that would once have ended a frayed old age has created a population of almost functionless, often demented old people seemingly abused by their medical care.14 Despite a great deal of thought by many, however, it is not clear how physicians should act.15 The training of generalists cannot evade the responsibility for teaching about ethical problems in medicine in general, and speci¤cally in the case of the aged and the very young. One concrete question raised by an individual patient may be “Should this patient’s disease be treated?” (The question is not whether the patient should be treated; all medical care is treatment.) Primary care physicians must be trained to provide care at all stages of human development. In the early days of their specialty, it took pediatricians a long time to make it clear that children are not just little adults. Their special physiology, psychology, needs, and problems are now universally recognized. Equally, we must understand the special problems of the care of the aged. Similarly, awareness of the fact that many aged patients are remaining active in all spheres of human behavior, even at advanced age, gives impetus to the understanding of developmental medicine as important to the old as to the young and all those in between. The evaluation of disability, and the promotion of improved functional status and return to work, exemplify a category of knowledge and skills that are important to the training of primary care physicians. Persons with disabilities have achieved special status in the society that shifts their medical problems from peripheral to central importance. In both the aged and the disabled, thinking in terms of function rather than disease states is crucial to successful doctoring. Understanding disability is impossible in the absence of a knowledge of the relationships of individ-

[166]

doctoring

uals to their families and society. None of us exists in isolation, but because all of us function physically and mentally within the societal norm, many of the demands, limitations, stipulations, and opportunities of social life are invisible to us; we think they arise from us as individuals. Such automatic interaction is often not true of those with disabilities, and physicians must be aware of this. Also necessary in the care of both groups is a knowledge of musculoskeletal disease and dysfunction that is currently rare in physicians trained in the United States. Schools of osteopathic medicine have been successful in teaching the necessary concepts and skills. Rehabilitation has been one of the major advances in American medicine since World War II, yet it remains peripheral in the educational process. In understanding disability, physicians must look beyond the abilities of the individual to the functional demands of the workplace and the community. The world of work is an essential aspect of the lives of all people. Its importance in the life of a patient demands a knowledge not only of patients’ function, but also of what can be done to help patients perform the social roles of their age group, and what modi¤cations in the society, the home, and the workplace might be required for their performance. Measuring functional ability is a learnable skill for which criteria are in various stages of development.

Preventive Medicine Preventive medicine has become increasingly important in discussions of primary care.16 The ¤eld of public health has long used the concept in relation to preventing disease in individuals (e.g., immunization) and populations (sanitation and other environmental controls) or in preventing illness in persons already diseased (e.g., treating a tuberculin-positive person with antibiotics). The modern de¤nition of preventive medicine is broader and has its intellectual basis in the understanding of process, described above, in which virtually all illness is a process arising from progressive alterations in function that are in¡uenced by the nature of the sick person and the context—environmental and social forces that affect the person and in which the past inheres in the present and the present in the future. It follows from this that the effect on individuals of the biological processes of disease can be altered—

What Should Be Taught?

[167]

illness can be reduced or prevented—by changes in the behavior of the person and/or by acting on the context of that person’s life. Thus, for example, changing the American diet and promoting healthier lifestyles take their place alongside smoking cessation and daily aspirin use in reducing the prevalence of coronary artery disease and the incidence of coronary events. As another example, changing the home or work environment can have salutary effects on conditions as disparate as asthma and depression. Prevention includes treating persons in whom important disease is present so that their function is returned or preserved, allowing continued activities of daily living and, optimally, return to work or other activities. As a ¤eld of knowledge, the surface of preventive medicine has hardly been developed. (It is important, however, to avoid getting caught up in popular ideas that can lead to an overenthusiastic embrace of questionable prevention strategies.17) Virtually all good medical care has an element of prevention. The effective action of physicians in preventive medicine requires a knowledge not only of conventional medical interventions, but also of the life context and of behavioral or social interventions that contravene the impact of habits or activities that could lead to disability or dysfunction. Simple examples are helping patients stop smoking, eat a healthier diet, or start an excercise program. The patient’s skin is no longer the boundary of medical knowledge. The modern sense of prevention is heavily oriented toward the active participation of patients, with the individual physician (if doctors are involved at all) as teacher. The modern sense of prevention involves a different and more active understanding of the relation of persons to their bodies, their well-being, and their future. It also needs a greater understanding of a normal function as opposed to abnormal function and structure. It requires a greater understanding of human development and extends the spectrum of the potential for change into old age rather than, as is usual, seeing childhood as the paradigm. Despite what I have said above, increasing public awareness of the importance of prevention, and sound clinical reasons for emphasizing it, studies have shown that clinicians often fail to provide recommended clinical preventive services.18 This demonstrates the importance of making prevention a speci¤c area of training for primary care physicians.

[168]

doctoring

Psychiatry Many studies have shown the frequency with which patients with psychiatric disorders seek help from primary care physicians.19 These disorders—for example, depression—are associated with substantial psychosocial and economic costs such as a diminished sense of well-being, excessive disability days, days lost from work, and income de¤cits.20 Sometimes patients’ symptoms or distress are obviously emotional; at other times, the psychiatric disorder is masked by somatic symptoms. Patients who abuse substances frequently are poorly recognized as such and often go ¤rst to generalists inadequately trained for their treatment.21 In addition, virtually no serious ongoing or chronic disease is without an important emotional component. Generally, the training of primary care physicians poorly prepares them to handle either the psychological component of organic illness or manifest psychiatric disease. Excellent diagnostic questionnaire methods have been devised and tested speci¤cally with primary care medicine in mind.22 Preventive medicine, when it is concerned with promoting general well-being and optimum function, cannot disregard psychological aspects of everyday life and the relationship between overall health and mental health. The care of the dying and of suffering patients requires an understanding of how inextricably intertwined psychological, social, and physical aspects of illness are. All of this points to the special importance of the relationship of general medicine to psychiatry. Primary care physicians must learn to manage basic psychiatric disease just as they learn to manage disease in other specialties.23 Current levels of ignorance frequently evidenced by general internists and family physicians are no longer acceptable. On the other hand, there are real dif¤culties to be faced both conceptually and practically, in the integration of psychological and medical skills.24 Psychiatry itself is changing. Departments of psychiatry are beginning to abandon psychodynamic models of treatment and illness for the more descriptive stance represented by DSM-III, DSM-III-R, and DSM-IV and for the psychobiological understandings that underlie modern psychopharmacology. The dif¤culties posed by the changing understanding of mental illness, as well as the destructive in¡uence of managed care on the ¤eld of mental health, must be faced directly in training. Otherwise, the generalist is merely being exhorted to “be holistic,” without being given the skills or a

What Should Be Taught?

[169]

way of caring for patients that can be successful in the current era. A literature exists that is concerned with training primary care physicians in mental health practice.25

The Health Care Team and the Coordination of Care The days of the doctor as a solitary ¤gure in health care are long over. Nurses, nurse practitioners in many ¤elds, physician assistants, social workers, technicians of many stripes, nutritionists, physical therapists, occupational therapists, home health aides, administrators, health bene¤t specialists, and others, including the patients’ families, all take part in the care of patients. This is often true in acute disease but is always the case in the care of the chronically ill. Family physicians frequently ¤nd themselves working with community organizations and health agencies because of their greater involvement in the community and public health problems—a trend that is increasing.26 The imperious individualistic physician, never very attractive, may be downright dysfunctional today. It requires skill to work as a team member with so many different people who have diverse perspectives. The fundamental element in good teamwork is respect for other team members, their knowledge, and their dedication. Teams require guidance to maintain ef¤ciency and purpose, which the physician may have to provide without being overbearing. The combination of leadership skills and respect for others will require repeated emphasis during training. This again brings up the importance of teaching primary care physicians in an atmosphere that does not foster competitiveness (and defensiveness) in the old mold. Another caution is that overdependence on other professionals can produce an inadequate physician. For example, reliance on social workers to handle the psychosocial aspects of illness can lead physicians to avoid learning ¤rsthand how to deal with patients.

The Relationship of Primary Care Physicians to Specialists Otto Guttentag, who, virtually alone, explored the philosophical basis of medical practice and the role of the physician for decades before others followed, always spoke from the perspective of the

[170]

doctoring

attending physician. He believed that the attending physician is the central ¤gure because his or her cardinal and ultimate concern is the welfare of the individual patient.27 From this fact ¡ows the need for one physician, the attending physician, who has primary responsibility. This belief seems somewhat old-fashioned in the present era because of its taint of hierarchy and the notion of responsibility. In a time of almost radical egalitarianism and of the dominance of rights language, Guttentag’s beliefs seem passé. However, re¡ection on the effects illness has on people, particularly serious illness, immediately brings his ideas back to life. The return to primary care physicians is not for their organizational or administrative functions, but for their role as the patient’s attending physician. It is from this perspective that their relationship to specialists should be viewed. Patients frequently self-refer, or insist on referral to specialists even when their primary physician does not concur. This is because most people believe that specialists know more about diseases than the generalist does. While this may be true, who knows the most about the patient’s disease is not the central question. Does the patient have the disease of the specialty? Is this disease the source of the patient’s problem? Is the specialist better able to help the patient than the primary physician? If more than one specialist or disease is involved, who is coordinating the patient’s care? The ¤rst two are diagnostic questions that should be answered by a patient’s primary physician, with or without the help of a specialist. Specialists tend to think in terms of their specialty, have less knowledge of contextual or personal factors in the illness, and use technology earlier and more extensively in the diagnostic process than generalists do. There are many specialists who are superb clinicians and to whom these caveats do not apply, but our question is more general. Who, by training, should be better able to make a complete diagnosis—not only the name of the disease but also the detailed, particularistic questions, including the nature of the threat to the patient, why it is threatening this patient at this time, its relationship to the patient and the patient’s other dif¤culties, and how it ¤ts into the patient’s environmental and relational contexts? Finally, and most important in the long run, who is most prepared to care for the patient? When these questions have been answered, it is possible to specify the role of the specialist. Referral to a specialist is appropriate if a stage of a speci¤c disease is identi¤ed whose complexity is beyond

What Should Be Taught?

