Promotion of Work Ability towards Productive Aging: Selected papers of the 3rd International Symposium on Work Ability, Hanoi, Vietnam, 22-24 October 2007

PROMOTION OF WORK ABILITY TOWARDS PRODUCTIVE AGING

PROMOTION OF WORK ABILITY TOWARDS PRODUCTIVE AGING Selected papers of the 3rd International Symposium on Work Ability, Hanoi, Vietnam, 22–24 October 2007 Editor

Masaharu Kumashiro Department of Ergonomics, University of Occupational and Environmental Health, Kitakyushu, Japan

Taylor & Francis is an imprint of the Taylor & Francis Group, an informa business © 2009 Taylor & Francis Group, London, UK Typeset by Charon Tec Ltd (A Macmillan Company), Chennai, India Printed and bound in Great Britain by Antony Rowe (A CPI-group Company), Chippenham, Wiltshire All rights reserved. No part of this publication or the information contained herein may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, by photocopying, recording or otherwise, without written prior permission from the publishers. Although all care is taken to ensure integrity and the quality of this publication and the information herein, no responsibility is assumed by the publishers nor the author for any damage to the property or persons as a result of operation or use of this publication and/or the information contained herein. Published by:

CRC Press/Balkema P.O. Box 447, 2300 AK Leiden, The Netherlands e-mail: [email protected] www.crcpress.com – www.taylorandfrancis.co.uk – www.balkema.nl

ISBN: 978-0-415-48590-6 (Hbk) ISBN: 978-0-203-88251-1 (eBook)

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Table of Contents

Preface

IX

Chapter 1 Overview Achievements in Aging and Work “period 1989–2007; Challenges after 2007” W.J. Goedhard Measures for the aging workforce in Japan from the perspective of the occupational health, safety and ergonomics M. Kumashiro

3

9

The effects of work-related and individual factors on work ability: A systematic review T.I.J. van den Berg, L.A.M. Elders & A. Burdorf

15

Systematic review for assessing job demands and physical work capacity in safety jobs A. Ropponen, J. Smolander & V. Louhevaara

19

Chapter 2 Work Ability Index The place of age in organisational policymaking: Evidence from an Australian qualitative survey L. Brooke, P. Healy, J. Jiang & P. Taylor Establishment of a Work Ability Index network in Germany H.M. Hasselhorn, B.H. Müller, R. Tielsch, B. Gauggel, G. Freude & J. Tempel WAI among workers in SMEs at Wholesale, Fruits, Vegetables and Flower Market in Brazil – from research to action I. Monteiro, K. Tuomi, J. Ilmarinen, J. Seitsamo, E. Tuominen & H.R. Corrêa-Filho Factors associated to the work ability among Brazilian teachers T.G. Vedovato & I. Monteiro Factors that predict work ability: Incorporating a measure of organisational values towards ageing J. Palermo, L. Webber, K. Smith & A. Khor

27 33

37 41

45

Effect of Japanese employment system on Work Ability Index M. Tokuhiro, H. Izumi, J.-L. Malo, N. Uehara & M. Kumashiro

59

Study on work ability of Vietnamese workers in selected industries D.K. Van & N.N. Nga

65

WAI among young employees in Brazil with new scores I. Monteiro, J. Ilmarinen, K. Tuomi, J. Seitsamo, E.P. Goes, A.P. Fernandes & E.P. Hodge

71

V

Verifying of the theoretical model of perceived work ability in the field of teaching V. Louhevaaral & S. Järvelin

75

An assessment of the Work Ability Index (WAI) and its usefulness in predicting and promoting continued work in staff employed by a major UK Charity T. Reilly, A. Rees & M. Tipton

83

Individual and work related determinants of work ability in white-collar workers T.I.J. van den Berg, S.M. Alavinia, F.J. Bredt, D. Lindeboom, L.A.M. Elders & A. Burdorf

89

Workplace trauma exposure, emotional imbalance and work ability G.P. Fichera, L. Neri, S. Sartori & G. Costa

93

Work ability and all cause mortality: A 25-year longitudinal study among Finnish municipal workers J. Seitsamo & R. Martikainen

101

Impacts from occupational risk factors on self reported reduced work ability among Danish wage earners L. Sell, A. Faber & K. Søgaard

105

Work ability and work quality as indicators for a longer and more productive working life M. Melon, P. Cocco & G. Costa

113

Disabled people at work and work ability: A study in a Brazilian company E.P. Hodge & I. Monteiro

117

Work ability of a population of 40+ in Luxembourg N. Majery

121

Chapter 3 Staying at Work Older worker career plateau: Issues and remedies P. McCarthy & S. Moore

127

Can organisations influence employees’ intentions to retire? J. Oakman & Y. Wells

133

Extending the working life R.G. Goedhard & W.J.A. Goedhard

139

Pension preferences and work environment M. Stattin

143

Younger managers and considerably older subordinates I. Johansson

155

Training older workers and long-term development: Needs and obstacles J.-C. Marquié & L.R. Duarte

161

Chapter 4 Promotion of Health and Work Ability Promoting health and workability in Vattenfall AB Nordic, Sweden R.J. Mykletun & T. Furunes VI

169

Managers’ decision making latitudes in relation to managing ageing workers T. Furunes & R.J. Mykletun The DRUVAN-project: A major increase in Occupational Health Service based on the Metal Age method in a Finnish municipality improved the work ability and gave significant financial return O. Näsman & G. Ahonen

177

183

A study on effects of support systems for KAIZEN in casting production of non-ferrous alloy by aged foundry workers Y. Mizuno, N. Motegi, M. Sugiura, F. Matsuda, T. Yoshikawa, K. Sakai & T. Misawa

191

Discovering the treasure: The use of the Work Ability Concept and the WAI in a bus company within a workplace health promotion process J. Tempel & J. Schramm

195

Prolonging working life in intellectual work Ü. Kristjuhan

209

Ergonomic and safe design of railway vehicles for elderly and handicapped people M. Rentzsch & D. Seliger

213

Female workers’ superior peculiarity and consideration for aging for the activation of “super-advanced age and fewer children” society K. Mikami, K. Iida & M. Kumashiro

219

Usability research on the older person’s ability for Web browsing D. Kobayashi & S. Yamamoto

227

Chapter 5 Age Affected Functions Impact of psychosocial work environment factors measured by the COPSOQ on the need for recovery after work in aging workers. Preliminary results P. Kiss & M. De Meester Age differences in mental workload while performing visual search task M. Takahara, T. Miura, K. Shinohara & T. Kimura A study of clinical assessment of Unilateral Spatial Neglect using a Head Mounted Display system (HMD) for elderly stroke patients in a virtual reality technology T. Tanaka, S. Ino, T. Ifukube, S. Sugihara, S. Shirogane, Y. Oyama & Y. Maeda Postural control using a vibratory feedback system for balance training in the elderly T. Tanaka, S. Ino, T. Ifukube, S. Shirogane, Y. Maeda, Y. Oyama, S. Sugihara & T. Izumi A study of static and dynamic balance abilities in the elderly Y. Maeda, T. Tanaka, S. Ino, T. Ifukube, S. Shirogane & Y. Oyama Differences of visual information processing between younger and aged person K. Yamanaka, Y. Nakanishi & M. Kawakami VII

239 247

253

263 271

275

Chapter 6 Nurse Effects of disturbed sleep on work ability and well-being among European nurses D. Camerino, P.M. Conway, S. Sartori & G. Costa

283

Physical/mental recovery and work ability of nurses in Japan C. Miyamoto, Y. Suenaga, M. Ando, N. Noda, Y. Okui & M. Hashimoto

293

Relationship between depersonalization syndrome and medical malpractice among Japanese nurses M. Sugiura, M. Hirosawa, Y. Yamada, Y. Nishi, S. Tanaka & M. Mizuno

299

Effect of continuous exercise on job stress among Japanese nurses Y. Kawata, Y. Yamada, M. Sugiura, Y. Nishi, M. Mizuno, M. Hirosawa & S. Tanaka

303

Relation between Typus Melancholicus and burnout among Japanese nurses Y. Yamada, M. Hirosawa, M. Sugiura, Y. Nishi, S. Tanaka & M. Mizuno

307

Relationship between depression and depersonalization among Japanese nurses S. Tanaka, M. Sugiura, M. Hirosawa, Y. Yamada, Y. Nishi & M. Mizuno

311

Effect of obsessive personality traits and impulsiveness on obsessive-compulsive disorder and eating disorders among hospital nurses Y. Nishi, M. Hirosawa, M. Sugiura, Y. Yamada, S. Tanaka & M. Mizuno Work-family balance and stressors among Japanese administrative nurses M. Mizuno, Y. Yamada, M. Hirosawa, M. Sugiura, Y. Nishi, Y. Kawata & S. Tanaka

315 319

Chapter 7 Others Time constraints and age: Health impact on musculoskeletal problems and perceived health B. Norma, S. Rosa & S. Francesco

325

Age-related differences in insurance claim rates for work-related injuries and diseases for different occupations and industry sectors K. Munk, P. Congdon & W. Macdonald

331

Age-related differences in patterns of return to work and compensation costs following work-related injury or illness K. Munk, P. Congdon & W. Macdonald

343

Keyword Index

355

Author Index

357

VIII

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Preface

This publication is a collection of selected papers from the 3rd International Symposium on Work Ability—Promotion of Work Ability Towards a Productive Aging, which is an international conference that was held for three days from Monday, October 22 to Wednesday, October 24, 2007. This conference was planned by the ICOH Scientific Committee for Aging and Work (ICOH SC for Aging and Work) and was jointly sponsored with the IEA Technical Committee for Aging, Vietnam Association of Occupational Health, and Department of Ergonomics, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health (UOEH), Japan. The ICOH SC for Aging and Work was founded in 1989. Since its founding the ICOH SC for Aging and Work has held two international conferences. The first was the International Conference on Aging and Work. The first Conference was held in The Hague in 19901 ), the second one was held in Helsingor, Denmark2 ) in 1998, the third in Kitakyushu, Japan3 ) in 2001, and the fourth in Krakow, Poland in 2002. The other was the International Symposium on Work Ability. The first Symposium on Work Ability was organized in Tampere, Finland in September 20014 ). The second Symposium on Work Ability was organized in Verona, Italy in October 20015 ), and irregularly held workshops on aging workforces have also been held For example,6 ). With the exception of the conference in Krakow, the proceedings of all ICOH SC for Aging and Work conferences have been published to propose measures for aging workforce societies from the standpoint of occupational health and ergonomics and other areas of the social sciences. The first ICOH SC for Aging and Work chairman is W. J. Goedhard. The first paper in this publication, Achievements in Aging and Work “Period 1989-2007; challenges after 2007”, uses the history of this conference to introduce readers to challenge of an aging workforce from an occupational health perspective. The aging workforce issues addressed by the occupational health and ergonomics fields are of particular interest to the well developed regions. Many of these areas are European countries in Northern Europe, and in Asia it is Japan that is actively working on these issues. In view of this, Vietnam in Southeast Asia was selected as the conference location for the ICOH SC for Aging and Work. There was a reason for holding the conference in Vietnam. The reason is that it is forecast that in the near future Asia will be unable to avoid the issue of aging workforces. The problems related to an aging workforce differ among countries. For example, Japan is dealing with an aging society with a low birthrate, which is reducing the working population and increasing the ratio of aging workers in the productive age population. This demographic distortion is impacting the foundation of the pension system. In response, the labor administration is working to create an employment environment where people work until age 70. In particular, Japan is working to improve the workplace. Europe, on the other hand, is concentrating on human resources and work ability evaluation methods for their extension. Differences can also be seen in “work ability”, which was used in the title of this publication, between Japan and Europe in how the evaluation results are used. This publication addresses in a major way the Work Ability Index (WAI) as an index for evaluating work ability. This was developed by the Finnish Institute of Occupational Health as a tool for evaluating work ability of workers. The ICOH SC for Aging and Work has promoted the use of this index through international conferences, publications and other means and construction of an international database based on this index. As a result, today the WAI is used in 25 languages. WAI is becoming an international and multi-cultural technique. However, to advance the promotion of work ability, which is the subtitle of this publication, how to use work ability evaluations and how to achieve the sought after productive aging must be studied. Currently, unfortunately, there is a major gap between Japan and Europe regarding how the evaluation obtained from WAI is used. Europe appears to be using this index to say, “You have this much work ability, so please continue IX

working instead of retiring early.” On the other hand, Japan is using this index to allow workers to say, “I have this much work ability, so please allow me to continue working without making me retire because of my calendar age.” This publication is a collection of papers that discuss from a variety of angles the goal of developing a truly international standard tool that can be used in common internationally taking into account differences in circumstances among countries. It is our hope that soon there will be a common guideline that exceeds national and regional boundaries that can be used for the diagnostic evaluation of work ability and employability. In conclusion, I wish to thank Mr. Jean-Luc Malo for the great effort he demonstrated in editing this publication. The majority of the time he spent as an assistant professor in my department was devoted to the work of holding this conferences and editing the corresponding papers. This publication is the result of his efforts. I also wish to thank Mr. Osami Hagiwara, president of Alphacom, for picking up where Mr. Jean-Luc Malo left off and assisting in the organization and editing of the transcripts for this publication. References: Materials issued by the ICOH SC for Aging and Work 1. AGING AND WORK, Willem J.A. Goedhard, 1992, ICOH Scientific Committee “Aging and Work”, Den Haag, ISBN 90-9005032-9 2. Experimental Aging Research, Jeffrey W. Elias, Vol.25, No.4, 1999, Taylor & Francis, USA, ISSN 0361-073X 3. AGING AND WORK, Masaharu Kumashiro, 2003, London: Taylor & Francis, ISBN 0-41527478-8 4. Past, Present and Future of Work Ability, -Proceedings of the 1st International Symposium on Work Ability- , Juhani Ilmarinen, Suvi Lehtinen, 2004, Finnish Institute of Occupational Health, Helsinki, ISBN 951-802-581-9 5. Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers, Proceedings of the 2nd International Symposium on Work Ability held in Verona, Italy between 18 and 20 October 2004-, Giovanni Costa, Willem J.A. Goedhard, Juhani Ilmarinen, 2005, Elsevier, The Netherlands, ISBN 0-444-51989-0 6. AGING AND WORK 4, -Healthy and Productive Aging of Older Employees-, Willem J.A. Goedhard, 2000, ICOH Scientific Committee “Aging and Work”, The Netherlands, ISBN 90803145-3-6 Masaharu Kumashiro, PhD, Professor of University of Occupational and Environmental Health (UOEH), Japan Chairman of ICOH Scientific Committee for Aging and Work

X

Chapter 1 Overview

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Achievements in Aging and Work “period 1989–2007; Challenges after 2007” Willem J. Goedhard The Netherlands Foundation of Occupational Health and Aging, Middelburg, The Netherlands

ABSTRACT: The Scientific Committee “Aging and Work” of The Int. Commission on Occupational Health (ICOH) was raised in 1989. Since the beginning the activities of the committee and its members have been growing resulting in many fruitful workshops, symposiums and a series of conference proceedings. The following officers of the committee were nominated by the Board of ICOH: Prof. (em.) Willem J.A. Goedhard (Middelburg, The Netherlands) as chairman and Prof. Juhani Ilmarinen (Helsinki, Finland) as secretary. In 2005 these officers were succeeded by Prof. Masaharu Kumashiro (Kitakyushu, Japan) as chaiman and Prof. Clas-Håkan Nygård (Tampere, Finland) as secretary. Aim of the committee: To preserve health and work ability of aging employees. The committee focuses on: • Older Employees • Studies on the relationship of Work Environment and the Aging Process of employees • Studies about the changes in Work Ability of aging employees Early developments: In the period before the start of the committee attention for older employees, for example during ICOH congresses was relatively small. This is possibly an indication that ageassociated changes of worker’s health were considered of little importance. Research concerning aging or age associated topics in occupational health was usually aimed at physiological changes in work capacity. From the beginning of the work of the committee research was more directed towards the challenges that were emerging from foreseeable aging of the workforces. This implied attention for prevention aspects of diseases, the importance of the aging process on the performance of aging workers and the relation between aging and mental capacity. Furthermore much research was dedicated towards aspects of (early) retirement. Also aspects of work stress in relation to worker’s age were receiving closer attention. It became clear that research concerning aging workers was explained rather differently in many countries possibly due to the different problems prevailing in these countries. Work ability and aging: A major break-through was the development and gradual implementation of the work ability index (WAI) by the Finnish Institute of Occupational Health in Helsinki. This questionnaire was translated in many languages and very soon study results emerged in which the WAI had been applied. The various study results among different cultures showed that the WAI was a valuable instrument to determine a worker’s ability to perform his/her job. Cross-sectional studies with the WAI showed negative trends of WAI scores in relation to age. However it became obvious that the work ability rate with age is relatively small explaining only about 5–10% of the variance of WAI. It can therefore be concluded that other variables, such as endurance capacity, are contributing considerably more to the level of work ability than calendar age. The scatter of WAI scores increases with advancing age. 3

Challenges after 2007: With a greater number of older participants in tomorrow’s workforces much knowledge about aging will be needed among Occupational Health (OH) physicians and OH personnel. Special programs aimed at older workers will have to be developed. No doubt the WAI will be a helpful instrument. Main activities will have to be directed towards age management. Then, it will be important to control prevention of chronic diseases which usually have detrimental effects on work ability. Last but not least control of excess job stress will be required since work stress may negatively affect aging and work ability. 1 ICOH SCIENTIFIC COMMITTEE “AGING AND WORK” FROM 1989 TO 2007 During the 20th century life expectancy increased considerably, not only at birth but also life expectancies at advanced ages (Fries, 1983). Improved living and working conditions theoretically allowed longer and more productive work lives. However, opposite developments were observed in many developed countries. Instead of retirement at 65 years, the retirement age that was already established during the 19th century; in the 70s and 80s of the 20th century shrinking of manpower in many industrialized countries was observed. Much unemployment followed and with financial aid of governments workers were entitled to early retirement around the age of 60 or even earlier (RedayMulvay, 2005). These developments were leading to the paradox of increased and improved life expectancies and health the percentages of people working beyond 60 years decreased considerably. Yet it could be foreseen that these developments had to reach a limit since another important development was observed, i.e. decreased birth rates in many countries. This would no doubt lead in the near future to shortages of manpower and the return of older workers in the workforces. This was more or less the situation in the beginning of the 1980s. An illustration of the situation is minimal attention for age-associated aspects of occupational health at that time (e.g. at the ICOH congress in 1984 in Dublin) only a few papers were dealing with topics pertaining to the aging process of workers: There was only one session on Aging, chaired by Prof. Asa Kilbom (Sweden) that comprised two papers related to aging (one from Finland by Ilmarinen et al. and one from the Netherlands by Goedhard et al.). In 1987 I wrote a letter to the late Professor Murray the president of the ICOH in which I suggested to raise a scientific committee that would concentrate on matters of aging in the working population. Professor Murray wrote 27th August 1987 “. . . My own personnel interest in the subject derives from the fact that I believe it is very useful to continue working as you get older. It is those who do not have a consuming interest, either in their work or in equally important leisure activity who die early. I can recognize that older people some times have to step aside to allow younger people to be promoted, but in many cases a life time of experience is wasted if they are merely given a retirement pension”. The board of ICOH responded positively and in 1989 the scientific committee “Aging and Work” was raised and officially started in Helsinki by prof. Sven Hernberg, president of ICOH. Nominated as officers of the committee were: myself as chairman and Prof. Juhani Ilmarinen (Finland) has secretary. The aim of the committee was to preserve health and work ability of aging employees. I am much obliged to my colleague Juhani Ilmarinen with whom I was able to work together for many years. We had many stimulating discussions and were able to organize a series of international workshops and conferences (see Table 1). The committee focuses on: Health of Older Employees, Studies on the relationship of Work Environment and the Aging Process (bridging the gap between occupational health and gerontology), and Studies about changes in Work Ability of older employees. Aging of the population – Some consequences of population aging are: (1) Aging of the workforce (2) Changes in retirement policies (3) Increased numbers of disabled workers Is aging of the workforce a problem? – (J-F. Caillard, former President of ICOH, 1994) wrote “. . . the survival of greater numbers of people into old age has great consequences for the health of the 4

Table 1.

1990 1992 1994 1994 1996 1998 1999 2001 2001 2002 2004 2007

List of venues of workshops and conferences under the auspices of the ICOH scientific committee “Aging and Work” between 1989 and 2007 The Hague (Netherlands) Haikko (Finland) Wijk aan Zee (Netherlands) Kitakyushu (Japan) Stockholm (Sweden) Helsingor (Denmark) The Hague (Netherlands) Tampere (Finland) Kitakyushu (Japan) Krakow (Poland) Verona (Italy) Hanoi (Vietnam)

Aging and Work Aging and Work Aging and Work Paths to Productive Aging Work after 45? Aging and Work Aging and Work Work Ability Aging and Work Aging and Work Work Ability Work Ability

labor force, as many remain in employment with the deficiencies and disabilities that accompany ageing”. Early developments – The start of the new ICOH committee: “Aging and Work” in 1989 was certainly not the beginning of research interests in older workers’ problems. In social gerontology much earlier than 1989 researchers were interested in topics concerning older workers. An example is the publication of a book in 1970: by professor Harold Sheppard (USA): He was White House counselor on aging in the Carter Administration (end of the 70s). In 1970 he edited and published a book “Towards an industrial gerontology, an introduction to a new field of applied research & service”: This book was considered of particular use for personnel management. A balance should be maintained between exertion of capabilities and performance (Sheppard, 1970). An important conclusion was: “it may someday be proven (or convincingly argued) that continued activity in some form of work (compensated or not) is an important way in which to slow down the process of aging”. In the Netherlands important research on aging workers was done by Dirken and co-workers (1972). They concentrated on physiological aspects of aging which led to the concept of functional age that may differ from calendar age in many people. Johan Dirken (Netherlands): 1970: Publication of his book “Functional Age of Industrial Workers: Functional age: i.e. measurable characteristics of an individual vs. his functioning in a physically or socially determined (work) environment (Dirken, 1972). Finally, in Jan Baart (1973) wrote a thesis about older dock workers. One of his conclusions was “Older dock laborers should be allowed to adjust their work speed. What is an older worker? – The committee focuses on older workers. The question arises what is an older worker? Without giving some exact definition it is mores sensible to provide some characteristics of an older worker. It is now generally accepted that the following characteristics are appropriate: • Older than 45 years [1996: congress in Stockholm: theme: “Work after 45?”, At this congress Hjort (1997) proposed an official retirement age of 70 years]. • In the final third of professional career. • Age-associated changes which may affect work ability. • Increasing risk of chronic diseases. • Aging of professional skills and knowledge. • More experienced. 2 HIGHLIGHTS FROM CONFERENCES AND WORKSHOPS Below I will shortly refer to some highlights of conferences and Workshops as indicated in Table 1. The selection of papers is very much subjective. 5

WorkAbility Index (WAI) – This questionnaire which was developed and introduced in Finland by researchers of the Finnish Institute of Occupational Health can be considered an important breakthrough in the international research on aging workers. (1992: Haikko) Ilmarinen and Tuomi (1993): • Work Ability Index for Aging Workers: Report on a longitudinal study that started 1981 on 6,257 workers (45–58 years). • It took some more years before WAI was introduced in other countries. • The presentation of the WAI was followed by the launch of the concept of “age management”. These developments allowed the study of aging workers in different countries with the application of a standardized instrument. In the meantime this has been proven a very useful development. The WAI has been translated in many languages and is used in many countries. The results of studies with WAI have given rise to the organization of several international conferences, like Tampere (2001); Verona (2004) and Hanoi (2007). Age Management – It has clearly been demonstrated by Ilmarinen (1999) that good age management is a very useful tool in order to maintain employability of aging workers. Age Management implies promotion of Work Ability. It is one of the main objectives of the committee Aging and Work to disseminate this view among employers and employees. Is this necessary? The answer is Yes, because: • In the EU it was decided: [EU summit conferences in Barcelona 2002, Stockholm 2001] to the 2010 objectives: i.e. • More than 50% of the 55–64 yr old population should be active in the workforces. • Reduction of early retirement policies. • Age of retirement to be increased to 65 years in all EU countries. It is rather questionable whether the EU targets will be realized. In the Netherlands it was found that employees and employers are not in favor of extending the working life (van Dalen et al., 2005). Most employees expect to work till 62 years. Most employers consider extending the working life of their employees to 65 years as undesirable. The central issue here is the question whether older workers will be able to continue working. At the conference in Stockholm (1996) an important contribution to this discussion was presented by Salthouse: “Implications of adult age differences in cognition for Work Performance needed” (1997). There is little relation between age and measures of work performance. This conclusion is consonant with Dirken (1972). The good news is: Increased age is not a liability in most employment situations. However the “bad news” is: age-related decline in cognitive abilities is to be expected. An important aspect of Age management is the maintenance of knowledge and professional skills. This requires continuous learning which may become difficult in case of reduced cognitive abilities. An important contribution at the conference in Helsingor 1998 was by Näsman & Ilmarinen (1999): “Metal-Age: A process for improving well being and productivity. Methods for starting interventions in enterprises e.g. improvement of the work organization: The important key words are: participatory planning, i.e. finding solutions together. This method derives to get more attention. It is applied to find the most crucial development targets of well-being at work. This is comprehensively discussed in the publication: “Well-being creates productivity” (Rissa, 2007). The problems of aging workforces are not limited to Europe or the USA. Also Asian countries have to deal with reduced birth rates and subsequently aging workforces. For example in Japan the age group 55–64 year old workers was estimated to increase from 15.5% of the population in 1990 to 20.8% in 2010. Kumashiro described a strategy to retain productive aging. Key aspects in this strategy are: estimation of functional age, maintenance of Work Ability and an optimum work motivation (Kumashiro, 1999). Winn (1999) pleaded for a more efficient use of older workers in the United States. He indicated that social minorities are more likely than non-minorities to retire 6

early because of poor health. Yet many work longer than the normal retirement age in order to maintain their living standards. Vitality – Vitality is not a usual biological concept. In many Gerontology textbooks the word is often not indexed. The word expresses something like energy or “vital force” (i.e. the power to live or go on living. At a conference in The Netherlands in 2006 (Wijffels, 2003) stated that that modern societies should invest in vitality of elderly people and the community. Elderly people should be enabled to participate actively and enjoy a high quality of life. This will imply that there should not be any age limit to education. Work Ability of men and women – Application of the WAI in several countries has resulted in numerous studies among workers in many countries. Differences between workers in different professions were visualized. Also gender differences in work ability can become clear. In an evaluation of functional working capacity by the WAI in Italian workers it was found that female workers have lower WAI scores than men and workers under stress (strain) are more prone to decrease in fitness and work ability (Costa et al., 1999). An important comprehensive study is the NEXT study (nurses early exit study). This study that was setup in nine European countries among 40,000 nurses and provided better understanding of the variables contributing to work ability scores. Camerino et al. (2005) showed that stress factors like harassment at work (uncertainty about treatment), low job control, and high job demands, tend to give low WAI scores and job alienation. 3 CHALLENGES AFTER 2007 It can be reasoned that in many industrialized countries more older workers (60–65 + yrs) will re-enter the workforces. Extending the working life is an important challenge for occupational health physicians in the future. They will have to deal more than in the recent past with increases of age-associated disorders (risk of disability). Disability rates will possibly increase. It will become more difficult to maintain skills and knowledge of these aging workers. It must also be realized that age-associated declines in cognitive abilities (Salthouse, 1997) may affect the possibilities of workers to keep up with new developments in their profession. The key-item is employability. This is preceded by promotion of work ability (Ilmarinen, 1999). An other important aspect of work in modern societies is the level of perceived stress. It is therefore essential to realize that on-going work ability control and on-going stress control are warranted in order to maintain employability of older workers. Variability in work ability – Work ability has the tendency to decrease with advancing age. In cross-sectional studies usually negative trends are observed in WAI scores of the studied population (Goedhard, 2004). The scatter in observed WAI scores also changes over time resulting in increased 50 45 40 35 30 25 20 15 10 5 0

WAI
47.9  0,14  age R2
0,91 (p.01)

Variance
0,84  age  9,6 R2
0,71 (p.05) 0

10

20

30 40 Age (years)