[171]

the capacity of the primary care physician. Suppose, for example, that a patient has adult-onset diabetes mellitus that has remained poorly controlled despite the best efforts of the primary physician; referral is indicated. As another example, a patient with type I (childhood-onset) diabetes has been successfully cared for (excellent control, good cooperation by the patient, and no or wellcontrolled diabetic complications) by the primary care physician for some time, but recently there has been evidence of advancing renal or neurological complications. This patient should also be seen by a specialist. Whether the specialist should continue to care for the patient or provide guidance to the primary care physician depends on the complexity of the problem presented by the disease. In general, specialists should give their opinion to the primary care physician, the patient’s attending physician and, together with the patient, decide who can best care for the patient. The relationship between specialist and primary physician should be ¡exible and consensual enough so that the patient can be seen by both, or remain with the primary physician with help obtained by telephone or other means, or the patient can remain with the specialist until the particular need has passed. The specialist should be teaching the primary physician about the disease, and the primary physician should be teaching about the patient. On the other hand, if primary care physicians are to care for patients with diseases in other specialties and not merely refer on the presumption of a diagnosis, they must be taught and take responsibility for the care of patients who have these diseases.28 There is an etiquette that governs referral and consultation (what used to be called medical ethics) that serves patients and physicians well. It is proper to call a specialist, explain the patient’s problem and provide enough information about the patient to facilitate the consultation. It is often helpful for the referring generalist to make the appointment. After seeing the patient, except in routine matters, the specialist should call the primary care physician and follow up by letter. The relationship between patient and primary physician should be preserved by the specialist. When differences of opinion occur they should be honestly presented, but in such a manner as to leave room for another consultation. Other sources of friction and dif¤culty also exist but they must be handled with the patient’s well-being as the primary consideration. Old texts on this subject (e.g., Percival’s Medical Ethics, cited in note 30) deserve review

[172]

doctoring

because the rules were designed to protect patients even in those very different times. The patient whose disease is well managed, whose other problems are handled satisfactorily, and who is able to talk comfortably with both the specialist and the primary care physician will probably be comfortable with whatever arrangement is reached. Ultimately, if the patient is not happy, a change in the method of care is essential. Nobody except patients themselves can make the ¤nal judgment about whether their needs are being met. Some diseases are so complex that a doctor speci¤cally experienced in their treatment should assume the role of attending physician and be the patient’s primary caregiver, assuming that the consent of both patient and primary physician is given. Examples are late-stage congestive heart failure or chronic obstructive pulmonary disease, very brittle diabetes mellitus, advanced or rare neurological disease or injuries, severe in¡ammatory bowel diseases, and diseases unique in their manifestation and treatment, such as cystic ¤brosis or end-stage renal disease. In such instances, however, the specialists should also attempt to meet their patients’ other simple problems. Patients should not be returned to the primary care physician for colds or other such conditions, as if such minor problems are not worthy of their interest. This suggests a poor relationship with the patient and the primary care physician. The specialist should periodically contact the primary care physician with an update. In treating patients with less complex problems, primary care physicians should have access to the technology commanded by specialists without having to turn the patient over to them. For example, decisions about endoscopy, one of the diagnostic technologies in whose indications for use the primary care physician should be speci¤cally trained, should rest with the primary physician. As always, of course, the person whose hands actually rest on the instrument has the last word. In general, however, when there is disagreement between a specialist and the primary physician, another consultation should be sought. Cancer is a disease whose treatment often exempli¤es both the best and the worst features of specialist referral. It is present practice to refer patients with cancer to an oncologist. Usually, if the patient requires ongoing treatment—radiation therapy or chemotherapy—the oncologist will take over the care. If the patient develops another, nonmalignant disease, he or she may be referred back to the primary

What Should Be Taught?

[173]

physician or, equally likely, to another specialist without the knowledge of the primary care physician. Thus, at a time of real sickness when patients require the care of someone with whom they have developed a relationship, they ¤nd themselves under the care of a stranger or, more commonly, a group of strangers. For the most part, oncologists are not better trained than other physicians in communicating with patients, in the psychosocial dimensions of illness, or in the general care of the sick. In fact, they may be less well-prepared. As the development of the hospice movement shows, they do a poor job of caring for the dying and have only lately begun to provide adequate symptom control. Oncologists should not be faulted for lacking these skills; they are trained in the treatment of cancer. They know what treatment is currently favored, which disease state is best suited for it, and how the treatment is optimally administered. Many physicians who are not oncologists seem afraid of treating patients with cancer. They do not know how to administer chemotherapy, even under an oncologist’s direction, and are no better at communication, psychosocial issues, symptom control, or the care of the dying. For the primary physician, as distinct from oncologists, these shortcomings are reprehensible because the need for these skills can arise in the care of any patient. In general, primary physicians should be able to care for patients with cancer under the direction of and in consultation with oncologists. With access to chemotherapy and radiation units or by administering chemotherapy personally, the primary care physicians should remain the patient’s attending physician. Occasional patients, such as those with hematological cancers requiring bone marrow transplantation, are better transferred to an oncologist, who then becomes the patient’s attending physician. Recent evidence has shown that patients treated in cancer centers have increased survival as compared to those treated in community hospitals. This is also true of Acquired Immunode¤cicncy Syndrome (AIDS).29 Survival, even disease-free survival, is not the only measure of success in treating cancer, but it is certainly an important index that must be matched by treatment in which the primary care physician plays a part. The treatment of cancer deserves special attention as the role of generalists evolves. We will know that we are achieving our goals when patients do better when cared for by generalists with specialist consultation than by specialists alone. The relationship between primary physicians and specialists should be one of cooperation, mutual dependence, and teaching, all

[174]

doctoring

in the service of their patients. It is of interest that the relationship between specialist and attending physician pictured here is similar to that presented in “The Principles of Medical Ethics of the American Medical Association,” adopted in 1912.30 This does not indicate a belief in a return to a fancied past. The patient-care problems of the older medicine arose from a paternalistic view in which patients were seen as children, voiceless, and stripped of their personhood by illness. That view of patients made medical care impersonal—not truly related to the sick person. In the last two decades, the 1950s phrase “Treat the patient as a person” has become a truth. The patient has become a person. Strangely, the high-technology specialist medicine that grew up along with the growth of civil rights and the evolution of the personhood of patients again depersonalized the patient (and the physician) because of its focus on disease states and individual organ systems. The emphasis here on the centrality of the attending physician in relationship to patients and specialists is changed by an understanding of the difference between the modern patient and the patient of the past.

What Program Could Teach All of This? No program can teach all the components of primary care medicine that I have described. There isn’t enough faculty, the teaching facilities are not properly organized, the concepts are underdeveloped, there is insuf¤cient teaching material, the students may ¤nd it too demanding and perhaps even old-fashioned, and the patients may not yet be ready. In fact, no teaching program at any stage of development can teach everything. No matter. Read Horace Davenport’s discussion, in Doctor Dock, of the transformation of the University of Michigan Medical School from the old-style lecture curriculum to the beginnings of a modern school where students learn by taking care of patients.31 Transitions take time. More than anything else, however, they take dedication to an idea. The emergence of a new generalism will produce an exciting amalgam of medical science and dedication to the care of persons, sick and well. A central task for twenty-¤rst-century medicine will be the discovery of the person and ¤nding ways to join the new understandings with medical science’s knowledge of the body. Those of us dedicated to the teaching of clinical medicine must discover the ways in which this new

What Should Be Taught?

[175]

knowledge can instruct the actions of physicians. This book describes the beginnings. It doesn’t matter how long it takes. It isn’t some ultimate goal that counts, it is the path. Consider this story: Sylvia Kalisch is a seventy-eight-year-old businesswoman who is the mother of a family of patients. She has had rheumatoid arthritis for many years. Her son had a small bowel obstruction that took the family’s physician a painfully long time to diagnose and get to surgery. (But the illness made the patient and his wife take stock [with a little help] of a life that had become all business.) Her grandson developed appendicitis just when he was about to go to college. The diagnosis was made on the telephone (because of the family’s con¤dence in the physician) and con¤rmed in the emergency room. Mrs. Kalisch’s daughter has been sturdy and well, but her daughterin-law is the family’s medical expert who calls about nutrition and other issues. Over the ¤fteen years that they have been patients, other problems have arisen, some minor and others more important. Some are purely physical, others are privately social, and still others are socioemotional familial matters. Two years ago in June, Sylvia presented with an acute abdomen that seemed at ¤rst to be appendicitis but was actually a ruptured colonic diverticulum. That September she had upper gastrointestinal bleeding that turned out to be a MALT lymphoma of the stomach. She tolerated her chemotherapy poorly, but multiple follow-up endoscopies and biopsies have shown no evidence of recurrent lymphoma. Last September her children came to see the physician because they were worried about her increasing reclusiveness and debility. In continuing consultation with the oncologist, rheumatologist, and gastroenterologist, but with basic responsibility shifting back to the primary care physician, a program was developed. First, she was transfused; then chronic erythropoetin was started to reverse the severe anemia that had developed from the chronic hypoplastic marrow failure, a result of her chemotherapy. Methotrexate with leucovorin marrow protection (methotrexate alone almost led to disaster) helped bring the rheumatoid arthritis under control, and her steroids were tapered very slowly to zero. A physical therapist visiting her at home helped restore enough muscle in her legs for everyday activities. Dietary intervention (with the help of her daughter-in-law) returned her serum proteins to normal. Her family called frequently during this period. She returned to work part-time. She has signed a health care proxy and a living will, and with her family present, has discussed what she wants done if she loses

[176]

doctoring

decision-making capacity. New York is cold this year, so she is going to stay with a relative in Phoenix, Arizona. An internist there was called, and she was given a referral letter. Recently, her granddaughter visited her primary care physician for a precollege physical examination.

Family members have many diseases, but what understanding of disease would have prepared a physician for all the medical tasks, including the diseases, presented by this family? Think about the kinds of knowledge required for all of these tasks, including integration of the many specialists and other caregivers into a uni¤ed whole. Consider the amount of information that has ¡owed from and to the family and the physician, and entrance of that information into so many small and large judgments. We are training generalists to care for patients such as these. We must accept that patients such as these, all patients from anywhere and everywhere, are going to show us the way to the future.

Epilogue

[177]










Epilogue










I

n 1953, during my third-year surgical clerkship, the chief, Dr. Mulholland,* told us about an operation for intestinal obstruction before the days of long tubes. As he watched, the surgeons tried to stuff yards of balloon-like, in¡ated intestines back into the patient’s abdomen. He said, “I thought, this cannot be what surgery is about.” Why? Because of the grossness, the disregard of the body that the surgery of his youth was already leaving behind as it entered the scienti¤c era. Twenty-¤ve years ago I saw a dying old man, confused, trailing his tubes and intravenous line behind him struggling to get out of bed over the side rails. The aides were yelling at him, pushing on his buttocks, and pulling on his arms to get him back into bed. As I watched, Dr. Mulholland’s words came back. This cannot be what medicine is meant to be when a patient comes to the end of his life— shoved around like a piece of meat. Now we are more than a half a century into the era of the centrality of the person in medicine, and the basic disregard of the patient still exists. Sometimes it takes the form of the awful treatment of that old man, but mostly it consists of

* John Hugh Mulholland, M.D., chairman of the Department of Surgery, New York Medical School from 1928 to 1970.