Figure 1. WAI-score mean and variance per 5-years age-group

7

50

60

70

variances of WAI scores (see Figure 1). This implies that differences between workers increase with advancing age. Work stress control and age – Based on earlier observations the preliminary conclusion was reached that perceived work stress increases with age. Will older workers be able to cope with these stresses? What could be the negative outcomes of work stress? An other conclusion was that baroreflex function is possibly affected by work stress (Goedhard, 1993). Recently it was found that prolonged stress affected telomere length in caregivers. Women with the highest levels of perceived stress have telomeres shorter on average by the equivalent of at least one decade of additional aging compared to low stress women (Epel et al., 2004). These findings have implications for understanding how, at the cellular level, stress may promote earlier onset of age-related diseases. Such findings illustrate the serious effects of work stresses on important biological systems. REFERENCES Baart, J., (1973). De oudere havenwerker (the older dock worker) (Dutch). Thesis, Erasmus University, Rotterdam Caillard, J.F., (1995). In: Cox, R.A.F. et al. (eds) Fitness for Work. Oxford University Press, Oxford, Foreword, pp. iv–v Costa, G. et al., (2000). Evaluation of functional working capacity by the work ability index in Italian workers. In : Goedhard, W.J.A. (editor), Aging and Work 4, ICOH SC Aging and Work, printed: Pasmans, The Hague, pp. 53–61 Dirken, J.M. (editor), (1972). Functional Age of Industrial Workers. Wolters-Noordhoff, Groningen, Netherlands Epel, E.S. et al., (2004). Accelerated telomere shortening in response to life stress. Proceedings National Academy of Sciences; 101/49: pp. 17312–17315 Fries, J.F., (1983). Compression of morbidity. The Milbank memorial fund Quarterly 61/3: pp. 397–419 Goedhard, W.J.A., (1993). Work stress and the aging process. 24th Int. Congress ICOH, Nice, France, Book of Abstract, pp. 67 Goedhard, W.J.A., (2004). WAI scores and its different items in relation to age: a study in two industrial companies in the Netherlands. In: Ilmarinen, j. and Lehtinen, S. (editors). Past, Present and Future of Work Ability, FIOH, Helsinki, Research Reports 65: pp. 26–40 Hjort, P.F., (1997). Age and Work-good or bad for whom? In: Kilbom, Å et al. (editors), Work after 45? Arbete och Hälsa, 29: pp. 3–13 Ilmarinen, J., and Tuomi, K., (1993). Work ability index fora ging workers. In: Ilmarinen, J. (editor), Aging and Work, Finn. Inst. Occup. Health, Helsinki, pp. 142–151 Ilmarinen, J., (1999). Ageing workers in the European Union. Status and promotion of work ability, employability and employment, Finn. Inst. Occupational Health, Helsinki. Kumashiro, M., (1999). Strategy and actions for achieving productive aging in Japan. Experimental Aging Research, 25/4: pp. 461–470, 379–384 Näsman, O., and Ilmarinen, J., (1999). Metal-Age: A process for improving well being and productivity. Experimental Aging Research, 25/4: pp. 379–384 Reday-Mulvey, (2005). Working beyond 60. Key policies and Practices in Europe, Palgrave MacMillan, New York Rissa, K., (2007). Well-being creates productivity (the Druvan model). Centre for Occupational Safety, Helsinki, pp. 80. ISBN: 978-951-810-340-3 (PDF) http://www.tsr.fi/files/Julkaisut2007/wellbeing productivity. pdf (valid fev. 2007) Sheppard, H.L. (editor), (1970). Towards an Industrial Gerontology. Schenkma Publ., Cambridge, Mass. Salthouse, T.M., (1997). Implications of adult age differences in cognition for work performance. In: Kilbom, Å et al. Work after 45? Arbete och Hälsa, 29: pp. 15–28 Van Dalen, H.P. et al., (2005). The double standard in attitudes toward retirement- the case in the Netherlands. The Geneva Papers, Palgrave MacMillan, 30/4: pp. 693–710 Wijffels, H., (2006). Investeer in de vitaliteit van mensen en de samenleving (Dutch). In: Tielen, G. and Dortland, B (editors). Van Nazorg naar voorzorg, Kon. Holl. Maatsch. Der Wtrenschappen, Haarlem, Netherlands, pp. 19–28 Winn, F., (1999). Structural impediments to the efficient use of older workers in the United States. Experimental Aging Research, 25/4: pp. 451–459

8

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Measures for the aging workforce in Japan from the perspective of the occupational health, safety and ergonomics Masaharu Kumashiro Department of Ergonomics, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, Kitakyusyu, Japan

ABSTRACT: The aging of population has major repercussions on the workforce population. The basic issues of such an aging workforce society that must considered from the perspective of the occupational health, safety and ergonomics are (1) enhancing the health resources of the citizenry, (2) human resources management that includes setting wages and benefits based on objective evaluation of workability and employability, and (3) work and working environment management focusing on job design that takes aging workers into account. To resolve these issues requires making an employment environment in which employer and employees can work comfortably without being conscious of calendar age while maintaining the high labor productivity that can be thus created. This is to create a so-called “age free” workplace. The core strategy for creating an age free workplace is to avoid as much as possible mismatches between work ability and the job requirements placed on the workers. To do this, how work ability is evaluated and the factors that hinder maintaining and increasing work ability must be identified. At the same time, working conditions and working environment that allow the work ability acquired by workers to be fully utilized must be created (set). In this paper, the former is positioned as a strategy at the company level. In particular, the difference from the West of the work ethic held by Japan has given birth to employability that is unique to Japan. An evaluation for the Japanese style work ability that is the source for this must be considered. On the other hand, the later is positioned as a strategy at the workplace level, and it can be used to weigh the actual condition of Japanese company KAIZEN activities. This will also serve as a reference for strategies that support future workplaces.

1 JAPAN AND THE AGING WORKFORCE SOCIETY The average lifespan for Japanese people announced in 2005 was 78.53 years for men and 85.49 years for women. For men, this figure is second to that of Iceland, while women live longer in Japan than anywhere else. Compared to the figures for 1950 (58 and 61.5, respectively), the average lifespan has lengthened by more than 20 years for both men and women. Meanwhile, the productive-age population peaked in 1995, after which the trend has been toward decline. According to projected population statistics released in 2002, the productive-age population comprised 66.4% of the total Japanese population in 2005. This is estimated to fall to 58.5% in 2030. Focusing on middle-aged and older workers aged 45 and over, the percentage has increased since 1990, and in particular, a dramatic increase is foreseen for 2015 and after. Japan has truly entered the era of an aging society with a low birthrate. One can relate to the negative aspects of this phenomenon when observing it from the viewpoint of maintaining a productive-age population. This means that 4.1 million people between the ages of 60 and 64, or about 2/3 of that age group, will have to be working in 2010 in order to maintain a productive-age population between ages 15 and 60 comparable to that in 1998 (58.7 million). Similarly, by 2015, everyone up to age 64 and 10% of the over-65 population will have to be working. In 1973, the Japanese government set the target retirement age at 60, and subsequently examined legislating this target, which in 1994 became law. The next step was the start in 1999 of a campaign 9

to promote model businesses in which people could work until age 65, and then seven years later in April 2006 the Revised Law for the Stabilization of Employment for the Aged came into force. As a result, businesses are under pressure to prepare for introducing a continuing employment system up to age 65. In spite of the fact that the law came into effect after a preparation period lasting 33 years, any negative judgments made by employers regarding employment up to age 65 relate to skepticism about the health, physical capabilities, motivation to work and the like of aging workers. The aging phenomenon is unavoidable as long as human beings are animals. The fact that aging brings about diminished fluid ability can be felt in daily life. As a result, there is a general tendency to dismiss aging workers as being out of the running. This tendency results in a vicious circle by creating an unhealthy aging society that has lost its motivation toward life and work. Meanwhile, according to a survey of aging people (25,224 randomly selected people between ages 55 and 69) regarding their employment status conducted by the Ministry of Health, Labor and Welfare in 2004, 71.5% of the men and 45.6% of the women were gainfully employed in September 2004. By age group, these figures were 68.8% for men aged 60–64, 49.5% for men aged 65–69, 42.3% for women aged 60–64, and 28.5% for women aged 65–69. It is unclear as to whether these percentages indicate people who were forced to work to support themselves, or those who consider that work gives them something to live for. However, it can be easily seen that a large number of aging people are seeking work. This means that the role of occupational health and safety is to provide lifestyle and health guidance to enable aging people to lead reasonable working lives, and to supply the know-how required to avoid mismatches between their working capabilities and the job requirements. Thus, the basic issues that occupational health and safety specialists will have to consider from now on are (1) enhancing the health resources of the citizenry; (2) human resources management that includes setting wages and benefits based on objective evaluation of workability and employability; and (3) work and working environment management focusing on job design that takes aging workers into account. The fundamental concept for resolving these issues is the building of a company that has a healthy workforce regardless of age or gender. This means making an employment environment in which employer and employees can work comfortably without being conscious of calendar age, and maintaining the high labour productivity that can thus be created.

2 PERCEPTIONS OF WORKABILITY IN JAPAN In looking at measures for aging workers in Europe, and particularly in the northern European countries, lively research is ongoing in the area of human resources and workability. In northern Europe, human resources research is viewed from the perspective of the natural sciences, starting with physiology and medicine. For Japan, where human resources and labour management is carried out from the viewpoint of social sciences, this dimension is probably as great a shock as the coming of Commodore Perry’s Black Ships in 1853. Japan also has focused on the relationship between unavoidable human aging and fluid ability. But diminished workability that accompanies aging has always been perceived visually. Specifically, the emphasis has been on developing and preparing support systems (working conditions, working environment, etc.) to compensate for decline that can be confirmed with the eyes, such as diminished muscular strength, failing stamina, fading vision and the like. Why has Japan not perceived this human aspect biologically, and evaluated human resources objectively on that basis? Perhaps they were working on the basis of Japan’s organizational operation and the labor management practice of “protecting one another” – for example: “I know you’re having a hard time. So am I. So let’s try together”. The “protect one another” concept resulted in the creation not only of bonding, arising from the life wisdom of agricultural people, but also in the concept of ostracism. Perhaps this encouraged the development of organizations that are oblivious of the distorted nature of this method of ability evaluation. Moreover, the “protect one another” concept covers up potential problems. The introduction of support equipment and tools may aggravate this aspect. So this is probably why an equation for evaluating the workability of aging workers has not been created outside the arena of calendar age. If a competitive society had resulted in a “pep talk” concept among the Japanese people, such as “If it’s so tough, you can stop. 10

If you stop, it’ll be easier for you, and it will mean one fewer competitor for me, so it will be easier for me to win”, then Japan might have turned to objective human resource evaluation based on biological thinking. 3 THE TIME HAS COME TO DEVELOP JAPANESE-STYLE WORKABILITY In order for individual human beings to maintain workability, they must first enhance the skills, experience, knowledge, interpersonal relationship building capabilities and the like that are required in their current jobs. Once a certain level of ability is achieved, then it is important to increase their capabilities to cope with peripheral work and gradually expand their work scope. Workability fostered in this way must eventually be carried forward into employability that can be utilized in the external labour market, not just within the company of the current employer. At the same time, companies as part of their role should be proactive in providing educational and training systems for the purposes of improving the quality of employee workability and expanding its applicable scope. Basically, companies must be aware that they are each a member of the single organization “Japan Inc.”, which is becoming an aging society with a low birthrate. And they must work together as one to improve and educate Japan’s human resources. Companies must become aware of the fact that unless they eradicate the insularity exemplified by the phrase “our own employees”, the deed will come back upon them when they end up with “only our own employees” having low workability. Essentially, Japan appears to have the grounding for developing workability into employability. In other words, employees in Japan usually spend their entire working life in the same company, in contrast to other countries. However, within that same company they move about from sales to production to office work, for example, in order to gain a variety of experiences. If this grounding could be exploited, this environment of changing workplaces within one company need only be standardized for application to employment in other companies. As described above, a wide range of employability becomes possible when one possesses workability in relation to various individual jobs. The issue for Japan going forward is whether or not the individual work abilities cultivated within just one company can be applied outside that company. In other words, the standardization of workability will be a crucial labor policy issue for the future. The three sources of workability are health, motivation and skill. Among these three sources, the most basic is the health of the individual. Health is composed of mental and physical capacity, an awareness of its role in social lifestyle including work and daily life, and the functions carried out. Then, the source of the required workability is motivation to work. This means the attitude to one’s work. The final aspect is technical skills, which is what many companies are most concerned about. These can be expressed in terms of a combination of knowledge, experience and technology (in terms of blue collar workers), or a combination of specialized abilities, adaptability to environmental change, and management abilities (in terms of white collar workers and managers). 4 OCCUPATIONAL HEALTH AND SAFETY STRATEGIES IN THE ERA OF AN AGING WORKFORCE SOCIETY WITH A LOW BIRTHRATE Many different measures are put forward when discussing how occupational health and safety should intervene in Japan’s aging workforce society with a low birthrate. From the perspective of occupational health, safety and ergonomics, the following four strategies can be proposed. The characteristic of strategies in occupational health, safety and ergonomics is that they can be developed at all levels, from the individual to the state level. (1) Strategies at individual level Examples: Ensuring health, rejuvenating the physical age (preventing reduction in life functions) (2) Strategies at workplace level Examples: Workplace improvement (KAIZEN) activities (including job re-design), developing support equipment and tools 11

(3) Strategies at company level Examples: Evaluating workability and building workability-based wage and benefit systems, horizontal development of success stories (among workplaces) (4) Strategies at state and regional administration levels Examples: Creating database of company success stories, publicizing the database, establishing a certified employability evaluation system

5 STRATEGIES AT WORKPLACE LEVEL WORKPLACE IMPROVEMENT (KAIZEN) – A CHARACTERISTIC AGING WORKFORCE MEASURE IN JAPAN Tables 1–3 illustrate the results of aging workforce measures as classified and arranged by the author, based on the joint research reports (April 1986–March 2007) published by the Japan Organization for Employment of the Elderly and Persons with Disabilities (JEED). Table 1 gives an interesting glimpse of the kinds of issues that concern Japanese companies in terms of aging workforce measures. This joint research system is broadly divided into four categories (job re-design, health care, human resources/wages, and capability development). Looking at the joint research results for the past 21 years, job re-design at 55.9% was most often cited, followed by capability development at 20.5%, human resources/wages at 15.3% and health care at 8.3%. This indicates that Japanese companies have shown great interest in workplace improvement activities carried out under the slogan of job re-design.

Table 1. Japan Organization for Employment of the Elderly and Persons with Disabilities Joint Research Reports (April 1986–March 2007) No. of cases Job re-design Health care Human resources/wages Capability development Participating companies

194 29 53 71 234

In fact, most companies encourage workplace improvement from the bottom up, under such names as KAIZEN Dojo, karakuri KAIZEN activities and the like. In Table 2, the focus is on measures taken by Japanese companies to lighten the workload on their own employees who are aging workers, in particular the number of times job re-design was implemented. The result shows that the issue of greatest concern is the load on the musculoskeletal system, at almost 60%.

Table 2.

KAIZEN activities in Japanese companies as measures for aging workers (April 1986–March 2007)

Vital function

KAIZEN cases (%)

Musculoskeletal functions Sensory organs Judgment functions Other (including unclassifiable) Total

470 (59.4%) 74 (9.3%) 50 (6.3%) 198 (25.0%) 791 (100%)

12

Note: The figures given in Table 1 relate to multiple joint research topics and thus represent combined totals. Also, even if improvement measures related to an individual function, they were classified under “Other” if the function was not a main target of improvement. Next, Table 3 focuses on this issue of greatest concern, the musculoskeletal system, and shows the elements that cause the load and the status of countermeasures taken. The first preventive countermeasure against musculoskeletal system disorders taken on the front line was improving defective work posture (36.1%), followed by transport of heavy objects (29.4%), and improved methods of lifting of heavy objects (27.6%). To summarize, the main concerns of Japanese companies in terms of aging worker measures are improving defective working posture and handling of heavy objects. Table 3.

Musculoskeletal system disorder preventive activities in Japanese companies as measures for aging workers (April 1986–March 2003)

KAIZEN target work

KAIZEN cases (%)

Lifting of heavy objects Transport of heavy objects Defective working posture Walking and ascending/descending staircases Isometric exertion Total

132 (27.6%) 141 (29.4%) 173 (36.1%) 21 (4.4%) 12 (2.5%) 479 (100%)

Further, one other keyword that can be found in Table 2, and indicated in almost 10% of KAIZEN activities, is dealing with sensory organ changes. Specifically, this means KAIZEN activities relating to age-induced reduction in visual functions such as near-point adjustment, night vision and dynamic visual acuity. In addition, a characteristic change in KAIZEN activities in Japanese companies is that they have started to be adapted to training and handing down of job know-how. Thus, some KAIZEN targets are no longer demanding the optimum KAIZEN results. This type of trend makes it increasingly difficult to fit corporate results from the past two to three years into the categories of Table 2 and Table 3. In any case, in Japanese companies all of these activities are carried out at the workplace level, by front line workers and managers. The most elementary and common KAIZEN activity carried out in Japan’s workplaces is the development and introduction of support equipment and tools from the production technology aspect. Next, workplace groups that have advanced one level further have acquired production control knowledge and skills, and are working on KAIZEN activities that incorporate method improvement, process improvement and other IE (Industrial Engineering) concepts. The procedure actually followed by mature KAIZEN teams is first to test low-cost method improvements, then to introduce equipment and tools. Meanwhile, recently not only are improvement measures being introduced for current work, but KAIZEN tools are also being developed. One of these that is often used is the application of the ergonomics checklist. In most of these cases, a checklist most suited to the workplace in question is created based on a sample of a general ergonomics checklist. For example, a formula is woven whereby the focus of the checks is divided into safety, human aging characteristics, product quality, and production efficiency, with scores allocated for each, and the KAIZEN priority is obtained from total scores. Finally, a formula has been developed whereby KAIZEN priority is determined according to the degree of workload.

6 CONCLUSION Measures for aging workers in Japan mix two types of tactics – those to be proud of and impoverished ones. Those in which to take pride exist as strategy at the workplace level. In other words, workplace 13

level tactics include the frequent utilization of workplace improvement and job design. Japanese companies excel at these tactics. Even on a global level, Japan’s level of knowledge, technique and experience for KAIZEN activities cannot be approached by any other country. Furthermore, Japan has training programs that allow this to be entrusted to the front line workers. In contrast, in terms of corporate level tactics, Japan is not very good at workability evaluation, which must be utilized proactively from now on. This is a tactic at which European countries excel, in particular northern European countries. In observing aging worker measures in northern Europe, one sees lively practical research being carried out in relation to human resources and workability in the field of occupational health and safety. Northern Europe perceives human resources research from the viewpoint of the natural sciences, starting with physiology and medicine. This clearly differs from Japan, which has carried out human resources research from the social sciences perspective. The background to this is the clear difference in thinking on workability and employability in Japan as compared to Europe. For Japan, with its seniority system and lifelong employment, evaluating workability using figures was difficult. Turning to the thinking on employability, as evidenced by the Japan Economic Federation in 1999, there is resistance to equating employability with the ability to make a career change. Employability could be defined as the ability fully to utilize one’s workability within the current employer and thus remain capable of continuing to be employed. In such an environment it is difficult to create standards such as those that exist in the West, for capability evaluation and development targets for the external labor market. However, it would appear that the very issue that Japanese companies must deal with urgently as they confront a rapidly aging society with a low birthrate is that of developing a workability evaluation method that incorporates the occupational health and safety perspective, and creating a workability bank that contributes to employability standardization. REFERENCES Joint Research Report (April 1986–March 2007), Japan Organization for Employment of the Elderly and Persons with disabilities, – This report is published annually in Japanese Juhani Ilmarinen, Aging workers in the European Union – Status and promotion of work ability, employability and employment, Finnish Institute of Occupational Health, Finland, 1999 Juhani Ilmarinen, Tuomi K, Past, present and future of work ability, Proceedings of the 1st International Symposium on Work Ability, Finnish Institute of Occupational Health, Finland, 2004 Juhani Ilmarinen, Towards a Longer worklife! – Aging and the quality of worklife in the European Union, Finland, 2005 Kumashiro, M., Strategy and Actions for Achieving Productive Aging in Japan, Experimental Aging Research, Vol. 25, No. 4, p. 461–470, 1999 Kumashiro, M., Japanese Initiatives on Aging and Work: An Occupational Ergonomics Approach to Solving this Complex Problem, “AGING AND WORK”, p. 1–8, Taylor & Francis, U.K., 2003 Population Statistics of Japan 2006 by PDF, National Institute of Population and Social Security Research Population Projects for Japan 2001–2050, National Institute of Population and Social Security Research, January 2002

14

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

The effects of work-related and individual factors on work ability: A systematic review Tilja I.J. van den Berg, Leo A.M. Elders & Alex Burdorf Department of Public Health, Erasmus MC, University Medical Center Rotterdam, The Netherlands

ABSTRACT: Objectives – This paper systematically reviews the scientific literature on the effects of individual and work related factors on work ability. Methods – Studies on work ability published from 1985 to 2006 were identified through a structured search in PubMed, and Web of Science. Studies were included if the Work Ability Index was used as measure of work ability and if quantitative information was presented on determinants of work ability. Results – In total, 20 studies were included. Important factors for a decreased work ability were lack of leisure-time vigorous physical activity, poor musculoskeletal capacity, older age, obesity, and high physical and psychosocial work demands. Conclusion – Work ability is influenced by individual characteristics, life style, demands at work and health status.

1 INTRODUCTION Aging of the working population increases the importance of physical and mental capacities of workers. In order to prolong the working life among older workers the concept of work ability has been developed in the early 1980s in Finland, and later adopted in various other European and Asian countries. The Work Ability Index (WAI) (Tuomi, 1998) is by far the most used, and well-accepted instrument to measure work ability. Although several studies in different occupational settings have been conducted, there is a need for a systematic evaluation of the relative importance of important determinants of work ability. The aim of this systematic review is to identify the individual and work-related determinants of a poor work ability in the workforce. 2 METHODS 2.1 Identification of the studies Relevant articles were identified by means of a computerized search of the bibliographical databases PubMed January 1985–December 2006, and Web of Science January 1988–December 2006. The following search string was used: “work ability”. The search was restricted to studies published in the English language. The literature search identified 337 abstracts with 124 abstracts in both databases, resulting in 213 unique abstracts. 2.2 Selection Studies were excluded when (1) not using the work ability index to describe work ability in an occupational population, and (2) not presenting quantitative information on associations between 15

individual and work-related factors and work ability. There remained 17 publications that met our selection criteria (Aittomaki et al., 2003, Eskelinen et al., 1991, Fischer et al., 2006, Goedhard et al., 1998, Kaleta et al., 2006, Laitinen et al., 2005, Monteiro et al., 2006, Nygard et al., 1991, Pohjonen 2001, Pohjonen 2001, Pranjic et al., 2006, Punakallio et al., 2004, Sjogren-Ronka et al., 2002, and Tuomi et al., 1991, 2001, 1997, and 2004). One publication was included after an additional search in personal archives of the authors (Martinez and Latorre Mdo, 2006). Since two publications reported results of both a cross-sectional study and a follow-up study, in total 20 studies were included. 3 RESULTS 3.1 Individual characteristics The demographic factor most studied was age (seven studies). Four out of seven studies (57%) reported a decreased WAI with older age (Goedhard et al., 1998, Monteiro et al., 2006, Pohjonen, 2001, and Tuomi et al., 1991), two studies (29%) demonstrated no association (Martinez and Latorre Mdo, 2006, and Tuomi et al., 1997) and one study found a higher risk for a poor WAI among younger workers (Fischer et al., 2006). All four studies on poor musculoskeletal capacity reported a significant association with a poor WAI with risk estimates varying from 6.4 to 9.1 (Eskelinen et al., 1991, Nygard et al., 1991, Pohjonen, 2001, and Sjogren-Ronka et al., 2002). One out of three studies found a positive association for a poor cardio respiratory fitness expressed by maximum oxygen uptake (VO2 max) (Goedhard et al., 1998). Poor functional balance in home care workers was associated with poor WAI (Pohjonen, 2001), whereas this association was not observed among fire fighters (Punakallio et al., 2004). Both studies on cognitive performance showed no significant associations (Eskelinen et al., 1991, and Nygard et al., 1991). Lack of leisure-time physical activity was associated with a lower WAI in four out of five studies. Overweight was positively associated with a poor WAI in four out of six studies. In one study smoking was associated with lower WAI (Tuomi et al., 1991), whereas in two studies no significant association was found (Kaleta et al., 2006, and Tuomi et al., 2001). 3.2 Work-related characteristics A large variety of psychosocial factors at work were addressed, varying from poor management to satisfaction with supervisor. Five out of seven studies (71%) reported a positive association between high mental work demands and a poor WAI (Pranjic et al., 2006, Sjogren-Ronka et al., 2002, and Tuomi et al., 1991, 2001 and 2004). Regarding autonomy (i.e. poor possibilities of job control, lack of freedom in work tempo) three out of four (75%) studies reported an increased risk for poor WAI with lack of autonomy (Tuomi et al., 2001, 1997 and 2004). High physical demands, such as increased muscular work, and poor work postures were positively associated with a lower WAI in four out of seven (57%) studies (Pohjonen, 2001, and Tuomi et al., 1991, 1997 and 2004). Three out of seven studies (43%) could not find any association between high physical work demands and poor WAI (Aittomaki et al., 2003, Fischer et al., 2006, and Tuomi et al., 2001). Regarding the physical work environment, two out of four (50%) studies reported a lower WAI with thermal discomfort and poor physical climate (Tuomi et al., 1991, and 2001), whereas another two studies did not find any association (Fischer et al., 2006, and Tuomi et al., 1997). 4 DISCUSSION Important factors for a decreased work ability were lack of leisure-time vigorous physical activity, poor musculoskeletal capacity, older age, overweight, and high physical and psychosocial work demands. 16

For individual determinants the range in magnitude of associations was larger in cross-sectional studies than in longitudinal studies. A cross-sectional study design is more sensitive to bias, which may explain the larger differences in risk estimates. For some determinants the available number of studies was too small to draw meaningful conclusions; gender (two studies), education (two studies), mental performance (two studies). Despite the large differences in definition of the determinant and the validity of the measurement techniques applied, the studies showed a consistency in important determinants for WAI high mental work demands, poor autonomy, and high physical work demands. All work related determinants were measured by means of self-report. This assessment technique may lead to some spurious results because subjects with a poor WAI may overestimate their physical and mental work load in the workplace relative to those with an excellent WAI. It is unclear if an objective measurement of the work demands would show similar results.

5 CONCLUDING REMARKS Research on work ability is strongly focused on risk factors for poor work ability and more often on individual than work related factors. Important factors for health promotion in aging workers would be increasing leisure-time physical activity, prevention of overweight, increasing musculoskeletal capacity and decrease of physical and psychosocial work load. REFERENCES Aittomaki, A., Lahelma, E., and Roos, E., (2003). Work conditions and socioeconomic inequalities in work ability. Scandinavian Journal of Work Environment & Health 29: pp. 159–165 Eskelinen, L., Kohvakka, A., Merisalo, T., Hurri, H., and Wagar, G., (1991). Relationship between the SelfAssessment and Clinical-Assessment of Health-Status and Work Ability. Scandinavian Journal of Work Environment & Health 17: pp. 40–47 Fischer, F.M., Borges, F.N., Rotenberg, L., Latorre Mdo, R., Soares, N.S., Rosa, P.L., Teixeira, L.R., Nagai, R., Steluti, J., and Landsbergis, P., (2006). Work ability of health care shift workers: What matters? Chronobiology International 23: pp. 1165–1179 Goedhard, W.J., Rijpstra, T.S., and Puttiger, P.H., (1998). Age, absenteeism and physical fitness in relation to work ability. Stud Health Technol Inform 48: pp. 254–257 Kaleta, D., Makowiec-Dabrowska, T., and Jegier, A., (2006). Lifestyle index and work ability. International Journal of Occupational Medicine and Environmental Health 19: pp. 170–177 Laitinen, J., Nayha, S., and Kujala, V., (2005). Body mass index and weight change from adolescence into adulthood, waist-to-hip ratio and perceived work ability among young adults. International Journal of Obesity 29: pp. 697–702 Martinez, M.C., and Latorre Mdo, R., (2006). Health and work ability among office workers. Revista Saude Publica 40: pp. 851–858 Monteiro, M.S., Ilmarinen, J., and Corraa Filho, H.R., (2006). Work ability of workers in different age groups in a public health institution in Brazil. International Journal of Occupational Safety and Ergonomics 12: pp. 417–427 Nygard, C.H., Eskelinen, L., Suvanto, S., Tuomi, K., and Ilmarinen, J., (1991). Associations between Functional-Capacity and Work Ability among Elderly Municipal Employees. Scandinavian Journal of Work Environment & Health 17: pp. 122–127 Pohjonen, T., (2001). Age-related physical fitness and the predictive values of fitness tests for work ability in home care work. Journal of Occupational and Environmental Medicine 43: pp. 723–730 Pohjonen, T., (2001). Perceived work ability of home care workers in relation to individual and work-related factors in different age groups. Occupational Medicine-Oxford 51: pp. 209–217 Pranjic, N., Males-Bilic, L., Beganlic, A., and Mustajbegovic, J., (2006). Mobbing, stress, and work ability index among physicians in Bosnia and Herzegovina: Survey study. Croatian Medical Journal 47: pp. 750–758 Punakallio, A., Lusa, S., and Luukkonen, R., (2004). Functional, postural and perceived balance for predicting the work ability of firefighters. Int. Arch. of Occup. & Env. Health 77: pp. 482–490

17

Sjogren-Ronka, T., Ojanen, M.T., Leskinen, E.K., Mustalampi, S.T., and Malkia, E.A., (2002). Physical and psychosocial prerequisites of functioning in relation to work ability and general subjective well-being among office workers. Scan. J. of Work Environment & Health 28: pp. 184–190 Tuomi, K., Eskelinen, L., Toikkanen, J., Jarvinen, E., Ilmarinen, J., and Klockars, M., (1991). Work Load and Individual Factors Affecting Work Ability among Aging Municipal Employees. Scandinavian Journal of Work Environment & Health 17: pp. 128–134 Tuomi, K., Huuhtanen, P., Nykyri, E., and Ilmarinen, J., (2001). Promotion of work ability, the quality of work and retirement. Occupational Medicine-Oxford 51: pp. 318–324 Tuomi, K., Ilmarinen, J., Martikainen, R., Aalto, L., and Klockars, M., (1997). Aging, work, life-style and work ability among Finnish municipal workers in 1981–1992. Scandinavian Journal of Work Environment & Health 23: pp. 58–65 Tuomi, K., Ilmarinen, J., Jahkola, A., Katajarinne, L., and Tulkki, A., (1998). Work ability index. In: Health, F. I. o. O., (Ed.), Helsinki. Tuomi, K., Vanhala, S., Nykyri, E., and Janhonen, M., (2004). Organizational practices, work demands and the well-being of employees: a follow-up study in the metal industry and retail trade. Occupational Medicine-Oxford 54: pp. 115–121

18

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Systematic review for assessing job demands and physical work capacity in safety jobs Annina Ropponen Institute of Biomedicine, Physiology/Ergonomics, University of Kuopio, Kuopio, Finland

Juhani Smolander ORTON Research Institute and ORTON Orthopaedic Hospital, Helsinki, Finland

Veikko Louhevaara Finnish Institute of Occupational Health, Kuopio, Finland

ABSTRACT: Introduction – A number of physical demanding jobs include unavoidable health and safety risks. The aim of this review is to consider worker selection and the assessment by the job-related tests for the prerequisites of physical work capacity in safety jobs. Material and methods – The literature search identified references from Medline, and ISI Web of Science databases. First the titles were assessed for relevance and then abstracts by one of the authors. Then the articles were evaluated for the assessment of physical work demands in safety jobs. There were 41 articles fulfilling the selection criteria of fire fighter and security guard jobs and 11 of the articles had relevance and quality to be presented this paper. Conclusions – The selection of safety workers based on physical performance and job-related tests seems to guarantee an acceptable level of physical fitness up to middle-age. Both individual and societal measures are needed to keep the physical work capacity at the acceptable level for the age of 55–60 years. The discrimination of the tests due to gender and age could be avoided in physically demanding high-risk jobs if the tests are relevant and valid with respect to the actual job demands.