[177]

[178]

doctoring

diagnostic and therapeutic actions seemingly indifferent to the person within the patient. John Leopold, a seventy-seven-year-old man with long-term asthma, had increasing dysphagia and for the next three months was sent by his primary care physician from one specialist to another, being told about his esophageal cancer by this one, how bad his lungs were by that one, and how his cancer was probably inoperable by yet another. Who was taking care of Mr. Leopold, and where was his doctor? Listen with new ears, open your eyes, and look around you at the medicine of our times. Everywhere you will see the patient subservient to medicine and technology, not the reverse. I know very well that there are many exceptions—some ordinary, some wonderful, some even brilliant. Exceptions, however, not the rule. If you believe I am exaggerating, then read the pathographies of their illnesses written by physicians who had to experience it themselves before they discovered that it is persons who get sick. I am also taking for granted the technical sophistication and advances of the medicine of our times. If you believe that I am complaining that physicians—even Mr. Leopold’s doctors—are not kind or compassionate, you could not be more wrong. In medicine we get the best, brightest, and most motivated students the country has to offer. I do not doubt that our students and physicians-in-training care about their patients. They feel bad when things go wrong. They are upset at the amount of pain they sometimes have to cause. They are often deeply troubled and sad at a bad death. Here is the paradox. The physicians care, but their medicine is uncaring. When students apply to medical school, they are embarrassed to say that they want to be a physician because they want to take care of people. It sounds corny and unsophisticated—and it is, but it is also often true. In their ¤rst year, they cannot wait to have contact with patients. The students are deeply interested, captivated by the patients’ stories of illness and medical care. They are concerned and have solicitude for the well-being of patients they see or hear about. In their anatomy course they approach the cadaver with trepidation; it was a person, and now it is dead. In some schools they are encouraged to remember that the cadaver was once alive and the parent or child of living, grieving humans. At the end of the year in these schools they may have a memorial service for the cadavers. Mercifully for students everywhere, the cadaver soon becomes a depersonalized entity, a bag of body parts of numbing

Epilogue

[179]

complexity and number. It still provokes, however, the grisly humor common to dissecting rooms everywhere, a negative reminder that the cadaver is human. I was a subway student of medicine. My friends and I were happy to have a few bones stick out of our bags while riding the subway to demonstrate our indifference to what frightened others. Equally, we wanted a stethoscope to drape nonchalantly from our pockets so that we would be thought of as doctors. Still, on the occasions when we had contact with patients, that mixture of intense interest in their problems and anxiety for their persons reappeared to belie our studied impersonality. My students now are little different, except for being quicker to blame patients for their illnesses or distressed state. The smoker with cancer of the lung is “more to be censured than pitied.” After all, it was his smoking that did it, and he is responsible for his smoking— with little regard for their naive simplicity. Some don’t believe this patient should be cared for at public expense because he is to blame. In public health courses, at ¤rst the students ¤nd reasons to blame the poor and sick for their unhappy state, and many have sympathy with current attempts to reduce public assistance and medical care. As they learn more, however, they become moved by the plight of the disadvantaged. In the emergency room, for example, when these ¤rst- or second-year students see a very sick, poor patient, that curious mixture of ardent but dispassionate interest and troubled concern emerges again, the latter almost like a contaminant of the former. Learning physical diagnosis requires that students interact with patients. They are apprehensive about intruding on the patients, since they are aware that they contribute nothing to their care. No matter how ¡ustered they may be, however, they want to examine the patients and ask them questions. On the other hand, they are acutely aware of their own inept clumsiness. In the third year, this mixture of con¡icting feelings is even more apparent. The impersonality of the medical scientist vies with the empathy of a physician, the personal interest of an onlooker, and the voyeuristic impulses and sexual curiosity aroused in anybody when secret behaviors and private body parts are revealed. There are endless jokes about how medical students acquire the symptoms of their patients, much as physical therapy students develop symptoms in the same body part as their patients. This suggests involvement with patients at a level deeper than the emotional one— sadness, happiness, gratitude, concern—strong and troublesome as

[180]

doctoring

these emotions may be. Such commitment at several levels of mental life probably underlies the well-known dif¤culties that students encounter when their patients die. Sandra Bertman has explored the reactions of students to death using graphic images drawn by the students and by others. Her studies have revealed the depth and extent of involvement of the students.1 How deeply students are prone to becoming involved with their patients is well known to faculty. From my own observation, and from what I am told by students and house of¤cers, the most common response of faculty is that students must learn to suppress their emotions or “not get involved.” This is bad advice on both counts—the blind leading the blind. Suppressed emotions always ¤nd their way to the surface in some form, and unless directly managed, they may cause a lot of trouble. As for not getting involved, what does this mean? Does it have the same meaning for everyone? Does it produce good doctors? Would anybody work as hard and long, and worry as much as doctors often do, if they weren’t “involved”? Emotion swirls around me all day long. Kathleen Cotten (thirty years old and the mother of two) has just discovered that the sister who raised her when their mother died has Acquired Immunode¤ciency Syndrome (AIDS). She started retching and crying yesterday, and I gave her some medication. She is waiting to discover whether her sister’s baby, twenty-two months old, is positive for Human Immunode¤ciency Virus (HIV). The sister’s baby is HIV positive. John Leopold, described earlier, returned to the hospital with multiple pulmonary emboli after having done well for ten months after esophageal resection. As I was about to see him yesterday to talk about the studies that are showing recurrent tumor, his sister stopped me to say that John’s wife had suddenly died of asthma that morning. Who was going to tell him? Me, of course. I’m his attending physician. Sadly, he died inexplicably two days later. Wolf Josephson came in for a physical examination a few days ago and told me that his partner of twenty-three years (forty-seven years old) had died days earlier. He discovered her body when he came home. Dorothy Cantor died in the intensive care unit a few days ago. Unlike most such cases, her family was able to be with her for the last forty hours, and a morphine drip relieved the pain from her ischemic legs. Good things happened, too. People got better. A biopsy was negative. Other patients are back to work. Someone had a baby. People show me the pictures of their grandchildren, and I admit to

Epilogue

[181]

having some envy. It is impossible to pretend that these strong emotions do not exist. Sometimes I get sad—usually at the death of someone who has been a patient of mine for years. Mostly, I keep my sense of humor and an even keel. One of the reasons that I enjoy caring for the dying is the amount and depth of emotions that surround their care. Most of my colleagues and friends in the hospice movement feel the same way. I know that the dying person is not a member of my family. I acknowledge the emotion in the other person and try to be quiet and just listen—making room for the emotion of the patient’s family members to be expressed. Mostly I use the emotion evoked in me by all of this as the instrument by which I know what is happening in these patients. I do not suppress the emotion; I turn it to use. Medical school graduates do not know how to do this, yet they also live and work in a sea of emotion. These are men and women in their twenties and thirties, a time of life with many emotional strains apart from medicine. They are beginning to discover themselves as individuals—somewhat more slowly than their nonmedical peers because of the consuming intensity of medical education. They struggle for recognition in their training programs while being acutely aware of their inadequacies. They make relationships and break them, marry and have children, worry about money because of their large educational debts, worry about the future and their part in it, and try to establish their reputations. Like everybody else, they have emotional problems that generally remain unresolved. This is not an era that urges psychotherapy, so many who would bene¤t from help do not receive it. Their emotional dif¤culties— the source usually unrecognized—are played out at home and work. They are hard on themselves, their families, and their patients. The medical environment is not very kind to its workers. Physicians appear to be uncomfortable about acknowledging emotional distress in colleagues, and usually little comfort is given to a distressed resident. Communication barriers between attending physicians and house staff exacerbate the problem.2 All of this contributes to the belief that emotion is unwelcome in medicine and should be suppressed. Yet it exists. One evening a ¤rst-year resident stopped by my of¤ce on the way home, barely suppressing her tears until the ¡ood broke through in my consulting room. She had given a patient the wrong intravenous dose of 5-¡uorouracil. The error was of little conse-

[182]

doctoring

quence, but she was inconsolable. Her terror at the ease with which potentially fatal errors could occur and her shame at being unable to admit it to anyone overwhelmed her. Her demeanor was a far cry from her usually contained, competent-appearing self. What physician has not made a more or less serious mistake? When I make a mistake, the ¤rst thing I try to do is show myself that it wasn’t my fault. That doesn’t work, so next I try to push it out of my mind. That’s even less successful. I have learned to tell another doctor as quickly as I can while still trying not to look like a fool. This doctor usually says something like “Don’t worry about it; everyone makes mistakes,” at the same time acting as though he or she would not like to hear anything more. It’s not much consolation, but at least I’ve said it aloud and kept myself from pretending it doesn’t exist. I’ve spent more than forty years with the emotional meaning that goes with living in constant fear of error. I often raise this issue when I lecture or have discussions with other physicians, and it is true of all of us. Why are we haunted by this fear? Certainly not out of concern for malpractice suits. Most errors are never revealed, and much malpractice has little to do with error. The worry arises because of the possibility of hurting persons who put their trust in me—of doing irreparable damage to another human being. Emotion is never far from the surface in both physicians and patients. It arises from the modern de¤nition of medicine by which the patient is the object and subject of medicine and the physician’s immediate focus is the illness, not the disease. You may wonder why I pause to emphasize this so close to the end of this book when it has been an underlying assumption from the beginning. Because this Epilogue is about us, the physicians. It is about our emotions and our passions, not only the patients. Looking back over the years of training and practice, and seeing how often feelings are pushed aside or buried, takes on added meanings. It not only hurts us, causing direct pain and unhappiness, it also robs us of an important dimension in the care of patients. Emotion is present, above all, because physicians are in a relationship with each patient. There is nothing strange about that. We are always in relationships with others. There is no such thing as no relationship. Even strangers may share language, cultural meanings, and rules of behavior, all of which govern how we relate, speak, and behave toward others. Roles—for example, doctor, nurse, mother, child—are always guided by the rules contained within us about the obligations, responsibilities, limitations, and

Epilogue

[183]

rewards associated with the role. Often a role has an overriding ethos. Physicians, for example, are expected to be benevolent, and kind, not to harm their patients, and to have respect for them. Even when these things have been said about relationships, not enough has been said. All relationships exist at all levels of the human condition: physical, emotional (feeling), social, cognitive (intellectual), and spiritual. Physicians, in their relationships with patients, may participate at only one level, or two, or three, or all, knowingly or unknowingly. In fact, a whole culture can rule out a level of relating. The dominant culture of medicine, for example, has blinded itself to the spiritual dimensions of relationships, and even feelings are suspect. Physicians may work at only one level of a relationship in order to hide from the others. The exclusive concentration of some physicians on science and their patients’ diseases, as though feelings were not present, is an example. Sometimes even the ordinary social rules of politeness may be pushed aside—as in any emergency room. When other aspects of the relationship with patients, such as feelings, are brushed aside or pushed below the surface, their potential for providing both information to the physician and an avenue of healing is lost. The physician also loses a source of the rewards and pleasures of clinical medicine. It should be obvious at this point that I believe that physicians, patients, and medicine itself lose a lot when the dimension of emotion in the care of patients is shut out, repressed, stamped on, derogated, or just plain pretended away. When I go further and say that the thesis I am advancing is that learning to love in medicine is a key to knowledge, healing, even wisdom—and by loving I mean not only patients, but also disease, sickness, and even dying—I have far to go to prove my point. Love is the problematic word. To many physicians, the juxtaposition of the words love and medicine is anathema. It smacks of a kind of Bernie Siegel triumphalism or a dangerous romanticism.3 On the other hand, in the Hippocratic Corpus it says, “Love of the art starts with love of humankind.” The word love is clearly being used in a different sense than that of Ella Fitzgerald singing “What Is This Thing Called Love?” The idea of love has been a source of dif¤culty and confusion throughout recorded time. Its meaning has varied from desire and the drive for sexual ful¤llment to the power that Aristotle believed held the planets together. The meanings include the consummated love (eros) of an individual, as well as the love (philia) that is the bond of friendship and the desire for knowledge and wisdom. If love is all of these things