1 INTRODUCTION There are a number of jobs such a fire fighter, a paramedic worker, a police officer, and a security guard whose physical demands include unavoidable health and safety risks. Physically demanding safety jobs often require heavy dynamic muscle work, manual handling of materials and static muscle work in poor or dangerous work environment. Typically also the intensity of the jobs may reach near maximal or maximal levels in an unpredictable manner (Lusa, 1994). In these high-risk safety jobs a worker with a low physical work capacity is considered to have a high individual risk (Taylor and Groeller, 2003). Therefore, the workers need to have capacities to control or cope with the risks in order to avoid overstrain and to improve safety at work for themselves and others (Shephard and Bonneau, 2002). The necessity of the tests for the assessment of the prerequisites of physical work capacity is commonly agreed, and the use of tests has been required by employers, workers and civil right organization (Jackson, 1994, Shephard and Bonneau, 2002, and Taylor and Groeller, 2003). Matching the work to the worker needs to be based on reliable and accurate analysis of the critical and important elements of work. There is, thus, a need to investigate if the published studies can provide the basis for methods of assessment of job demands and physical work capacity. 19

The aim of this review is to consider worker selection and the assessment by the job-related tests for the prerequisites of physical work capacity in safety jobs. The focus is on the selection of healthy individuals with various capacity and/or job-related tests, which assess physical performance or various dimensions of physical work capacity. The examples of safety jobs include a fire fighter and a security guard.

2 METHODS A systematic review was made including studies published until March 2006. The literature search identified a total of 573 references from Medline, and ISI Web of Science databases. First the titles were assessed for relevance for assessing job demands and physical work capacity in safety jobs and then abstracts by one of the authors (AR). Based on the relevance of the abstracts, fulltext articles were gathered and 81 articles were identified of safety jobs including jobs of fire fighters, paramedic workers, police officers and security guards. The articles were evaluated for the worker selection and the assessment by the job-related tests for the prerequisites of physical work capacity in these four different safety jobs. Particularly, the selection of healthy individuals with various capacity and/or job-related tests by assessing physical performance or various dimensions of physical work capacity was evaluated for the selection of the articles. There were 41 articles fulfilling these selection criteria of fire fighter and security guard jobs which are presented as an example in this paper. In respect with the 41 articles of fire fighters and security guards, 11 articles were determined having relevance and quality to be included in this paper.

3 RESULTS 3.1 Job demands of fire fighters Fire fighters meet high physical and psychological demands in operative tasks. In different firefighting and rescue operations with fire protective clothing and a self-contained breathing apparatus (SCBA) such as smoke-diving, fire-suppression, ladder climbing, rescuing a victim, dragging a hose, and raising a ladder, the mean oxygen consumption levels were 2.1–2.8 l/min (Louhevaara, 1985, Lusa, 1994, and Holmér and Gavhed, 2007). During peak loads the mean oxygen consumption was 3.6–3.8 l/min and heart rate 179–180 beats/min in young and healthy fire fighters (Lusa, 1994, and Holmér and Gavhed, 2007). While the aerobic demands of fire fighting tasks are well documented, the strength and motor coordination demands are less well quantified. The biomechanical features were studies in a simulated rescue-clearing task requiring the lifting of a 9-kg power saw (Lusa et al., 1991). In the task the mean dynamic compression force at the disc of L5/S1 was 6228 N. The mean peak torque for the back extension was 242 Nm and for knee extension 120 Nm. The peak values corresponded to over 90% of their maximal isokinetic muscle strengths. The results showed that lifting and handling of a heavy power saw produced a high load on musculoskeletal system. Other typical fire fighting tasks having high demands on muscle strength and endurance are carrying, pulling, pushing and dragging. Regarding to motor coordination of fire fighting tasks, Punakallio et al. (2003) showed that the use of fire-protective clothing and SCBA in particular, significantly impaired both postural and functional balance, and more negatively the balance of older fire fighters than that of younger ones. The authors recommended that valid balance tests should be developed for fire fighters. 3.2 Tests of physical work capacity for fire fighters Several studies have recommended that a fire fighter should have a maximal oxygen consumption of 2.7–3.0 l/min and/or 34–45 ml/min per kilogram of the body mass (Lemon & Hermiston, 1977, 20

Table 1. Tests and their classification for assessing male and female fire fighters’ cardiorespiratory\fitness and muscular performance. Classification Test

Poor

Moderate

Good

Excellent

VO2 maxa (l/min) ml/min/kg Bench press (45 kg) (reps/60 s) Sit-up (reps/60 s) Squatting (45 kg) (reps/60 s) Pull-up (max reps)

2.4 29 9 20 9 2

2.5–2.9 30–35 10–17 21–28 10–17 3–4

3.0–3.9 36 18–29 29–40 18–26 5–9

4.0 50 30 41 37 10

a

Bicycle-ergometer or treadmill test, Maximal oxygen consumption = VO2 max.

Louhevaara et al., 1994, and Holmér & Gavhed, 2007). In Finland, the Guide for Smoke-diving (1991, 2002) gives a recommendation for a four-stage scale to classify fire fighters’ maximal oxygen consumption (Table 1). Male and female fire fighters at the age range of 20–63 years are considered to have a sufficient cardiorespiratory capacity for smoke-diving tasks when they attain a result of 3.0 l/min and/or 36 ml/min/kg in the tests for the maximal oxygen consumption (Lusa, 1994, and Holmér & Gavhed, 2007). The demands of fire fighting and rescue tasks on muscle strength and endurance have not been quantified adequately. The tests of muscular performance in Table 1 have mainly designed to followup the fire fighters’ physical work capacity in Finland. These tests are also used for the selection of applicants for the training courses of the Emergency Service Institute of Finland. In the selection of applicants for basic training courses, the required minimum level has been between “good” and “excellent” in each test. A job-related test drill has been developed for Finnish fire fighters (Louhevaara et al., 1994). It simulates the physiologically heaviest work activities often required by smoke-diving operations, and the purpose is to assess the circulatory strain during the tasks with heart rate measurements. The test drill with fixed maximal working time of 14.5 min consists of five activities (walking and carrying, climbing/ascending stairs, hammering, moving over and under bars, and hose rolling) done with full personal protective clothing and SCBA of 25.5 kg. The test drill is instructed to be carried out using a habitual work pace within the fixed time. This is exceptional as compared to other job-related tests or test drills that are advised to perform as fast as possible (e.g., Shephard and Bonneau, 2002). 3.3 Job demands of private guards There are no scientific reports available in ergonomics or work physiology literature about the job demands and physical work capacity of security guards. In Finland, the number of security guards is about 7,000, and every fourth of them is a woman. Usually security guards are young, and their length of the career often remains quite short. However, the guards meet risks and dangers in the street such violent clients usually due to abuse of drugs, unpredictable and poor work conditions and situations, shift work and night work as well as shortcomings in their protective equipment and weapons. During the work shifts the most common operative tasks of security guards are the catching of a client by running and controlling him or her by using arms and trunk. (Laine et al., 2004). In the job of the security guards there are demands for muscle strength and endurance, cardiorespiratory capacity, and dynamic balance of the body (Laine et al., 2004). 3.4 Tests of physical work capacity for private guards There are no commonly approved health or physical capacity qualifications for the security guards. Laine et al. (2004) suggested that the assessment of the security guards’ physical work capacity 21

would include two job-related tests. Firstly, a 1.5-mile walking-running test with the work clothing and equipment of 7 kg that consists of light protective clothing, a bullet vest, a weapon belt and protective boots. The performance time and heart rate are registered in the test. According to the time and heart rate the average and maximal oxygen consumption is estimated and related to mass of the body and equipment. The performance is accepted when the estimated maximal oxygen consumption is at least 36 ml/min/kg, which is also used as the qualified criteria for fire fighters’ smoke-diving operations (Lusa, 1994). The second job-related test is a functional (dynamic) balance test (Punakallio et al., 2003). According to Laine et al. (2004) suitable physical work capacity tests for evaluating security guards’ muscle strength and endurance are the sit-up test and the pull-up test The time limit is 60 s, and the accepted performance requires at least 38 sit-ups/min. The pull-up test is done with a reverse grip without the time limit. The acceptable performance requires at least 6 pull-ups. The minimum limits require “good” performance according to the classification for fire fighters (Lusa, 1994).

4 DISCUSSION The selection of safety workers based on physical performance and job-related tests seems to guarantee an acceptable level of physical fitness up to middle-age. Both individual and societal measures are needed to keep the physical work capacity at the acceptable level for the age of 55–60 years. The discrimination of the tests due to gender and age could be avoided in physically demanding high-risk jobs if the tests are relevant and valid with respect to the actual job demands. The fire fighters’ job demands on physical work ability are occasionally very high during the entire occupational career. Their cardiorespiratory and muscular fitness should be tested regularly for guaranteeing adequate work capacity for physically extreme operations. The use of fitness tests is necessary in the planning and carrying out preventive measures for maintaining health and work ability of fire fighters. A number of valid test batteries are available for the assessment of fire fighters’ physical work capacity (Lusa, 1994). According to Laine et al. (2004) the test battery developed for assessing the security guards’ physical work capacity is relevant and feasible both in the pre-employment screening and in the follow-up of the physical work capacity. The set “acceptable” limits of the tests are very strict and are needed to be modified when older (over 30 years) and female guards are evaluated. The tests have been used for two years in a small enterprise of security guards and the obtained results and experiences have been positive. When the physical performance or job-related tests are relevant they should guarantee an acceptable work performance in physically demanding jobs that are equal for men and women with all ages. This leads on the situation that the passing of tests is easier for men than women, and for younger than older individuals. For instance, about a half of the Finnish fire fighters aged over 50 years have serious problems to pass all physical fitness tests, and presently only one female fire fighter is carrying out operative fire and rescue tasks. On the other hand, there are many female security guards that carry out operative tasks in the street.

5 CONCLUSIONS The selection of safety workers based on physical performance and job-related tests seems to guarantee an acceptable level of physical fitness up to the age of 40 years. After that there should be alternative respectable career paths for older workers who could not pass the tests. The reasons for the negative test outcome usually relate to various disorders and diseases, which prevent regular physical exercise needed for maintaining sufficient physical fitness in the middle-aged and older safety workers. Intensive and regular fitness training alleviates the evident decline of cardiorespiratory and muscular fitness due to age, and may maintain the physical work capacity at the acceptable 22

level for the age of 55–60 years. After that most of the safety workers have problems to meet the high physical job demands in operative tasks. REFERENCES Guide for Smoke-Diving, (1991, 2002). Finnish Ministry of Internal Affairs. Helsinki Holmér, I. and Gavhed, D., (2007). Classification of metabolic and respiratory demands in fire fighting activity with extreme workloads. Applied Ergonomics, 38: pp. 45–52 Jackson, A., (1994). Pre-employment physical evaluation. Exercise and Sports Science Reviews, 22: pp. 53–90 Laine, K., Kolehmainen, M. and Louhevaara, V., (2004). Development of the promotion of work ability in an enterprise of security guards. In: Action program for safety occupations, edited by Lusa, S. and Louhevaara, V. (Helsinki, Finnish Institute of Occupational Health), pp. 33–44 (in Finnish) Lemon, P.W. and Hermiston, R.T., (1977). Physiological profile of professional fire fighters, Journal of Occupational Medicine, 19: pp. 337–340 Louhevaara, V., Soukainen, J., Lusa, S., Tulppo, M., Tuomi, P. and Kajaste, T., (1994). Development and evaluation of a test drill for assessing physical work capacity of fire fighters, International Journal of Industrial Ergonomics, 13: pp. 139–146 Lusa, S., Louhevaara, V., Smolander, J., Kinnunen, K., Korhonen, O. and Soukainen, J., (1991). Biomechanical evaluation of heavy tool-handling in two age groups of firemen. Ergonomics, 34: pp. 1429–1432 Lusa, S., (1994). Job demands and assessment of physical work capacity of fire fighters, University of Jyväskylä, Studies in sport, physical education and health, Jyväskylä (doctoral dissertation) Punakallio, A., Lusa, S. and Luukkonen, R., (2003). Protective equipment affects balance abilities differently in younger and older firefighters, Aviation Space and Environmental Medicine, 74: pp. 1151–1156 Shephard, R. and Bonneau, J., (2002). Assuring gender equity in recruitment standards for police officers. Canadian Journal of Applied Physiology, 27: pp. 263–295 Taylor, N. and Groeller, H., (2003). Work-based physiological assessment of physically-demanding trades: a methodological overview. Journal of PhysiologicalAnthropology andApplied Human Science, 22: pp. 73–81

23

Chapter 2 Work Ability Index

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

The place of age in organisational policymaking: Evidence from an Australian qualitative survey Libby Brooke, Patricia Healy, Joyce Jiang & Philip Taylor Business Work and Ageing Centre for Research, Swinburne University of Technology, Victoria, Australia

ABSTRACT: This report presents findings of qualitative research in four case study organizations as part of a larger study, Redesigning Work for an Ageing Society (RW4AS) Project, being undertaken by Business Work and Ageing Centre for Research, Swinburne University. The study will identify factors transforming the nature of work and posing risks to the work ability of an ageing workforce, examine existing policies and practice in managing workforce ageing in public and private organizations and test the applicability of the Work Ability model to Australian circumstances. The research found that changes engendered by globalization shaped the organization of work and the pressures experienced by individual workers. All the organizations function in a globalised, highly competitive market, are adopting new technologies and are increasingly reliant on work intensification and casualisation of labour. They had not come to terms with the need to retain their ageing workers as manifested by short term policy responses rather than proactive strategies. The areas in which ageing workers experienced difficulties, such as their competence with new technologies, changing work structures and occupational health risks indicate alignments between work ability domains and managing the vulnerabilities of ageing Australian workers in changing organizational environments.

1 INTRODUCTION The Redesigning Work for an Ageing Society (RW4AS) Project aims to adapt the Work Ability model to the Australian context within an Australian Research Council funded demonstration project carried out in four case study organisations. The project is conducted by the Business Work and Ageing Centre for Research, Swinburne University of Technology, in collaboration with industry partners, including public and private sector case study organisations and three government agencies interested in workforce ageing and work ability. The project aims to identify factors transforming the nature of work and posing risks to the work ability of an ageing workforce, examine existing policies and practice in managing workforce ageing in public and private organisations, and test the applicability of the work ability model to Australian circumstances. The demonstration project will also design, implement and evaluate workplace interventions to promote work ability and support the continued employment of older workers. This paper reports findings of a qualitative survey of key informants in the case study organisations.

2 METHOD In the initial stages of the RW4AS Project, a qualitative survey of company-selected key informants was conducted in the four case study organisations – the freight handling division of a large airline, a manufacturing firm, the roadside assistance division of a motoring organisation and a national 27

university. All have a large proportion of older workers with many years of employment in the company. The sample of 59 informants comprised senior and operational managers, specialist OH&S, human relations and training staff, employee representatives and six patrol staff from the motoring organisation. Interviews generally lasted 40–60 minutes and used a structured questionnaire to explore perceptions of the impact of workforce ageing on the company’s capacity to meet productivity challenges and opportunities, the company’s responses to the identified risks of employing ageing workers, and attitudes towards older workers. Interviews were transcribed and analysed using NVivo7.

3 FINDINGS This paper focuses on changes in work organisation in the four case study organisations. As a result of new globally driven dynamics these organisations are undergoing change processes which present ongoing risks to their ageing workforces. In each case the organisation of work and the pressures experienced by individual workers are shaped by globalisation. Interrelated processes involve shifts towards more intensive work practices and new forms of work organisation, particularly to casualised workforces. At the same time the organisations have retained ageing workforces which are more vulnerable to certain kinds of risk, including physical and psychosocial stress. 3.1 Case study organisations Case study 1 – Airline freight terminals: The two international air freight terminals have an ageing permanent full-time workforce (average 44 years) of 380 (220 warehouse and 160 clerical staff), with long tenure (average 10–15 years), and very low staff turnover (2%) despite a series of redundancies over the last 10 years. Warehouse employees are all male, work in small teams and rotate tasks. The permanent full-time workforce is supplemented by casual, part-time employees working five hour shifts at peak times. Base salary rates are relatively low but supplemented by overtime and shift penalty rates, which form a significant part of take-home pay. The mainly female clerical workforce, includes permanent full-time and part-time staff supplemented by casual employees as required. Senior managements are expected to work long hours, including extended periods on-call (up to 24/7). The workforce is strongly unionised – with around 90% union membership. Production extends over 24 hours as the company operates across global working time zones. Most employees work rotating 8-hour shifts on rosters dictated by international airline schedules – with often considerable overtime when there are unscheduled delays. Shift rosters are designed to minimise fatigue but are often changed to accommodate personal needs and unexpected work demands and there is no cap on overtime. Although much of the work is mechanised, using cranes, forklifts and motorised trollies, there is a lot of manual handling involved in packing and loading freight in accord with strict safety and quarantine regulations and aircraft requirements. As a consequence, manual handling related injuries are the most common. Older and less physically able permanent employees are given preference for lighter manual work and plant operation, with the heavier work more frequently allocated to casual staff. The firm is in transition between industrial era work practices and changing human performance criteria to improve competitiveness in global markets. Management practices have changed significantly since the company was privatised in the mid-1990s, with introduction of measures such as merit based promotion, performance management, more external recruiting and increasing management awareness of the need for career management over the life course and managing ageing workers’ injuries and health. Appropriate management of people resources was frequently identified as a continuing key organisational challenge within a context of pressures to contain costs and improve customer service in order to maintain position in a dynamic and highly competitive global market. 28

Case study 2 – Manufacturing company: Over the last decade this long established company has been through a series of takeovers, mergers and acquisitions and is now part of an international firm with interests across New Zealand and Australia. It manufactures building materials and components and sells on the domestic market and exports to Asian manufacturers of home and commercial componentry – much of which returns as cheap imports to supply the Australian market. In the domestic market the company is a major supplier to small manufacturers of home and commercial fittings, which are increasingly challenged by the same cheap imports. The company places high value on its ascendancy in the local marketplace which is becoming increasingly insecure and particularly open to challenge by global firms practicing vertical product integration which cut across its componentry. In response to these challenges it is repositioning its market role and adapting its production and marketing of products and services. The company has an Australian workforce of over 2,000, predominantly male, in 13 production plants and multiple distribution centres, many in rural areas. Employees have an average age of 42 years and over 10 years employment with the company. Most employees work rotating shifts of 8–12 hours on 24/7 or 24/5 rosters. The 12 hour shifts are unlikely to change, despite management opposition, due to workers’ preference for working fewer days. There is a high performance culture throughout the company with a strong emphasis on staff training and career development and commitment at all levels to good OH&S practice. Overall the company was seeking to maintain its competitive advantage though lean staffing and high performance utilisation of its workforce, and was concerned that its ageing workforce could potentially constrain this broader agenda. Workforce risks which applied particularly to ageing workers were identified as long hours of work, manual handling injuries, exposure to hazardous substances and injuries related to plant operation. Case study 3 – Motoring organisation: The roadside assistance service has been a major part of the motoring organisation for over 100 years but now faces declining demand for services due to the improved design and reliability of vehicles. The organisation is restructuring its workforce through attrition and by shifting to contractors, who are flexibly employed to decrease workforce down time. There are currently 119 male, patrol staff – a 19% reduction over two years due to retirement, voluntary redundancy and replacement by subcontractors, who now provide 60% of roadside assistance services. All are trained mechanics with an average age of 49 years and an average 16 years employment with the organisation. The job requires both the technical skills to diagnose and fix problems in a large range of vehicles and well-developed communication and people management skills to negotiate with people in often difficult and stressful roadside situations. The job involves eight or nine hour rotating shifts on an 18/7 roster (06.00 h – midnight). The work is largely sedentary (long in-vehicle periods of driving and waiting for jobs) but interspersed with tasks involving high physical demands (lifting, applying force, bending, static postural loading) performed in variable and often poor working environments (bad weather, poor lighting, hazardous roadside locations) and with limited access to assistance. This creates high risk of musculoskeletal injuries. As all work communications are via computer, staff are also socially isolated with little opportunity for interaction with colleagues. The shift to maintaining a “balance” of contract staff with a “stable core” of permanent staff – no permanent patrol staff have been recruited for over eight years – suggests that management had not resolved problems connected with the ageing of its workforce. These include physical risks and demands, the changing composition of the job and the transition to more autonomous work structures with lower supervision and peer group support as it moves from core staff to subcontractors. Case study 4 – Multi-campus university: The university was established in the early 1990s with several small campuses and a focus on teaching, particularly in the areas of education and nursing. It has approximately 13,000 students and a workforce of around 1200 (70% female), including 950 permanent staff and a fluctuating number of casual staff which has been steadily increasing over the last decade. The increasing reliance on a casual/sessional workforce is largely a cost containment strategy in response to the decline in government funding, on which the university is heavily reliant. 29

The average age of staff is 47 years, with over 80% of academic staff and 50% of general staff older than 45 years and 50% of new recruits over 50 years. Changes in priorities for government funding and student demand are now forcing the university to rapidly adopt a stronger research focus and to increase on-line courses and teaching. This requires the established teaching staff to take on new, additional research roles and adopt new teaching methods while maintaining their existing workloads. The conflicting task priorities and intensification of work places increasing demands on the ageing workforce. A major organisational issue identified by respondents was the tension between retention of the ageing workforce and recruitment of younger academics more attuned to the new teaching and research environment. Whilst the university has introduced a program of phased retirement to encourage retention of older staff, it is not well supported by programs to preserve their health and wellbeing and manage the stress resulting from their high workload and conflicting work priorities of research and teaching. The combination of an increasingly casualised workforce structure, more conflicting work priorities and increasing workloads, has increased stress and insecurity, particularly amongst older staff. 3.2 Common issues Although they are located in different sectors with different risk profiles, the four case study organisations share characteristics that exemplify the contemporary workplace and increasingly drive labour requirements in ways that generate significant risks for the work ability of all workers, particularly older workers. All function in a globalised, highly competitive and changing market, and have developed a strong emphasis on growing productivity and rapidly adapting to changing market demands. They are all actively responding to and adopting new technology and are increasingly reliant on work intensification and casualisation of labour to maintain productivity, reduce labour costs and increase workforce flexibility. In all four organizations, global market demands and new technology are directly driving changes in work practices. For example, the working hours and shift rosters at the international air freight terminals are directly influenced by the speed necessary to keep pace with global freight and logistics operations. Similarly, in response to increased competition from cheaper Asian and European imports, the manufacturer is repositioning its market role – moving up the supply chain, developing new products and automating production – and in the process restructuring workforce deployment. An interrelated process of workforce restructuring, is also leading to more casualisation. The university is increasingly relying on casual staff to introduce skills needed to meet increased student demand for on-line courses deliverable globally. And new vehicle technology has reduced demand and changed the working conditions of roadside patrol staff. As a result of these combined pressures the organisations’ workers are functioning in a moreor-less constantly changing environment where long-established work practices are reshaped, and traditionally valued competencies and skills are becoming outmoded. There is increased pressure to work longer and more flexible hours, update competencies and learn new skills – often in employees’ own time. At the same time traditional job security is being undermined by a growing reliance on contingent workers to provide a cheap and flexible workforce better able to meet immediate production needs, while outsourcing labour costs and risks. This combination of dislocation, work intensification and job insecurity is both creating new OH&S risks and exacerbating old ones – particularly for older workers. 3.3 Perception of older workers The majority of respondents claimed a marked dissonance or mismatch between company labour requirements and the capacities and skills of older workers. Although there was considerable uncertainty about the immediate and future implications of workforce ageing, all respondents gave a predominantly negative assessment of older workers. Older workers were stereotyped as having a negative impact on the organisation’s capacity to meet new challenges and opportunities. Older workers were also perceived to be more expensive – engendering higher risks and costs – and less productive because of their reduced functional capacity and inability or reluctance to develop new 30

skills and adapt to new working conditions. The increased costs and risks were largely attributed to greater vulnerability to injury, more workers compensation claims and payouts, longer and more difficult rehabilitation after injury or illness, and increased sick leave. Yet there was also some acknowledgement that older workers had lower turnover rates and were generally regarded as more reliable with greater experience, knowledge and life-skills, and as holding valuable corporate knowledge. Several respondents also noted that older workers’increased sick leave was frequently planned and therefore less disruptive that the unplanned one-day ‘sickies’ more prevalent among younger workers – particularly when shift work is involved. Generational differences in attitudes to work were also noted by several respondents, with younger workers characterised as more demanding and less committed and reliable. 3.4 Responses to workforce ageing Perceptions of older workers reflected many commonly held negative stereotypes – although many stressed the effect of individual differences and referred to positive exceptions. Nevertheless, in part their views reflected existing problems which were poorly addressed at company level and managed mainly at local or operational level and often in a relatively ad hoc manner. A central theme was that the companies perceive and respond to workforce ageing as essentially an issue of better managing the exit of older workers – not as part of a more comprehensive workforce planning strategy. Generally responses are reactive rather than proactive, fragmented rather than systematic. For example, despite clear identification of risk factors for older workers, there were no age-specific OH&S policies or programs in place to avoid injuries and maintain their functional capacity. Similarly, retirement programs do not include formal arrangements for transfer or handover of skills and corporate knowledge and succession planning at other than senior levels is largely dependent on the initiatives of local operational managers. In large part the companies address the mismatch or misalignment between their current ageing workforce and their labour requirements in ways that meet their immediate needs – mainly through use of overtime or casual/contract labour to fill the gaps. There is relatively little effort to retrain and redeploy older workers except at operational level, where reasonable accommodation is frequently made on an ad hoc basis, particularly for older workers with long employment, but only as long as it does not significantly reduce productivity. 4 CONCLUSION The findings indicate both the need for and a lack of comprehensive, structurally based interventions, focusing on all the work ability domains, to cope with pervasive changes in work environments. In all the case study organizations there is considerable uncertainty about the implications and appropriate responses to workforce ageing yet respondents identified many issues requiring responses. In all four organizations the work environment is changing in ways creating ongoing misalignments with ageing workforces and indicating the need for long-term career management across the life course to ensure a continuing supply of workers with competencies appropriate to new technologies and restructured workplaces. Yet management responses currently tend towards the short-term, with increasing reliance on work intensification and casual labour to maintain productivity in the face of growing shortages of skilled labour. As yet there are no age-specific programs or policies in place to address the identified risks and better manage health and injuries across traditional OH&S boundaries. REFERENCES NVivo7. QSR International Pty Ltd, 2nd Floor, 651 Doncaster Road, Doncaster Victoria 3108 Australia. National Occupational Health and Safety Commission 2005. Surveillance Alert. OHS and the Ageing Workforce.

31

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Establishment of a Work Ability Index network in Germany Hans Martin Hasselhorn, Bernd H. Müller, Rainer Tielsch & Benita Gauggel University of Wuppertal, Wuppertal, Germany

Gabriele Freude Federal Institute for Occupational Safety and Health, Berlin, Germany

Jürgen Tempel Occupational Health Practice, Hamburg, Germany

ABSTRACT: In times of an ageing work force sustaining the work ability becomes increasingly important for a society. In 2003, the German Federal Authorities have therefore initiated the development of a National Work Ability Index (WAI) network with the aim to promote the use of the WAI in Germany. The article describes design, tasks and performance of the German WAI network. Main tasks are the development of (i) a network of WAI users, (ii) a WAI data base, and (iii) a WAI user software. In a critical review of the past 5 years of network performance, the authors identify a range of success- and non-success factors. While, for example, the strong increase of public interest in the WAI belongs to the success factors, the identification of a wide spread misconception of the WAI instrument (misunderstood as indicator for ‘health’) leads to substantial future challenges for the network.

1 INTRODUCTION In times of an ageing working population sustaining the work ability becomes increasingly relevant and can even be vital for a healthy society. Here, the Work Ability Index questionnaire (WAI) can play a central role in work place health promotion by: • promoting the discussion of “work & age”, • identifying risk factors, risk groups and individuals at risk, and • evaluating the effect of interventions. The question is, whether and how the use of the WAI can be promoted in a large country. The German Federal Authorities (Federal Institute for Occupational Safety and Health, Berlin) have therefore assigned this task to a research group which has founded a national WAI-Network. In this contribution, tasks, process and results of the WAI-Network are discussed. 2 DEVELOPMENT AND PERFORMANCE OF THE WAI-NETWORK The German WAI-Network was set up in 2003/4 by the “Work Group Empirical Work Research” (“Arbeitsgruppe Empirische Arbeitsforschung”, AEA) at the University of Wuppertal. Several additional organisations and individuals were included in a project council for sharing expertise. The WAI-Network has set three aims: • to develop a network of WAI users, • to collect a WAI data base, and • to develop a WAI user software. 33

2.1 WAI user network Having the limitation of own financial resources in mind, the organisers of the network understood their role as one initiating knowledge exchange between the participants rather than steering communication. As of October 2007, the German WAI-Network has 171 members, almost half of them representing enterprises (occupational health physicians, ergonomists, and work organisation psychologists). About 20% of the members are scientists and 10% work in health- or occupational health insurances. Membership is free of charge. The degree to which within group communication occurred is not easy to determine, but it may be estimated to be low in relation to the large amount of communication with the network organisers which turned out to occur in the coming years. In the course of the project the interest in the WAI and the use of the instrument has increased substantially. Interest in the WAI-Network was dependent on the degree of networks’ dissemination activities which have been high in the course of the project and included oral and poster presentation at conferences, numerous WAI presentations, teachings, seminars and workshops and the edition of 14 articles. Regular newsletters have kept the network informed about activities and the website has provided the public with access to the WAI instrument, the underlying concept, information material and an online WAI assessment possibility. Soon after the start of the project it became evident that a mere translation of the WAI reference publication (Tuomi et al., 1998) did not suffice for the use of the WAI in Germany. This has resulted in a more detailed guide book (Hasselhorn and Freude, 2007). In addition, the request for practical examples was met by the publication of a booklet with 16 detailed examples from Germany, Austria and Switzerland (BAuA, 2007).