[184]

doctoring

and more, how can we make sense of it in medicine and the care of patients? Further, how can the problem that so frightens people when they consider love in the context of medicine be solved: the confusion of love with desire and of desire with sex—thinking about it, acting on it, or running away from it? The ¤rst step to the solution is to acknowledge the problem that always lurks behind the issue of emotions—whatever they are called—between doctors and patients: the problem of sex and desire. Thoughts with sexual content, sexual references that are usually vague or oblique, sexual feelings, and sexual arousal that is generally brief or transient are extremely common in relation to patients. I am not speaking of sexuality per se, of the physician wanting to have sex with patients—taking advantage of his or her role—apart from the other motivations and emotions in medicine. This has been discussed and condemned from antiquity to the present. The problem of desire in the relationship between patient and doctor, or of desire creating sexual feelings, is not easily set aside. Yesterday, for example, I saw a thirty-eight-year-old woman who had fainted after coming out of her long-soaking, hot après ski bath and struck her side against the tub. Compressing her lower lateral chest wall by placing one hand under her breast and the other in back revealed that she had fractured several ribs. She had already seen her company doctor, who told her the problem was muscular. “He hardly touched me,” she said. “Why didn’t he examine me?” My guess is that he was constrained from touching her handsome breast and thus avoided performing a proper examination. The connection between doctor and patient, whether minimal, as in this instance, or powerful and intimate, as in the care of the dying, brings sexual feelings in its wake. The feelings ¡ow through the bond and may be sensed as much by the patient as by the physician. They may be strong or weak, suppressed, unacknowledged, or manifest, but they are there. I think I am correct in saying that most solutions to this problem consist of denying its existence, remaining silent about it, or acknowledging it and counseling carefulness. This is no longer suf¤cient because the patient has become the primary object of our concern. The need to know the patient to the degree necessary for the kind of primary care medicine described in this book requires that we nurture the bond with the patient as the conduit to the necessary information. David Bernard, chair of the Department of Humanities at Hershey, has shown the omnipresence of intimacy in terminal care. He has made clear the promise for physicians that this intimacy affords. He

Epilogue

[185]

shows that it provides physicians the opportunity to really hear their patients, to break down the walls of loneliness that are the hallmark of suffering, and to bond to the sick and suffering in ways that can even turn sorrow into joy. He has also described how the fear that goes with intimacy—fear of being overwhelmed by it—may keep physicians from experiencing it. Intimacy also carries the constant danger of desire. To deny the desire in any manner stunts the bond, while acting on it sexually destroys it. I think we will get the help we need in understanding this dilemma and its meaning for clinical medicine from Plato. At one point in Plato’s dialogue The Symposium, Socrates says: Well then, [continued Socrates] desiring to secure something to oneself forever may be described as loving something which is not yet to hand. . . . And therefore, whoever feels a want is wanting something which is not yet to hand, and the object of his love and desire is whatever he isn’t, or whatever he hasn’t got—that is to say, whatever he is lacking in.4

Love wants what it does not have, and what it does have it desires to keep. Desire and love cannot be divorced. There may be desire without love, but there is no love without desire. Here we are again with desire and its association with sex. We also desire, however, our morning coffee, a piece of chocolate, or a hot shower. We can want things like these without loving them, in which case a piece of chocolate (say) removes the desire. One might really love chocolate, in which case a piece of chocolate, satisfying as it might be, does not remove the desire. Or one might really love a hot shower—not just the water and cleanliness, but the whole restorative experience. Desire cannot be dissociated from love, but sex can be separated from desire. To love something is to desire it, but not necessarily to want to have sex with it. The issue here is patients, not chocolate or hot showers. When a physician loves a sick person—feels connected or bonded to the patient, even when the bond is as intimate as the one that may occur in the care of the dying—is it really the person who is the object of desire, in the sense that a partner in love is desired or that one desires the love of parents or friends? It is well known that the bond with a patient and sexual desire can become confused in the clinical setting— but it is a confusion with unfortunate consequences. Physicians, from their students days on, want to help their patients. From their earliest years and on into maturity, the behavior

[186]

doctoring

of physicians demonstrates not only an interest in diseases, medical science, and technology, but also a concern with their patients, their losses, and their suffering. When that interest is not manifest, patients and the public call them to task. In order to reach their goals, physicians must connect to their patients. The connection—that powerful bond—is the love of patients. I have discussed the problem of desire, but there is also the danger of being swallowed up or overwhelmed by the relationship. This danger also arises from the intensity of the connection required to know the patient. Here are the alternative hazards of the loving connection with patients. On the one side, there is the peril of succumbing to physical desire; and on the other, the threat of becoming lost in the patients’ pain, swallowed up by their needs and their losses. Both dangers keep the love of patients from its purpose. Drawing back from the bond diminishes the effectiveness of physicians. Yet another problem exists that helps de¤ne what is and is not meant by the love of patients. If the physician’s love of the patient becomes like the love of any person for another outside the role of physician, then the objectivity necessary for clinical action becomes compromised. Another case may clarify this issue. My colleague and I are discussing an older man whom we have been watching grow sicker every day. We are both worried because he looks as if he will soon die. We know there is a collection of pus somewhere, but where? His wife and daughter insistently press us about listening in on our discussion, but I won’t permit it. Why? Aren’t my colleague and I very concerned and uncertain, and aren’t the patient’s wife and daughter also concerned and uncertain? Yes, but the meaning of concern and uncertainty is very different for them and us. The pain of their feelings is born of spousal and ¤lial love. Ours arises from the love of medicine and the love of the patient. It is the love of the patient that binds us to him and his fate and drives our desire to know what is the matter. (As it happened, he had a right subphrenic abcess missed on the ¤rst computer tomography scan.) Without this love, we would merely be interested onlookers. In this circumstance our emotions are not confused with theirs, while in settings where sexual feelings are aroused, confusion is common. This is another situation in which the goal is not obvious. For that reason, more super¤cial attractions or apparent hazards divert physicians from the continuing pursuit of an important human capacity. What is the purpose of the love (philia) of patients? We bond to them in order to help, but not merely by listening or being

Epilogue

[187]

empathetic, although these are important. Physicians must make accurate diagnoses and provide appropriate treatment. It should be clear by now, at the end of this book, that diagnosis is not merely the naming of a disease, nor is treatment only drugs given to treat the disease. Good physicians desire knowledge and desire to make the patient better. They are seeking the information necessary to make the patient whole, even in the face of death. Knowledge ¡ows in one direction through the loving connection so that physicians can know what is the matter and in the other direction for accomplishing therapeutic aims. Physicians also desire the power that arises from knowledge and the ability to heal. In Chapter 5, I discussed the kinds of knowledge necessary to know the person. This is knowledge born of the loving connection. It is about the person and about the person’s sickness. Some of it is transmitted by the hands of the examining physician, some of it through skilled listening. Some of it is subjective—a feeling or an intuition—but all of it is enhanced and given added dimension by the connection to the patient. On the therapeutic side, the bond permits effective tuning or shaping of physicians’ technical medical actions to the patient. This kind of knowing through the agency of love distinguishes professional caregivers from those who only care. It differentiates care based on good intentions or unlettered compassion from that grounded in the physicians’ love of patients. There is something else. Two hundred years ago the Enlightenment, in defeating the smothering effects of received knowledge and religion, gave birth to modern science and brought it to prominence. The central fact of that era, however, was not the astounding knowledge of nature that began to accumulate, wonderful as that was, but recognition of the power of reason and inference about the objective world of nature. Not just new knowledge, but a new function of the mind became commonplace. Here, also, the fundamental fact is not the knowledge of humankind opened up by love and its connection to patients. Instead, it is the function of the mind that might become commonplace in medicine. This is the capacity that opens physicians up to the subjective knowledge of others and allows them to think about it, make it manifest, and act on it. Only then can this kind of knowing join with medicine’s knowledge of science in the care of patients. Another dimension would thus be added to the Hippocratic insight that the love of the art starts with the love of humankind.

This page intentionally left blank










Notes








Acknowledgments 1. Cassell, Eric J. Teaching the fundamentals of primary care: A point of view. The Milbank Memorial Quarterly. 73:373–405. 1995.

Introduction 1. See, for example, Katherine Riley, Wayne Myers, Michael J. Gordon, et al., A collaborative approach to primary care preclinical preceptorship for underserved areas. Academic Medicine 66:776–77, 1991; Paul R. Young, Residency training for rural primary care. Academic Medicine 65(12 Suppl):S25–7, 1990; Allan Roberts, Robert Foster, Margaret Dennis, Laurence Davis, et al., An approach to training and retaining primary care physicians in rural Appalachia. Academic Medicine 68:122–25, 1993; Allan Roberts, Laurence Davis, and James Wells, Where physicians practicing in Appalachia in 1978 to 1990 were trained and how they were distributed in urban and rural Appalachia. Academic Medicine 66:682–86, 1991; Nancy S. Jecker and Alfred O. Berg, Allocating medical resources in rural America: Alternating perceptions of justice. Social Science in Medicine 34:467–74, 1992; Charles T. McElmurray, D. Lindsie Cone, Sandra K. Kammerman, and Stanley D. Fowler, The Winnsboro Rural Primary Care Education Project: University of South Carolina School of Medicine. Journal of the South Carolina Medical Association Oct: 493–95, 1992; Marc L. Rivo, Tim M. Henederson, and Debbie M. Jackson, State legislative strategies to improve the supply and distribution of generalist physicians, 1982 to 1992. American Journal of Public Health 8 5:405–7, 1995; and Robert A. Conner, Steven D. Hillson, John E. Krawelski, Competition, professional synergism, and the geographic distribution of rural physicians. Medical Care 33:1067–78, 1995. 2. G. Gayle Stephens, The Intellectual Basis of Family Practice. Tucson, AZ: Winter Publishing Co., 1981, p. 8. 3. Eric J. Cassell, The nature of suffering and the goals of medicine. New England Journal of Medicine 306:639–45, 1982.