2.2 WAI data base The aim of a national WAI data base is to provide the users (individuals and enterprises) with valid reference scores. Valid reference scores are regarded to be of high relevance for the practical use of the WAI. Since no representative assessment of the German working population could be made, the network organisers had to collect as many (anonymous) WAI data as possible from different groups assessing WAI in Germany. As of October 2007, the network has collected cross sectional WAI data of about 8500 individuals. This data base has an uneven distribution of professions (esp. too few manual workers) and is therefore far from being representative for the German working population. Each group submitting a WAI data base receives a feedback of own results in relation to those of relevant occupational groups stratified by age (Figure 1).

45

Metal workers (low qualif.) Safety engineers

42 WAI mean

Leaders (metal industry) Nurses

39

Nursing aids

36

Rescue workers Teachers

33

NEW data: teachers Priests

30

Figure 1.

30

45 45 Age groups (years)

Example for feedback of WAI results to a group submitting new teacher WAI data.

34

2.3 WAI user software The network has developed a “WAI software” for the use in enterprises (Figure 2). It has the following features: • • • • •

WAI data entry (long and short version). WAI analysis (for individuals and for groups). Results feedback (individuals and groups). WAI benchmarking (individuals and groups). WAI longitudinal assessment (individuals and groups), and WAI data export.

The software is free for all members of the network. Since 2007 an English version is available. 3 CRTITICAL REVIEW At the end of the first project phase (2003–2006) the German WAI-Network has been evaluated very positively. Positive aspects may have been the following: • the large degree of dissemination activities initiated and performed by the network, • the substantial increase in public awareness and interest in the WAI in Germany, • the collection of profound expertise in the network core group resulting in teachings and in consultation of medium sized and large companies (free of commercial interests). • the development of the WAI software, and • the activation of further research and prevention activities adjacent to the work ability (e.g. the validation of the short WAI version, the initiation of the development of an equivalent assessment instrument for unemployed).

Figure 2. WAI software, version 2.3.1.

35

Another positive aspect has been a very controversial public debate about usefulness, limits and risks of the use of the WAI in Germany. Whereas the Germany unions have had a central role in promoting the WAI in Germany, a high ranked group of the German metal workers’ union (IG Metall) has had some major concerns about the instrument, questioning its’ validity, confidential use and worrying about the potential for misuse. The resulting debate has forced the WAI network organisers to in detail review the instrument and its’ use in Germany. This process has provided the organisers with an even more differentiated view on the instrument, considering both strengths and weaknesses and preventing the network from an undifferentiated “marketing” of an assessment instrument. Below, some problems of the network performance are listed: • Data collection of external WAI data was rather difficult. Joint efforts to perform a central WAI assessment of a representative sample of the working population may be more efficient. • Currently, the network does not have a good overview of the use of the WAI in Germany. • There is a lack of experience and knowledge about WAI in intervention studies in Germany. • It turned out a half academic post (with student support) was not adequate for keeping up the networks activities. • It may be difficult for scientists heading a WAI network to keep a scientific distance to the instrument and the topic. 4 OUTLOOK The network activities have been extended until 2009. It was considered that a continued support and guiding of the public use of the WAI was necessary. As it has turned out in the past years, the shortness and simplicity of this screening instrument is misleading. Many users believe to assess “health” or “physical or mental capacity”, instead it needs to be emphasised that “work ability” represents another concept based on the individual, his/her work and the interaction of both (this view in turn has implications for preventive action). In addition it may be necessary to – to higher degree – encourage the users to not only focus on the WAI instrument but to emphasise the underlying WAI concept which provides insight in starting points for preventive action. Another future activity of high necessity will be to promote research in the use of the WAI in intervention studies and in adjacent areas such as in rehabilitation and for long term unemployed. It is another future aim of the network is to promote international cooperation and to internationally address some validity questions. Many validation studies origin from older samples in Finland, additional new validation findings from other countries may be useful in the future. The most challenging aim will be to make the network independent of external funding by the year 2009. Since paid teaching and consultation activities (e.g. insurances, Occupational Health Services) are increasing, this aim may be realistic. ACKNOWLEDGEMENT The first phase of the German WAI Network (2003–2006) was funded by the Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, BAuA. The current activities are funded by the Initiative Neue Qualität der Arbeit, INQA, by the German Ministry of Labor. REFERENCES BAuA (Eds), (2007). Why WAI? – Der Work Ability Index im Einsatz für Arbeitsfähigkeit und Prävention – Erfahrungsberichte aus der Praxis. (BAuA, Dortmund, Berlin, Dresden) Hasselhorn, H.M., and Freude, G., (2007). Der Work Ability Index – ein Leitfaden, Schriftenreihe der Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, (NW Verlag Bremerhaven). Tuomi, K., Ilmarinen, J., Jahkola, A., Katajarinne, L., and Tulkki, A. Work Ability Index (2nd ed.). Helsinki: Finnish Institute of Occupational Health 1998.

36

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

WAI among workers in SMEs at Wholesale, Fruits, Vegetables and Flower Market in Brazil – from research to action Inês Monteiro Faculty of Medical Sciences, State University of Campinas – UNICAMP, Campinas, Brazil

Kaija Tuomi, Juhani Ilmarinen, Jorma Seitsamo, Eva Tuominen & Heleno Rodrigues Corrêa-Filho Finnish Institute of Occupational Health, Helsinki, Finland

ABSTRACT: The objective of this study was to evaluate the worker’s work ability, work characteristics and life style, which lead of towards the comprehension of certain consequences for health and well being related to the work environment. A cross-sectional study was carried out in Wholesale and Flower Market area with around 1,000 micro and small sized companies, including shops, inspection and cleaning services, administrative sector and autonomous porters. A questionnaire containing socio-demographic data, life-style, health and work aspects, living and work conditions and the Work Ability Index was administered. The sample included 1,006 workers. The male population represented 86.9% of the workers (range from 15 to 73 years old), and the mean age was 33.5 years (SD = 12.1). Young women had poor work ability than young men, while the opposite result was found in relation to older women and men. Work breaks had a positive correlation and “to have a work accident during the last year” and related risks/hazards at work (lifting and transporting heavy weight, time pressure, tiredness and stressful job) were negatively correlated with work ability. In the life style model, leisure time, physical activities, number of hours of sleep and “sleeping well” were correlated with work ability. The study indicated that work conditions are quite important in relation to work ability and should be considered when planning workplace health promotion and intervention actions. Keywords: Work Ability Index, SMEs, work conditions.

1 INTRODUCTION The International Labour Organization presents a positive perspective regarding the potential generation of decent jobs in small and medium businesses (ILO, 2003). Small and micro sized companies had been playing an important role in the economic development in many countries. In the late nineties, there were about 2.7 billions workers, with one billion employed in small businesses and one million self-employed in agriculture, as reported by Rantanen (1999). The businesses productivity became a fundamental factor in in the survival of many countries, directly affecting their market competitiveness. This has been a basic concern in some countries such as in Scandinavia and Europe, where short, medium and long-term actions have been developed. The costs of sick leave, precocious retirement due to invalidity, and work accident are important not only for the institution and the workers, but also for the country, as they interfere with the productivity, increase health expenses and interfere with workers’ lives outside of work (Monteiro-Cocco, 2003). The worker health system is structured only for the formal market, whose contingent has decreased, thus creating a gap on the provision of health care services for workers. 37

The Constitution and Brazilian law obligate the State assure this right. However, the existent service structure is not designed to comply with it. The issue is even more aggravated when dealing with large workplaces controlled by private companies, such as the Wholesale and Flower Market (Ceasa Campinas) in Brazilian cities, including the one located in Campinas, a large city in São Paulo State. What is the main issue regarding work ability and the ageing workforce in developing countries? Ilmarinen (2006) reported that the main problem concerned to the maintenance of the work ability, is due to the fact that sometimes workers begin working before legal age and, depending on the work sector are subjected to difficult situations related to work conditions and work environment. The country has a young workforce and the challenge is how to maintain workers in good conditions until the age of retirement, and at the same time, take care of older workers. The objective of the study was to evaluate the work ability, work characteristics and life style among workers from small and medium sized enterprises.

2 METHODS This study is part of a large public policies research project entitled “Basis for the implementation of worker’s health public service in informal work and SMEs of Wholesale and Flower Market in São Paulo State- Brazil” (Monteiro, 2005). A cross-sectional study was carried out in a large wholesale produce and flower market area consisting of around 1,000 micro and small sized companies, including shops, inspection and cleaning services, administrative sector and autonomous porters. A questionnaire with socio-demographic, life-style, health and work aspects, living and work conditions (Monteiro, 1996) and the Work Ability Index (Tuomi et al., 1997) was administered. The random sample included 1,006 workers and the response rate was 85%. Statistical analysis was performed in SAS 9.1 with general linear models. The research was approved by the Ethics Committee of the Faculty of Medical Sciences – State University of Campinas – UNICAMP.

3 RESULTS AND DISCUSSION The male population represented 86.9% of the workers (range from 15 to 73 years old), and the mean age was 33.5 years (SD = 12.1). The majority was married (58.4%) and 37.4% were engaged in physically, 35.8% mixed (physically and mentally) and 26.8% mentally demanding work. Women presented higher frequencies in the poor/moderate work ability categories (2.3 and 9.9% respectively) than men (0.6 and 7.6% respectively), using the chi-square test (p-value = 0.0002). Figure 1 shows decreased work ability of women aged down 40 years, when compared with men of the same age; and within the men’s group, work ability was lower for older workers. Figure 2 shows the work ability categories among men, women and all workers. The poor category was more frequent among young women and for young and old men. The mean WAI for women was 40.9 points (SD = 4.7) and mean age was 32.8 years (SD = 11.0), and for men, the mean WAI was 42.6 points (SD = 4.4) and mean age was 33.6 years (SD = 12.2). The gender difference was similar to found by Torgen (2005) in her research. Although difference were found in relation to age brackets and gender, with good results for older women and poor results for older men, it is important to consider the healthy worker effect described by Checkoway et al. (2004). Work breaks were positively correlated with work ability. Having a work accident during the last year and related risks/hazards at work (lifting and/or transporting heavy weigh, time pressure, tiredness and stressful job) had negative correlation with work ability in the general linear model. In the life style model, leisure time activities, physical activities, number of hours of sleep, and sleeping well were correlated with work ability in the general linear model. 38

WAI

44 43 42 41 40 39 37 36 35 34

Men 20

20–29

30–39

Women 40–49 Age

50–59

All 60–69


70

Figure 1. WAI scores by age groups at Wholesale vegetables, fruits and flower market. WAI by age group and sex

All all

Age and sex

Men

Women

Poor Moderate Good Excellent

50 or  40–49 30–39 20–29 20 All 50 or  40–49 30–39 20–29 20 All 0%

20%

40%

60%

80%

100%

WAI categories

Figure 2. WAI categories by age group and gender.

The results give support to design and to initiate intervention actions in the wholesale, fruits vegetables and flower market.

4 CONCLUSIONS The study showed that work conditions were quite important in relation to work ability and should be considered when planning workplace health promotion and intervention actions. Actions related to intervention should ongoing in the workplace related to basic information of health and safety aspects (ergonomic issues, sleep counselling, and nutrition). REFERENCES Checkoway, H., Pearce, N., and Kriebel, D., (2004). Research methods in occupational epidemiology. 2nd. (New York: Oxford). Ilmarinen, J., (2006). The ageing workforce – challenges for occupational health. Occupational Medicine, 6: pp. 362–364.

39

International Labour Office. ILO, (2003). Infocus Programme Boosting Employment through Small Enterprise Development – SEED. [Accessed in 2003 Oct 01]. Available on: . Monteiro-Cocco, M.I., (2002). Capacidade para o trabalho entre trabalhadores de uma empresa de tecnologia da informação. [Livre Docência]. Campinas (SP): Universidade Estadual de Campinas – UNICAMP. [Work ability among information technology workers]. Monteiro, M. I., (1996). revisado em 2004. Instrumento para coleta de dados sociodemográficos, trabalho, saúde e estilo de vida. [Socio-demographic, lifestyle, work and health questionnaire]. (Campinas: Universidade Estadual de Campinas). Monteiro, M. I., (2005). Basis for the implementation of worker’s health public service in informal work and SMEs of Wholesale and Flower Market in Sao Paulo State- Brazil. Public Policy Research Project – Phase II. (Campinas: State University of Campinas). Rantanen, J., (1999). Future perspectives in developing high-quality occupational health services for all. In: Good occupational practice and evaluation of occupational health services, edited by Lehtinen, L., (Helsinque: FIOH), pp. 32–45. Torgen, M., (2005). Experiences of WAI in random sample of the Swedish working population. In: Assessment and promotion of work ability, health and well-being of ageing workers, edited by Costa,G, Goedhard, W. J. A., Ilmarinen, J., (Amsterdam: Elsevier, International Congress series 1280), pp. 328–332. Tuomi, K., Ilmarinen, J., and Jahakola, A. et al., (1997). Índice de capacidade para o trabalho. [Work Ability Index]. (Helsinki: Finnish Institute of Occupational Health).

40

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Factors associated to the work ability among Brazilian teachers ∗ Tatiana Giovanelli Vedovato & Inês Monteiro Faculty of Medical Sciences, State University of Campinas – UNICAMP, Campinas, Brazil

ABSTRACT: The aim of this research was to evaluate the work ability of school teachers and its relation to health, individual characteristics and working conditions. A cross-sectional study with a sample of 258 teachers (fundamental education and high school) from nine public schools was performed in two cities of Sao Paulo state. A standardized self-applied instrument in order to assess work ability and social- demographic data, life style and work status questionnaire and the Work Ability Index was utilized. Statistical analysis was performed with SAS® and tested associations for logistic regression unvaried models and logistic multivariate. Sample was composed in the majority by females (81.8%), with age ranging from 20 and 65 years (Mean 41.9 SD = 9.4) and most of them with university level (95.7%). Work ability was considered as good by less than half (42.6%) of them and as moderate/poor by 35.3%. The most significant factors correlated with WAI moderate/poor categories were: female gender (p = 0.0481), health comparison to other people at the same age as being worse or equal (p < .0001), medication use (p = 0.0013), sleep less than six hours per night (p = 0.0341), bad sleep at night (p < .0001) and total working time as teacher or more then equal to 20 years (p = 0.0153). These data point the need of improvements in the working conditions of the teachers aiming at to the recovery and maintenance of the ability with respect to the work of these professionals. Keywords: Work Ability Index, teachers, health.

1 INTRODUCTION The teachers’ professions were really important for the development and culture of each country. Finish study showed that teachers had decreased work ability correlated with high mental demand at school, with time pressure, to take care of students leading towards psychosomatic diseases as reported by Parkatti, Kinnunen and Rasku (1999). The goal of this article is to evaluate the work ability of public school teachers and correlated with health, individual profile and work conditions. 2 METHOD A cross-sectional study was carried out in was realized in two cities – Campinas and São José do Rio Pardo in nine fundamental level school and/or high school. The random sample was composed by 258 teachers and data collection was performed from August until December in 2005. The research was approved by the Ethics Committee of the Faculty of Medical Sciences – State University of Campinas – UNICAMP, and all data was obtained with the agreement of the workers and school management. ∗ Grant

from CAPES and CNPq – Brazilian Federal Research Agencies

41

Female

Sex

Male Age bracket 30 (years) 30-49 50 and Marital status

 Married Single Divorced Widower

Kids

Yes No 0

10

20

30

40

50

60

70

80

90

%

Figure 1.