[189]

[190]

Notes to Chapter One

4. See Ian R. McWhinney, A Textbook of Family Medicine. New York: Oxford University Press, 1989; and Carl Edvard Rudebeck, General practice and the dialogue of clinical practice. Scandinavian Journal of Primary Health Care 1(Suppl): 1–85, 1991. 5. Kent J. Sheets, William A. Anderson, and Patrick C. Alguire, Annotated bibliography: Curriculum development and evaluation in medical education. Journal of General Internal Medicine 7:538–43, 1992. 6. Carol P. Tresolini, Daniel A. Shugars, and Linda S. Lee, Teaching an integrated approach to health care: Lessons from ¤ve schools. Academic Medicine 70:665–70, 1995. 7. Rudebeck, op. cit. 8. See W. T. Branch, Teaching models in an ambulatory training program. Journal of General Internal Medicine 5:S15–26, 1990; Annette M. Yonke and Richard P. Foley, Overview of recent literature on undergraduate ambulatory care education and a framework for future planning. Academic Medicine 66:750–55, 1991; Dorothy S. Lane, Meeting an expanding educational role for ambulatory care services. American Journal of Preventive Medicine 4:35–40, 1988; Susan F. Skochelak and Thomas C. Jackson, An interdisciplinary clerkship model for teaching primary care. Academic Medicine 67:639–41, 1992; Gabriel Smilkstein, Designing a curriculum for training community responsive physicians. Journal of Health Care for the Poor and Underserved 1:237– 42, 1990; John T. Philbrick, Julia F. Connelly, Eugene C. Corbett, Jr., et al., Restoring balance to internal medicine training: The case for the teaching of¤ce practice. American Journal of the Medical Sciences 299:43–49, 1990; Lesley Recs and John Wass, Undergraduate medical education. British Medical Journal 306:358–61, 1993; Ronald W. Richards and Rebecca C. Henry, Community partnerships: Educational linkages to increase the number of primary care practitioners. Academic Medicine 68:594–96, 1993; John E. Verby, J. Paul Newell, Susan A. Andresen, and Walter M. Swentko, Changing the medical school curriculum to improve patient access to primary care. Journal of the American Medical Association 266:110–13, 1991; G. T. Perkoff, Teaching clinical medicine in an ambulatory setting: An idea whose time has come. New England Journal of Medicine 314:27–31, 1986; and J. O. Wooliscroft and T. L. Shwenk, Teaching and learning in the ambulatory setting. Academic Medicine 66:750–55, 1989. 9. Jane L. Murray, Steven A. Wartman, and August G. Swanson, A national, interdisciplinary consortium of primary care organizations to promote the education of generalist physicians. Academic Medicine 67:8–11, 1992; Charles S. Bryan, General internal medicine as a 21st century specialty: Perspectives of community-based chairs of medicine. American Journal of Medicine 99:1–3, 1995; and Wylie Burke, James F. Wallace, and Paul G. Ramsey, A fourth year of training in ambulatory medicine: The University of Washington experience. American Journal of Medicine 96:463–68, 1994. 10. Eric J. Cassell, The Nature of Suffering. New York: Oxford University Press, 1991, chapter 7. 11. Ibid., chapter 5.

Chapter One 1. Quoted from Barbara Star¤eld, Primary Care: Concept, Education, and Policy. New York: Oxford University Press, 1992, p. 5. 2. Daniel M. Fox, Power and Illness: The Failure and Future of American Health Policy. Berkeley, CA: University of California Press, 1993, pp. 32ff. 3. Saad Z. Nagi, Disability concepts revisited: Implications for prevention. Appendix A. In Disability in America: Toward a National Agenda for Prevention, ed. by Andrew M. Pope and Alvin Tarlov. Washington, DC: National Academy Press., 1991.

Notes to Chapter One

[191]

4. Eric J. Cassell, The Healer’s Art. Cambridge, MA: MIT Press, 1985. 5. Walsh McDermott, Kurt Deuschle, and C. Barnett, Health care experiment at Many Farms. Science 175:23–31, 1972. 6. Mary O. Mundinger, Advanced–practice nursing—good medicine for physicians. New England Journal of Medicine 330:211–14, 1994. 7. Health Security Act. Title III, Subtitle A, Section 3012 (e) (3), 1993, p. 504. 8. Barbara Star¤eld, Primary Care: Concept, Evaluation, and Policy. New York: Oxford University Press, 1992; Aaron Manson, Why primary care physicians should not be restrictive gatekeepers. Journal of General Internal Medicine 10:145–6, 1995. 9. Eric J. Cassell, The Nature of Suffering and the Goals of Medicine. New York: Oxford University Press, 1991, chapter 7; Charles S. Bryan, General internal medicine as a 21st century specialty: Perspectives of community-based chairs of medicine. American Journal of Medicine 99:1–3, 1995; Wylie Burke, James F. Wallace, and Paul G. Ramsey, A fourth year of training in ambulatory medicine: The University of Washington experience. American Journal of Medicine 96:463–68, 1994. 10. See Salvatore Mangione, Linda Z. Nieman, Edward Gracely, and Donald Kaye, The teaching and practice of cardiac auscultation during internal medicine and cardiology training. Annals of Internal Medicine 119:47–54, 1993; Jay A. Block, A problem in academic internal medicine: We have taken the teaching out of the teaching hospital. The Pharos Spring 55:28–30, 1992; and Debra L. Roter, Karan A. Cole, David E. Kern, et al., An evaluation of residency training in interviewing skills and the psychosocial domain of medical practice. Journal of General Internal Medicine 5:347–54, 1990. 11. Howard S. Barrows and Gar¤eld C. Pickell, Developing Clinical Problem Solving Skills. New York: W.W. Norton, 1991. 12. See Dawn P. Lemcke, Julie Pattison, Lorna A. Marshall and Deborah S. Cowley, eds., Primary Care of Women. Norwalk, CT: Appleton and Lang, 1995; and Karen J. Carlson and Stephanie A. Eisdenstat, eds., Primary Care of Women. St. Louis: C.V. Mosby, 1995. 13. Robert D. Robertson, Reversing the decline and fall of the general internist. Indiana Medicine 86:128–30, 1993. 14. See Wylie Burke and Thomas S. Inui, Do we still need primary care tracks? Annals of Internal Medicine 116:1065–70, 1992; and Arthur M. Fournier, Mark Gelbard, and Laurence B. Gardner, Uni¤ed general internal training programs. Journal of General Internal Medicine 5:166–69, 1990. 15. John D. Stoekle and John D. Goodson, General internal medicine. Journal of the American Medical Association 265:3123, 1991. 16. Jay D. Orlander and Chistopher M. Callahan, Fellowship training in academic general internal medicine. Journal of General Internal Medicine 6:460–65, 1991. 17. See Federated Council for Internal Medicine, General internal medicine and general internists: Recognizing a national need. Annals of Internal Medicine 117:778– 79, 1992; Gerry N. Boccarossa and Susan G. Boccarossa, The proper de¤nition of a general internist. New England Journal of Medicine 119:440–41, 1993; and Robert G. Petersdorf, The undergraduate medical curriculum: Adequate preparation for a career in internal medicine? Journal of General Internal Medicine 7:213–16, 1992. 18. American College of Physicians, The role of the future general internist de¤ned. Annals of Internal Medicine 121:616–22, 1994; Harry R. Kimball and Paul R. Young, A statement on the generalist physician from American Boards of Family Practice and Internal Medicine. Journal American Medical Association 271:315–16, 1994. 19. See Robert G. Petersdorf, Financing medical education. New England Journal of Medicine 328:651–6o, 1993; Jerome P. Kassier, Primary care and the af¡iction of internal medicine. New England Journal of Medicine 328:648–50, 1993; Faith T. Fitzgerald, The case for internal medicine. New England Journal of Medicine 328:654–56, 1993;

[192]

Notes to Chapter Three

Norman C. Levinsky, Recruiting for primary care. New England Journal of Medicine 328:656–60, 1993; Jack H. Geiger; Why don’t medical students choose primary care? American Journal of Public Health 83:315–16, 1993; and Donald G. Kassenbaum, Philip L. Szenas, and Michael K. Schuchert, Determinants of the generalist career intentions of 1995 graduating medical students. Academic Medicine 71:198–209, 1996.

Chapter Two 1. Jerome P. Kassierer, The next transformation in the delivery of health care. New England Journal of Medicine 332:52–53, 1995. 2. Lucien L. Leape, Error in medicine Journal of the American Medical Association 272:1851–57, 1994. 3. George Engel, The need for a new medical model: A challenge for biomedicine. Science 196:129–36, 1977. 4. Stephen Toulmin, Cosmopolis: The Hidden Agenda of Modernity. Chicago; University of Chicago Press, 1990, pp. 30ff. 5. Deborah R. Gordon, Tenacious assumptions in western medicine. In Biomedicine Examined, ed. by Margaret Lock and Deborah Gordon. Boston: Kluwer Academic Publishers, 1988. 6. David Bidney, On the philosophical anthropology of Ernst Cassirer and its relation to the history of anthropological thought. In The Philosophy of Ernst Cassirer, ed. by Paul Arthur Schlipp. LaSalle, IL: Open Court Publishing Co., 1949, p. 491. 7. Alfred North Whitehead, Adventures of Ideas. New York: Free Press, 1933, p. 208. 8. H. Tristram Engelhardt, Doctor the disease, treating the complaint, helping the patient: Some of the works of Hygeia and Panacca. In Knowing and Valuing: The Search for Common Roots, Volume IV of The Foundations of Ethics and Its Relationship to Science, ed. by H. Tristram Engelhardt, Jr. and Daniel Callahan. Hastings-on-Hudson, NY: Institute of Society Ethics and the Life Sciences (The Hastings Center), 1980. 9. Robert Hunt Sprinkle, Profession of Conscience. Princeton, NJ; Princeton University Press, 1994, pp. 41 and 114. 10. Alfred North Whitehead, Symbolism: Its Meaning and Effect. New York: Macmillan Publishing Co., 1927. 11. Ernst Cassirer. The Philosophy of Symbolic Forms. Vol. 3: The Phenomenology of Knowledge. New Haven, CT: Yale University Press, 1957. 12. Arthur Kleinman, Illness Narratives: Suffering, Healing, and the Human Condition. New York: Basic Books, 1988.

Chapter Three 1. See Stanley J. Reiser, Technology and the eclipse of individualism in medicine. In In Search of the Modern Hippocrates, ed. by Roger J. Bulger. Iowa City: University of Iowa Press, 1987, pp. 212–21; and Eric J. Cassell, The sorcerer’s broom: Medicine’s rampant technology. Hastings Center Report 23:32–39, 1993. 2. See Neal G. Uren, Jacque A. Melin, Bernard De Bruyne, et al., Relation between myocardial blood ¡ow and the severity of coronary artery stenosis. New England Journal of Medicine 330:1782–88, 1994; C. W. White, C. B. Wright, D. B. Doty, et al., Does visual interpretation of the coronary arteriogram predict the physiologic importance of coronary stenosis? New England Journal of Medicine 310: 819–24, 1984; and Glenn N. Levine, John F. Keaney Jr., and Joseph A. Vita, Medical progress: Cholesterol reduction in cardiovascular disease. New England Journal of Medicine 332:512–21, 1995.

Notes to Chapter Five

[193]

3. See Charles B. Treasure, J. Larry Klein, William Weintraub, et al., Bene¤cial effects of cholesterol-lowering therapy on the coronary endothelium in patients with coronary artery disease. New England Journal of Medicine 332:481–87, 1995; and Todd J. Anderso, Ian T. Meredith, Alan C. Yeung, et al., The effect of cholesterol-lowering and antioxidant therapy on endothelium-dependent coronary vasomotion. New England Journal of Medicine 332:488–93, 1995. 4. Renee C. Fox, Training for uncertainty. In The Student Physician: Introductory Studies in the Sociology of Medical Education, ed. by Robert K. Merton, et al. Cambridge, MA: Harvard University Press, 1957, pp. 207–41. 5. Samuel Gorovitz and Alisdair MacIntyre, Toward a theory of medical fallibility, The Hastings Center Report 5:13–23, 1975. 6. Kay Toombs, The Meaning of Illness. Boston: Kluwer Academic Publishers, 1992, chapter 2.

Chapter Four 1. Sir William Osler, quoted in Pathophysiology: The Biological Principles of Disease, ed. by Lloyd H. Smith and Samuel O. Thier. Philadelphia: W. B. Saunders, Co., 1985, p. xi. 2. John Locke, quoted in The Encyclopaedia Britannica, article on medicine by Sir Thomas Clifford Allbutt, Vol. 18, 1911, p. 50. 3. Arthur Elstein and George Bordage, Psychology of clinical reasoning. In Professional Judgement, ed. by Jack Dowie and Arthur Elstein. New York: Cambridge University Press, 1988, pp. 109–29. 4. Eric J. Cassell, Talking with Patients. Vol. II: Clinical Technique. Cambridge, MA: MIT Press, 1985, chapter 1. 5. Maureen C. Jensen, Michael N. Brant-Zawadzki, Nancy Obuchowski, et al., Magnetic resonance imaging of the lumbar spine in people without back pain. New England Journal of Medicine 331:69–73, 1994. 6. Eric J. Cassell, Uses of the subjective in medical practice. In Changing Values in Medicine, ed. by Eric J. Cassell and Mark Siegler. University Publication of America, 1979, pp. 151–66. 7. Aristotle, Nichomachean Ethics. Book VI. 1143a 10–19. 8. Frank Davidoff, Kathleen Case, and Pamela W. Fried, Evidence-based medicine: Why all the fuss? Annals of Internal Medicine 122:727, 1995. 9. Alvan Feinstein, Clinical Judgment. Baltimore: Williams and Wilkins, 1967. 10. Hans Georg Gadamer, Truth and Method. New York: Continuum Publishing Co., 1994, pp. 13ff. 11. Hannah Arendt, Crises in Culture. New York: Harcourt Brace Jovanovich, 1969, pp. 220–21; quoted in Berstein, op. cit., p. 218.

Chapter Five 1. Eric J. Cassell, The Nature of Suffering. New York: Oxford University Press, 1991, chapter 5. 2. David A. Landis, Physician distinguish thyself: Con¡ict and covenant in a physician’s moral development. Prospectives in Biology and Medicine 36:628–41, 1993. 3. Mary Jean Huntington, The development of a professional self-image. In The Student-Physician, ed. by Robert K. Merton, George Reader, and Patricia Kendall. Cambridge, MA: Harvard University Press, 1957, pp. 179–87.

[194]

Notes to Chapter Six

4. Eric J. Cassell, Talking with Patients. Vol. I: The Theory of Doctor–Patient Communication. Cambridge, MA: MIT Press, chapter 3, 1985. 5. James Hillman, Emotion. Evanston, IL: Northwestern University Press, 1962. 6. See M. Balint, The Doctor and His Patient and the Illness. London: Pittman, 1957; and M. Balint, E. Balint, R. Yosling, and P. A. Hildebrand, Study of Doctors. London: Tavistock, 1966. 7. Richard V. Lee, The clinical picture, Journal of Clinical Epidemiology 43:527–31, 1990. 8. Benedetto Croce, Aesthetic as the Science of Expression and General Linguistic. New York: Noonday Press, 1958. 9. Rita Charon, Joanne Trautman Banks, Julia E. Connolly, Anne Hunsaker Hawkins, Kathryn Montgomery Hunter, Anne Hudson Jones, Martha Montello, and Suzanne Poirer, Literature and medicine: Contributions to clinical practice. Annals of Internal Medicine 122: 599–606, 1995. 10. Frank Davidoff, The future of the Annals. Annals of Internal Medicine 122:37–76, 1995.

Chapter Six 1. Martin Heiddeger, Hegel’s Phenomenology of Spirit. Trans. by Parvis Eman and Kenneth Maly. Bloomington: Indiana University Press, 1988, p. 20. 2. See, for example, Susan E. Skockelak and Thomas C. Jackson, An interdisciplinary clerkship model for teaching primary care. Academic Medicine 67:639–41, 1992; Gabriel Smilkstein, Designing a curriculum for training community responsive physicians. Journal of Health Care for the Poor and Underserved 1:237–42, 1990; John T. Philbrick, Julia E. Connelly, Eugene C. Corbett, Jr., et al., Restoring balance to internal medicine training: The case for the teaching of¤ce practice. American Journal of the Medical Sciences 299:43–49, 1990; Lesley Rees and John Wass, Undergraduate medical education. British Medical Journal 306:358–61, 1993; Ronald W. Richards and Rebecca C. Henry, Community parternships: Educational linkages to increase the number of primary care practitioners. Academic Medicine 68:594–96, 1993; and John E. Verby, J. Paul Newell, Susan A. Andresen, and Walter M. Swentko, Changing the medical school curriculum to improve patient access to primary care. Journal of the American Medical Association 266:110–13, 1991. 3. Jane L. Murray, Steven A. Wartman, and August G. Swanson, A national interdisciplinary consortium of primary care organizations to promote the education of general physicians. Academic Medicine 67:8–11, 1992. 4. Kenneth S. Ogle and James D. Plumb, The role of the primary care physician in the care of the terminally ill. Clinical Geriatric Medicine 12:267–78, 1996. 5. Robert M. Wachter and Lee Goldman, The emerging role of “hospitalists” in the American health care system. New England Journal of Medicine 335:514–16, 1996; Jerome P. Kassirer; Redesigning graduate medical education—location and content. New England Journal of Medicine 335:507–8, 1996. 6. Herb Lamberts and Inge Hofmans-Okkes, Values and roles in primary care. Journal of Family Practice 42:178–80, 1996. 7. Lucien L. Leape, Error in medicine. Journal of the American Medical Association 272:1851–57, 1994. 8. William Branch, Notes of a small-group teacher. Journal of General Internal Medicine 6:573–78, 1991. 9. Maria Callas, Maria Callas at Juilliard: The Masterclasses. EMI Classics ZDMC 7243 5 65802 2. Three CDs.

Notes to Chapter Seven

[195]

10. Irwin J. Schatz, Janet P. Realini, and Evan Charney, Family practice, internal medicine, and pediatrics as partners in the education of generalists. Academic Medicine 71:35–39, 1996. 11. Constance D. Baldwin, Harold G. Levine, and David P. McCormick, Meeting the faculty development needs of generalist physicians in academia. Academic Medicine 70(1 Suppl):s97–103, 1995. 12. Maurice Lemon, Annette Yonke, Bonnie Roe, and Richard Foley, Communication as an essential part of program and institutional development. Academic Medicine 70:884–86, 1995. 13. See Ludwig W. Eichna, 1975–1979: A student’s perspective. New England Journal of Medicine 303:727–34, 1980; A medical school curriculum for the 1980s. New England Journal of Medicine 308:18–21, 1983; and A medical school education for whom student or patient? Bulletin of the New York Academy of Medicine 67:151–61, 1991. 14. Horace W. Davenport, Doctor Dock: Teaching and Learning Medicine at the Turn of the Century. New Brunswick, NJ: Rutgers University Press, 1987. 15. Donald G. Kassenbaum, Philip L. Szenas, and Michael K. Schuchert, Determinants of the generalist career intentions of 1995 graduating medical students. Academic Medicine 71:198–209, 1996. 16. See, for example, John K. Iglehart, Health care reform and graduate medical education. New England Journal of Medicine 330:1167–71, 1994; and Fitzhugh Mullan, Robert M. Politzer, Sandy Gamliel, and Marc L. Rivo, Balance and limits: Modeling graduate medical education reform based on recommendations of the Council on Graduate Medical Education. The Milbank Quarterly 72:385–98, 1994; Marc L. Rivo, Internal medicine and the journey to medical generalism. Annals of Internal Medicine 120:92, 1994. 17. Sharon Kaufman, The Healer’s Tale: Transforming Medicine and Culture. Madison: University of Wisconsin Press, 1993.

Chapter Seven 1. R. G. Collingwood, An Autobiography. New York: Clarendon Press–Oxford University Press, 1978, p. 320. 2. Mark Siegler, The nature and limits of clinical medicine. In Changing Values in Medicine, ed. by Eric J. Cassell and Mark Siegler. University Publications of America, 1979, pp. 19–41. 3. See, for example, S. A. Cohen-Cole, The Medical Interview: The Three Function Approach, St. Louis: Mosby Year Book, 1991; M. Lipkin, S. M. Putnam, and A. Lazare, eds., The Medical Interview: Clinical Care, Education, and Research. Springer-Verlag, 1995; D. L. Roter and J. A. Hall, Doctors Talking with Patients, Patients Talking with Doctors: Improving Communication in Medical Visits. Westport, CT: Auburn House, 1992; Andrew J. Billings and John D. Stoeckle, The Clinical Encounter: A Guide to the Medical Interview and Case Presentation. Chicago: Year Book Publishers, 1989; Moira Stewart and Debra Roter, eds., Communicating with Medical Patients. Newbury Park, CA: Sage Publications, 1989; and John L. Coulehan and Marion R. Block, The Medical Interview: A Primer for Students of the Art, 2nd ed. Philadelphia: F. A. Davis Co., 1992. 4. Eric L. Cassell, Talking with Patients. Vol. I: The Theory of Doctor–Patient Communication. Cambridge, MA: MIT Press, 1985. 5. Eric L. Cassell, The Nature of Suffering and the Goals of Medicine. New York: Oxford University Press, 1991, chapter 5. 6. Katherine Hunter, Doctors’ Stories: The Narrative Structure of Medical Knowledge. Princeton, NJ: Princeton University Press, 1991.

[196]

Notes to Chapter Seven

7. Hans Georg Gadamer, Truth and Method, 2nd ed. New York: Continuum Books, 1994, p. 310. 8. See Alvan Feinstein, Clinical Judgement. Baltimore: William & Wilkins, 1967; and Clinical Epidemiology: The Architecture of Clinical Research. New York: W. B. Saunders Co., 1985. 9. See, for example, Robert H. Fletcher, Suzanne W. Fletcher, and Edward H. Wagner, Clinical Epidemiology: The Essentials, 2nd ed. Baltimore: Williams & Wilkins, 1988; David L. Sackett and Peter Tugwell, Clinical Epidemiology: A Basic Science for Clinical Medicine, 2nd ed. Boston: Little, Brown, & Co., 1991; and Henrik R. Wulff, Rational Diagnosis and Treatment: An Introduction to Clinical Decision-Making. Boston: Blackwell Scienti¤c Publications, 1981. 10. Collingwood, pp. 102ff. 11. Clement J. McDonald, Medical heuristics: The silent adjudicators of clinical practice. Annals of Internal Medicine 124:56–62, 1996. 12. Marsden S. Blois, Information and Medicine: The Nature of Medical Descriptions. Berkeley: University of California Press, 1984. 13. Alexander Tomasz, Multiple-antibiotic-resistent pathogenic bacteria: A report on the Rockefeller University Workshop. New England Journal of Medicine 330:1247– 51, 1994. 14. Eric J. Cassell, The abuse of the elderly: Abuses of power. New York State Journal of Medicine 89:159–62, 1989. 15. Daniel Callahan, Setting Limits. New York: Simon & Schuster, 1987. 16. See Dorothy S. Lane, Developing primary care curricula in preventive medicine: Some practical considerations. American Journal of Preventive Medicine 8:389– 94, 1992; Cheryl L. Albright, John W. Farquhar, Stephen P. Portman, et al., Impact of a clinical preventive medicine curriculum for primary care faculty: Results of a dissemination model. Preventive Medicine 21:419–35, 1992; Merwyn R. Greenlick, Educating physicians for population-based clinical practice. Journal of the American Medical Association 267:1645–48, 1992; and Richard C. Wender, Cancer screening and prevention in primary care: Obstacles for physicians. Cancer 72 (3 Suppl): 1093–99, 1993. 17. Louise B. Russell, Educated Guess: Making Policy about Medical Screening Tests. Berkeley: University of California Press, 1994. 18. U.S. Preventive Services Task Force. Guide to Clinical Preventive Services, 2nd ed. Baltimore: Williams & Wilkins, 1996, p. xxvi; Henry Wechsler, Sol Levine, Roberta K. Idelson, Edward Schor, and Eugenie Coakley, The physician’s role in health promotion revisited—a survey of primary care practitioners. New England Journal of Medicine 334:996–98, 1996. 19. E. S. Higgins, A review of unrecognized illness in family care: Prevalence, natural history, and efforts to change the course. Archives of Family Medicine 3:908–17, 1994; Edward J. Shahady, Obsessive compulsive disorder in primary care. Journal of Clinical Psychiatry 55 Suppl: 79–82, 1994. 20. See W. E. Broadhead, D. G. Blazer, L. K. George, and C-J. K. J. Tse, Depression disability days, and days lost from work in a prospective epidemiologic survey. Journal of the American Medical Association 262:914–19, 1990; W. Coryell, W. Scheftner, M. Keller, et al. The enduring psychosocial consequences of mania and depression. American Journal of Psychiatry 150: 720–27, 1993; and K. B. Wells, A. Stewart, and R. D. Hays, The functioning and well-being of depressed patients: Results of the Medical Outcome Study. Journal of the American Medical Association 262:914–19. 21. Douglas B. Kamerow, Harold Alan Pincus, and Donald Ian McDonald, Journal of the American Medical Association 255:2054–57. 22. M. M. Weissman, M. Olfson, A. C. Leon et al. Brief diagnostic interviews (SDDS-PC) for multiple mental disorders in primary care. A pilot study. Archives of

Notes to Epilogue

[197]

Family Medicine 4:220–27, 1995; W. A. Apfeldorf, M. K. Shear, A. C. Leon, and L. Portera, A brief screen for panic disorder. Journal of Anxiety Disorders 8:71–78, 1995. 23. Kathleen S. France, Consultation-liason elective for primary care residents. General Hospital Psychiatry 15:171–76, 1993. 24. Herbert N. Brown and Norman E. Zinberg, Dif¤culties in the integration of psychological and medical practices. American Journal of Psychiatry 139:1576–80, 1982; and Michael Gropper, Family medicine and psychosocial knowledge: How many hats can the family doctor wear? Social Science in Medicine 25:1249–55, 1987. 25. See Harold Alan Pincus, Linking general health and mental health systems of care: Conceptual models of implementation. American Journal of Psychiatry 137:315–20, 1980; Harold Alan Pincus, James J. Strain, Jeffry L. Houpt, et al. Models of mental health training in primary care. Journal of the American Medical Association 249:3065–69, 1983; and ibid., Patient-oriented models for linking primary care and mental health care. General Hospital Psychiatry 9:95–101, 1987; and Kathleen S. N. Franco, Consultation-liason elective for primary care residents. General Hospital Psychiatry 15:171–76, 1993. 26. R. Williams and S. L. Foldy, The state of community-oriented primary care: Physician and residency program surveys. Family Medicine 26:232–37, 1994. 27. Ott0 Guttentag, The attending physician as the central ¤gure. In Changing Values in Medicine, ed. by E. J. Cassell and Mark Siegler. University Publications of America, 1979, pp. 107–260. 28. See Janet B. Henrich, Daniel W. Rahn, and Nicholas H. Feihack, Integrating general medicine and rheumatology training in the outpatient setting: A practice model. Journal of General Internal Medicine 7:43–46, 1992; and Barbara Rochen Renner, Brenda M. DeVellis, Susan T. Ennett, et al., Clinical rheumatology training of primary care physicians: The resident’s prespective. Journal of Rheumatology 17:666–72, 1990; Linda L. Emanuel and James Richter, The consultant and the patient–physician relation. A trilateral deliberative model. Archives of Internal Medicine 154:1785–90, 1994; J. A. Horiszny, Specialists and generalists: a platonic relationship. Pharos 57:14–18, 1994. 29. Mari M. Kitahata, Thomas D. Keopsell, Richard A. Deyo, Clare L. Maxwell, Wayne T. Dodge, and Edward H. Wagner, Physician’s experience with the acquired immunode¤ciency syndrome as a factor in patient’s survival. New England Journal of Medicine 334:701–6, 1996; Stephanie L. McFall, Richard B. Warnecke, Arnold D. Kauzny, Marlene Aitken, and Leslie Ford, Physician and practice characteristics associated with judgments about breast cancer treatment. Medical Care 32:106–17, 1994; E. Grunfeld, D. Mant, M. P. Vessey, and P. Yudkin, Evaluating primary care followup of breast cancer: Methods and preliminary results of three studies. Annals of Oncology 6 Suppl 2:47–52, 1995. 30. In Percival’s Medical Ethics, ed. by Chauncey D. Leake. Huntington, NY: Robert E. Krieger Publishing Co., Appendix V, 1975, pp. 259ff. 31. Horace W. Davenport, Doctor Dock. New Brunswick, NJ: Rutgers University Press, 1987, pp. 1–25.

Epilogue 1. Sandra Bertman, Facing Death: Images, Insights, and Interventions. New York: Hemisphere Publishing Corp., 1991, pp. 120ff. 2. J. D. McCue and K. J. Beach, Communication barriers between attending physicians and residents. Journal of General Internal Medicine 9:158–61, 1994. 3. Bernard Siegel, Love, Medicine, and Miracles. New York: Harper-Collins Publishers, 1990. 4. Plato, The Symposium, 200d.

This page intentionally left blank










Index








Acute disease centrality of physician in, 25 chronic disease distinction, 22– 27, 37 course of, 23 infectious disease model of, 37 medical view of, 22–23 specialism model, 30–31 Aesthetics, 119–21 Affect. See Emotion Aged patients, 164–65 AIDS, 173 Ambulatory settings place of training, 11, 128–29, 132 and teaching of values, 132 Analgesics, 163 Anatomy training, 160–62 Anecdotes, 122 Angioplasty, 63 Antibiotics, 163 Aristotle, 100

Atomistic assumptions, 49–50, 60 Attending physician, 170–74 Audiotapes, 133, 149–50 Automated blood chemistry, 72 Automatic skills, 134, 157–58 Back pain, 56–58 disease model limitations, 58 and MRI disk abnormalities, 96 technology in evaluation of, 56, 71–72 Balint, Michael, 118 Bedside training, 60–61 Beliefs, 115 Best interest of the patient best interest of the doctor con¡icts, 32–33 clinical wisdom role, 100 and technology, 79 Bias, 45, 87–90, 104 Biopsychosocial model, 6, 49 Body-as-nature, 10

[199]

[200}

Index

Cancer care, 172–73 Career path, 141–43 Chemistry panels, 161 Chest pain, 70–71 Chest X-ray, 161 Chief complaint, 53 Cholecystectomy, 73 Chronic disease, 22–27 acute disease distinction, 22–27 continuum of care, 37–38 disability role, 22–23 education and training implications, 164 impact on person, 23–24 infectious disease model limitations, 37 institutional care inadequacies, 30–31 patient-centered model, 51–53 case examples, 53–58 personal focus of, 25–26 primary care implications, 25– 27 variable expression of, 24 Classroom instruction, 132–33 Clinical empiricism, 59–61 centrality of patient in, 59–61 in clinical method, 87–88 educational requirements, 60– 61 observational skills in, 59–60 Clinical epidemiology and clinical judgment, 156–57 instruction in, 156–57 and postgraduate curricula, 139 Clinical interview. See also History-taking language in, 149–50 in postgraduate curriculum, 139 teaching of, 148–50 Clinical judgment, 99–104 medical practice role, 103 practical wisdom in, 100

subjectivity and objectivity in, 104 training, 102–4, 156–58 Clinical method, 81–126 and aesthetics, 119–21 description role, 123–25 and emotion, 108–10, 116–18 instruction in, 133, 135 and intuition, 121–23 judgment in, 99, 104 and knowledge of persons, 82– 86 meanings in, 111–15 observational skills, 89–94 question-answer pairs in, 146–47 small group teaching, 135 subjectivity and objectivity in, 94 training, 82–85 Commodity model, 38 Communication skills in clinical empiricism, 60 clinical interview importance, 148–50 training, 10–11, 60 Community health centers, 6 Community setting, and training, 128–29 Competitiveness, 132, 135 Complete blood count, 161 Computers. See also Technology self-perpetuating nature of, 72 in twentieth-century medicine, 46 Con¡ict of interest in fee-for-service medicine, 32 and gatekeeping, 32–33 morality in, 33 Consultation, 171–72 Coordination of care, 169 Coronary angioplasty, 63 Coronary arteriogram, 67–68

Index Coronary artery disease chest pain relationship, 70–71 complexity of, 68–69 and technology, 67–71 Cost factors and gatekeeping, 33 and practice guidelines, 16 Curriculum development and didactic instruction, 133 and generalism, 9, 141–43 recommendations, 139 Davidoff, Frank, 125 Death medical student reactions, 180 medicine’s focus, 22–23 Decision making, 99. See also Clinical judgment Delivery of health care, 39–42, 46– 47 Denial, 26–27 Depression, 54–55, 168 Description, 123–25, 153 Desire, 184–87 Diagnostic tests, 38–39. See also Technology inef¤cient use of, 38–39, 62 limitations of technology, 62– 80 training in use of, 159–62 Diagnostics, 159–63 Diffusion model, 138–39 Disability in chronic disease, 22–24 education and training role, 165 illness relationship, 24–25 impairment distinction, 22 Discursive thought, 121 Disease concept/model, 14–16 back pain explanation, 58 and categorization, 35–36 and diagnosis, 92 versus ecological model, 18–19

[201]

limitations, 14–16, 18–19, 35–36, 48–49, 51–53, 84–85, 155 meaning in medicine, 14–15, 48 versus person-centered model, 51–53 reductionist aspects, 85 and technology, 63–64 Distancing, 109 Doctor Dock, 140, 174 Doctor-patient relationship, 153–56 aesthetic aspects, 119–21 bilateral nature of, 154 disease model versus patientcentered model, 153 emotions in, 108–10 fundamental importance of, 13– 14 meanings in, 111–15 and training experience, 129 Dying patients. See Terminal care Ecological theory, 18–19 Education and training basic requirements, 7–13 clinical empiricism implications, 6o–61 and clinical judgment, 102–4 and clinical method, 82–85 competitive individualism emphasis, 132 current emphasis of, 47 diffusion model, 138–39 and intuition, 122–23 location of, 11–12, 126–43 in medical technology management, 8o objectivity-subjectivity balance in, 12 Oslerian model, 139–40 principles, 7–9 and soft facts, 118 and values, 131–32

[202}

Index

Elderly, 164–65 Electrocardiogram, 66, 161 Emergency medicine, 78 Emotion, 108–10, 116–18, 180–83 repression of, 109–10 and terminal care, 180–91 and physician training, 118 Empathy, 127–28 Empiricism, 59–61 in clinical method, 87–88 educational implications, 60–61 observation and description in, 59–61 Endoscopy, 73, 162, 172 Engel, George, 49 Entry-level medicine, 27–29 Ethics, 165 Evidence-based medicine, 101 Experience, 126–37 Family physician movement, 4–6 Family physicians ¤rst contact care, 34 and gatekeeping, 33 hospital training, 133–34 place of training, 131 primary care delivery role, 40–41 role of, 5, 12, 40–41 Fee-for-service medicine, 32–33 Feelings, 108, 116–18, 121 Fifteen-minute visit, 16 50/50–110 proposal, 142 First-contact doctor functions, 32 skills of, 38–39 First contact visits, 36–38 Functional continuum, 163–66 Galen, 59 Gatekeeping, 4, 32–33 con¡ict of interest in, 32–33 ¤rst contact doctor distinction, 32, 34–35

General internists ¤rst contact care, 34 hospital training, 133–34 perspective of, 12 place of training, 131 primary care delivery role, 41– 42 Generalists career path, 141–43 classroom instruction of, 132– 33 de¤nition, 4, 142–43 focus of, 5–6 hospital training, 133–34 income, 142 knowledge base, 28 Oslerian teaching model, 139– 40 place of training, 131–33 referral to specialist criteria, 170–73 relationship to specialists, 169– 74 training weaknesses, 6–7 Geriatrics, 164–65 Guttentag, Otto, 169 The Healer’s Tale (Kaufman), 143 Health changing de¤nition, 18–21 ecological theory, 18–19 Health care team, 169 Health maintenance organizations and con¡ict of interest, 32–33 and gatekeeping, 32–33 Heuristics, 158 Hippocrates, 59, 87, 125 History-taking clinical empiricism emphasis, 60 clinical method central role, 88, 96–97

Index and diagnostic hypothesis formation, 91–94 primary care physician skill, 38–39 skills in, 38–39, 148–50 teaching of, 148–50 videotapes/audiotapes in, 148– 50 “Holistic” medicine, 168 Home care, 131 Hospice care, 5 Hospital training, 129–32 “Hospitalist,” 130–31 House calls, 131 Illness and beliefs of patient, 115 versus disease, focus of medicine, 182–83 variable expression of, 82 Imaging techniques, training, 159– 62 Impairment, 22 Income, 142 Infectious disease model, 37. See also Disease concept/model Informatics, 159 Information systems, 46–47 training, 158–59 in twentieth-century medicine, 46–47 Institute of Medicine report, 20–21 Intensive care units, 78 Internet, 159 Interpretation bias in, 90–93, 113, 145 and meaning of symptoms, 113–15 and observation, 90–93 Interview, clinical. See Clinical interview Intimacy, 184–85

[203]

Intuition, 92, 121–23 accuracy of, 121–22 and aesthetics, 119–20 de¤nition, 121 skilled use of, 122 and training, 122–23 Judgment. See Clinical judgment Knowledge-based skills need for, 8 specialists versus generalists, 28 Laennec, Theophile Rene, 48 Language, 98–99, 149–50 in clinical interview, 149–50 medical experience clari¤cation, 98–99 Life world, 113–14, 120 Literature in medicine, 123 Locke, John, 59, 87–88, 94, 125 Logic, 114–15, 121 Love, 183–87 Low back pain. See Back pain Magnetic resonance imaging, 63, 69, 71, 76, 96, 160–62 Managed care, 16 Many Farms, Arizona, 26 Master class method, 135, 158 Meanings, 111–15 Medical interview, 139. See also Clinical interview Medical model, 48–49. See also Disease concept/model Medical schools curricular change process, 9 generalist tracks, funding, 140–41 primary care training, 136 Medlars, 159 Mental health, 168–69 Mentors, 138 Molecular mechanisms, 51

[204}

Index

Morality, and con¡ict of interest, 33 Multispecialty groups, 136 Narrative approach, 123–24, 134– 35 Nonverbal communication, 149 Nurse practitioners, 28–29 Observational skills clinical empiricism emphasis, 60 and clinical method, 89–94 components, 123 and interpretation, 92–93, 120 narrative approach, 123 theory relationship, 93–94 Obstetrician/gynecologists, 40 Oncologists, 172–73 Osler, William, 12, 86, 125, 139–40 Outpatient setting. See Ambulatory settings Pain management of, 163 technology in evaluation of, 75– 76 Palliative care, 5 Patient-centered medicine case examples, 53–58 clinical empiricism in, 59–61 de¤nition, 84 and disease model, 51–53, 155 and doctor-patient relationship, 155 educational implications, 60–61, 82–86 overview, 51–53 and technology, 77–80 Pediatricians ¤rst contact care, 34 place of training, 131 primary care delivery role, 41 Percepts, 6o, 97–98, 119, 123

The person, 51–53, 82–85. See also Patient-centered medicine Persona, 108–9, 150 Personality, 107 Pharmacology, 163 Physical examination, 150–53 clinical empiricism emphasis, 60 clinical method central role, 88, 151–52 and diagnostic hypotheses, 93– 94 as part of treatment, 151 primary care physician skill, 38–39 purpose of, 151 required skills, 38–39 Physician assistants, 28–29 Physician’s role, 108, 182–83 Place of training, 11–12, 126–43 Plato, 185 Positivism, 60 Positron emission tomography, 63 Postgraduate training. See Education and training Power, and technology, 73–75, 79–80 Practice guidelines clinical judgment relationship, 101 cost control emphasis of, 16 Preventive medicine, 166–67 Primary care delivery, 39–42 Primary care medicine de¤ning characteristics, 4, 19–21 primary health care distinction, 21 Primary health care, 19–21 Psychiatry, 168–69 Psychological factors, 49–51, 168– 69 Quality of care, 130 Question-answer method, 144–47

Index Rationalism, 59 Reductionism as obstacle to change, 49–51 versus person-centered model, 85 Referral and cancer care, 172–73 in primary care medicine, 27 and specialists, 170–74 Rehabilitation, 166 Repression, 109 Residency positions, 142 Robert Wood Johnson Foundation, 140 Role models, 137 Rule-governed thinking, 134, 157– 58 Scienti¤c empiricism, 59. See also Subjectivity-objectivity balance Secondary care, 29–31 Seminars, 132–33 Sexual feelings, 109–10, 184–86 Sigmoidoscopy, 162 Simulated patients, 148–49 Small group teaching, 135 Social factors, 49–51 Soft facts, 117–18 Sonography, 160 Specialists acute disease model of, 4, 30–31 cancer care referral, 172–73 and chronic disease, 30–31 fragmented care, 30–31, 172–73 generalist relationship, 169–74 knowledge base, 28 and medical problem complexity, 28 in primary care delivery, limitations, 38–39 referral to, 170–74 training of, 28

[205]

Stead, Eugene, 87, 125 Stereotypes, 45 Stoeckle, John, 142 Subjectivity-objectivity balance, 8–9, 12, 44–45, 85–86, 94–99, 124 Suffering, 55 and meaning of symptoms, 112, 115 and palliative care, 5 Surgery, 66 Survival, 130, 173 Sydenham, Thomas, 59, 87, 93–94, 125 Symptoms, meaning of, 112–13 Team approaches, 135–36, 169 Technological mis¤t, 26 Technology access to, 172–73 and certainty, 70–72, 78–79, 110–11 de¤nition, 63 disease model relationship, 63– 64 fascination with, 64–65 immediacy of, 65–67, 78–79 knowledge of persons con¡ict, 9 and patient-centered medicine, 77–80 and power, 73–75, 79 in primary care, limitations, 26 problem of, 62–80 self-perpetuating nature of, 72– 73 specialists’ use of, 30, 172–73 training in, 159–62 in twentieth-century medicine, 46–47 unambiguous values in 67–69, 79

[206}

Index

Terminal care emotion in, 180–81 and intimacy, 184–85 primary care physician’s role, 173 and training experience, 129 Tertiary care, 29–31 Therapeutic alliance, 105–107, 113, 120, 154 Therapeutics, 163 Training. See Education and training Urinalysis, 161

Values in doctor-patient relationship, 107 and meanings, 111–12 and patient’s behavior, 115 teaching of, 131–32 Videotapes, 133, 148–50 War on Poverty programs, 6 World Health Organization, 19 Worried well, 35 WYSIWYG, 65