Distribution of teachers in relation to socio-demographic characteristics, 2005.

The Work Ability Index (Tuomi, Ilmarinen, and Jahakola et al., 1997) and socio-demographic questionnaire (Monteiro, 2002) was utilized for data collection. For statistical analysis the SAS® – 8.02 was performed and the significance level adopted was p < 0.05.

3 RESULTS The female population represented 81.8% of the workers and the mean age of all workers was 41.4 years (SD 9.4). Figure 1 shows the socio-demographic characteristics distribution among the subjects. The overwhelming majority teachers had university graduation course’s (95.7%), and 15.1% had post-graduation course (specialization, master degree or doctorate). The WAI average among teachers was 38.6 points (SD = 5.8), and women had poor values of WAI – 38.2 points (SD = 5.2) when compared with men 40.3 points (SD = 5.6) (Mann-Whitney test p = 0.0481). Figure 2 shows the work ability categories distribution. When compared WAI and self-perception about health compared with people with the same age 40.7% of the teachers that considered their health rather poor or very poor they had moderate work ability (chi-square test p < 0.0001). In relation to life style 93.4% had at least one leisure time activity and 56.6% did physical exercise regularly; and alcohol intake was referred by 27.5% of the teachers. Work conditions as noisy environment, repetitive movement was considered by the workers as tiredness and source of stress. Pain during the last six months was related by 62.1% of the teachers, pain during the last week by 51.6% and medicine use by 50.8%. Teachers that use medicine had poor WAI value when compared with no users (Mann-Whitney test p < .0001). Diseases with physician’s diagnosis was related by 82.2% and in own opinion by 71.3% of the teachers. The women double journey’s (tasks at school and at home) probably reduce time available to take care of yourself (leisure activities, to sleep). 42

Work ability among teachers Poor Moderate Good Excellent 0

10

20

30

40

50

%

Figure 2.

Distribution of work ability categories among teachers, 2005.

Perhaps the decreasing in WAI was associated with psychosocial factors, stress and musculoskeletal diseases as related in German research with teachers as reported by Freude, Seibt, Pech and Ullsperger (2005).

4 CONCLUSION The study limitation was correlated to cross-sectional design. Teachers that were in sick leave related to diseases didn’t participate on the research. The majority of teachers were women, working more then 15 years in this profession. A gender difference was in relation to work ability with poor values among women when compared to men. The authors found that work ability decrease among older teachers and with more time working as teachers. Is fundamental to improve and restore the work ability of teachers intervening on work environment, work conditions and life style. REFERENCES Checkoway, H., Pearce, N., and Crawford-Brown, D. J., (1989). Research methods in occupational epidemiology. New York: Oxford University Press. Freude, G., Seibt, R., Pech, E., and Ullsperger, P., (2005). Assesment of work ability and vitality – a study of teachers of different age groups. Anais International Congress Series 1280: pp. 270–74 Monteiro-Cocco, M. I., (2002). Capacidade para o trabalho entre trabalhadores de uma empresa de tecnologia da informação. Campinas, [tese – Livre Docência]. Campinas, SP, Universidade Estadual de Campinas. [Work ability among information technology company workers – Free docence thesis] Parkatti, T, Kinnunen, U., and Rasku, A., (1999). Work, well-being and health among ageing teachers. In: J. Ilmarinen, V. Louhevaara. (Ed.). Finn Age. Respect for the ageing. Helsinki: FIOH, pp. 163–171 Tuomi, K., Ilmarinen, J., and Jahakola, A. et al., (1997). Índice de capacidade para o trabalho. [Work Ability Index]. (Helsinki: Finnish Institute of Occupational Health).

43

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Factors that predict work ability: Incorporating a measure of organisational values towards ageing Josephine Palermo, Lynne Webber, Kaye Smith & Anne Khor School of Psychology, Deakin University, Melbourne, Australia

ABSTRACT: This research conducted in an Australian public sector organisation aimed to identify the main factors that predict work ability for employees. According to Ilmarinen’s (1999) model of work ability, an individual’s work ability is influenced by their general health, attitudes, values and motivation interacting with workplace and other environmental demands. However what is unknown is the influence of value incongruence (i.e. the lack of fit between individual and organisational values), particularly when that incongruence results in age discrimination. This is important in an Australian context where youth and symbols of youth are over-valued in business environments and where older workers themselves perceive age discrimination as the single most important cause of early exit from the labour force. 109 participants completed a survey about work ability. Differences between work ability and health were not found between older and younger workers suggesting that strategies for improving work ability could be targeted at all employees rather than just older employees. However there were significant differences found between older and younger workers on reasons that would influence employees to stay longer in the organisation. Older workers tended to be more influenced by the provision of less demanding work, and positive attitudes towards older workers. Younger workers tended to be more influenced by opportunities to be employed in another section of the organisation, skills training opportunities and career advancement opportunities. Results from hierarchical regression analyses suggested that good physical and mental health, and low occupational stress related to workplace culture were significant predictors of increased work ability. Results also suggested that occupational stress is likely to decrease with: high work ability and work satisfaction; and high value congruence. Implications for wellbeing programs to include the development of targeted organisational values are discussed.

1 INTRODUCTION It is well known that the Australian population is ageing (Kryger, 2005). Indeed, due to falling mortality and fertility rates, Australia was one of the world’s most rapidly ageing workforce. This is especially true for the female workforce (ABS, 2004). In addition, Australia has a relatively high incidence of “involuntary” early retirement with 46 percent of people in the 50–64 age group unemployed and 33 percent of 50–64 year olds relying on some form of social security payment (Sheen, 2001). Kryger (2005) predicts that unless older workers are encouraged to stay in the labour force longer, labour shortages will occur. He suggests that this could have a slowing effect for the economy along with a decline in productivity. The Council on Ageing (Sheen, May 2001) supports this view suggesting that Australian living standards could be adversely affected without appropriate public policy interventions addressing solutions to the ageing workforce. Organisations may be required to re-examine their human resources policies and working environments to facilitate and 45

accommodate the increased participation of older workers in the workforce (The Economist 2006; Rappaport et al., 2003) and to address the negative stereotypes that appear to hinder older workers (The Economist, 2006). Negative stereotypes surrounding older workers have been well documented (Taylor and Walker, 1998; Platman and Tinker, 1998; Chiu et al., 2001; Brooke, 2002). However, despite the fact that most reviews of the relevant literature have concluded that there is little or no relation between age and psychological resources and competencies (Sneed and Whitbourne, 2005), they remain ingrained in modern Australian society (McGregor and Gray, 2002; Platman, 2003). Sneed and Whitbourne suggested that researchers should shift their focus away from loss and decline in ageing to an approach focussed on growth and gain in later life. In line with this view, attention has turned to a focus on understanding and improving older workers’ “work ability” (Ilmarinen, 2001). Work ability refers to ones subjective or objective assessment of the ability to do the work that needs to be done. There is evidence that improving work ability in older workers can help to keep workers in the workplace and reduce levels of early retirement (de Boer, van Beek, Durinck, Verbeek & van Dijk, 2004; Pohjonen, 2001; Tuomi, Huuhtanen, and Nykyri and Ilmarinen, 2001). 1.1 Factors affecting work ability Research conducted in Finland, shows that work ability is multidimensional with complex interactions between the dimensions of work, lifestyle, aging and health (Tuomi et al., 1997). The Finnish Institute of Occupational Health coined work ability as a process of resources in relation to work (Ilmarinen, 2004). They defined the individual’s human resources as comprising: health and functional capacities (physical, mental and social); education and competencies; values and attitudes; and motivation. According to Ilmarinen and colleagues these variables interact with a complex set of organisational factors involving: the work demands (physical and mental); the work community and management; and the work environment. The model predicts that work ability is dependent on both the organisational factors and the individual’s human resources, including: (1) rating of current ability to do the work compared to lifetime best; (2) ability to do the work in relation to the physical and mental demands of the work to be done; (3) number of current diseases; (4) estimated work impairment due to diseases; (5) sick leave during last 12 months; (6) own estimate of work ability in two years time; and (7) mental resources available. As can be seen there is a heavy reliance on physical health (diseases, work impairment due to diseases and sick leave) rather than on psychological health and wellbeing (e.g., coping strategies, optimism). Although work ability has traditionally been measured by data from experts (e.g., occupational therapists), more recently the Finnish Institute of Occupational Health (FIOH) (1998) found that self-perceptions of work ability and own estimate of work ability in two years time are valid measures of work ability. Age and work ability. There is considerable evidence that work ability declines with age. Ilmarinen (2001) found that work ability especially in relation to perceived functional and mental capacity, can begin to decline after the age of 45 years. Age declines are commonly found for employees in physically demanding work such as cleaning. For example, Pohjonen (2001) found significant age-related declines in work ability at 35 years with a second larger decline at 55 years of age in a group of home care workers. Not surprisingly, health related factors such as chronic health issues (de Boer, 2004); workers’ health; and functional capacity, including physical, mental and social capacity (Chan et al., 2000) have all been found to be related to work ability. Self-rated work ability has been found to be predictive of the duration of sick leave (Reiso, Nygård, Brage, Gulbrandsen and Tellnes, 2001). As expected poor health, physical work demands and lack of control over work all predict poor work ability (Pohjonen, 2001; Hopsu, Lepanen, and Ranta & Louhevaara, 2005). In contrast, a ‘healthy’ lifestyle such as physical activity has also been found to be positively related to work ability (Seitsamo & Ilmarinen, 1997). Good work ability has been found to be related to good quality of work and job satisfaction (Tuomi et al., 2001). In addition, Webber, Smith and Scott (2006) found that job satisfaction and organisational satisfaction were related to self-rated work ability. It is not clear whether job satisfactions leads to higher work ability 46

or higher work ability leads to better job satisfaction. Work ability has been found to be important for predicting early retirement or failure to return to work after a prolonged illness (de Boer et al., 2004). Webber et al. (2006) found that self-rated work ability was a better predictor of plans to leave than age or job and organisational satisfaction. Factors that promote work ability. There is evidence that even older workers can improve their work ability by increasing physical activity and their positive social relationships at work (Tuomi, Ilmarinen, Martikainen, and Aalto and Klockars, 1997). Other factors that appear to be related to work ability are socio-economic status (Aittomaki et al., 2002; the level of professional competence; and work environment factors such as type of work, work satisfaction and work conditions (Tuomi et al., 2001; de Boer et al., 2004). Work ability may be mediated by physical and psychological stress. Salonen et al. (2003) found a strong correlation between stressors, chronic disease, stress symptoms and work ability. Stress outcomes are an imbalance between demand perceived load versus perceived capability (Lazarus & Folkman, 1984). Perceived demand also includes the anticipation of adverse consequences arising from failure to cope with demand. Cotton (1996) argued that the reporting of stress is an outcome of a more interactive and dynamic system of variables that cannot be reduced to the linear model. Taken together the research on stress suggests that stress is multidimensional with its effects influenced by individual differences. Factors requiring consideration in a model of occupational stress include personality style, occurrence of events, perceived support, concurrent personal pressures and other organisational factors (Cotton, 1996). Folkman and Lazarus (1984) also suggested that antecedent conditions such as motivation (for example values, commitment, goals) and beliefs about self are important factors. More recently research on stress has incorporated these variables in multidimensional designs (see Hart, Wearing, and Griffin, 1996; Code and Langan-Fox, 2001). Much of the stress literature, while emphasising individual personality and organisational characteristics as being determinants of stress, relate these only to characteristics that are associated with ‘work’. Mitchell (1996) argued that employment is more than ‘work’, and therefore the individual’s deeper levels of psychological functioning should be incorporated in models of stress. He argued that an individual’s value structure, unconscious processes, sense of meaning (and therefore self-concept), are also important factors. According to Code and Langan-Fox (2001), evidence suggests that organisational constraints that prevent goal attainment have a negative impact on individual well being, thereby adversely impacting on stress vulnerability. Therefore, congruence between goals and behaviour, and motives and behaviour may be fundamental to the experience of occupational stress. Code and Langan-Fox (2001) suggest a personality integration model of stress with dual emphasis on goal progress and attainability, and unconscious need fulfilment. This involves implicit-explicit congruence whereby stressors increase with increased discordance between implicit and explicit motives. These tenets can be applied to individual-organisational motives, extending traditional person-environment (P-E) fit theories to levels of individual and organisational system dynamics. Edwards (1996) advocates a supply value model of stress that accommodates these differences. Stress is viewed as a mis-fit between an individual’s personal values and the environmental contingencies or supplies available to fulfil those values. It follows then that congruence between the structure and organisation of personality at the individual level, and culture at the organisational level, may also be important. This may explain why reporting of stress is more likely to occur when psychological distress is higher than usual and when morale is lower than usual (Cotton, 1996). Following this argument, the extent to which the organisational culture is perceived to value older workers may be an important determinant of stress as well. 1.2 Expanding the work ability model As the previous review has shown, work ability is influenced by individuals’ general health, attitudes, values and motivation interacting with workplace and other environmental demands. However what is generally unknown is the influence of organisational cultural processes. This may be particularly important in an Australian context where youth and symbols of youth are 47

over-valued in business environments (Sheen, May 2001). According to research conducted by the Council on Ageing (Sheen, 2001), older workers themselves perceive age discrimination as the single most important cause of early exit from the labour force. Age discrimination denotes a system of entrenched beliefs and practices include: stereotyping of mature age people as less adaptive and productive than young people; marketing images concentrated on youth; and industry restructuring targeting mature workers as soft targets for redundancy and retrenchment (Sheen, 2001). Evidence has been found that people use double standards when assessing employees for promotion (Smith and Webber, 2005) and when evaluating job candidates (Richardson, Webber, Smith and Webb, 2006). In both cases older adults were judged more harshly than younger adults. Furthermore, evidence of age discrimination in various areas of employment has been found in both blue and white-collar sectors of a diverse range of organisations, and data supporting its existence has been found in numerous developed countries (Adams 2002; McGregor cited in McGregor and Gray 2002; Cant et al., 2001; Henkens 2000; Platman and Tinker 1998; Still and Timms 1998; Boerlijst, 1994). Negative attitudes and age discrimination not only render older adults vulnerable and susceptible to redundancy if the company changes direction (Boerlijst, 1994), but also perpetuate the myth that older workers lack the required motivation and ability to participate in training (Taylor and Walker, 1998). A quantitative study conducted in New Zealand found evidence of discriminatory practices in relation to training (McGregor, 2001 cited in McGregor and Gray, 2002). Negative stereotypes regarding the inability of older adults to adapt to technological change were found to be particularly pervasive. If older workers are not encouraged or afforded the opportunities to participate in training and skill development they eventually become de-motivated and lack the enthusiasm required not only to inspire younger workers but also maintain their value and relevance within their respective organisations (Boerlijst, 1994). Although motivation and values has been incorporated into the workability model (Ilmainen and Rantanen, 1999), further contributions could be made by including a measurement of value congruence (between personal and organisational values). Palermo (2005) found that congruence based on values associated with gender tolerance to be a determinant of job satisfaction and occupational stress. It is therefore likely that similar patterns may be ascertained by an investigation of ‘fit’ between individuals and their organisation on values related to age tolerance and/or discrimination, especially for older workers. This study aimed to examine the influence of individual and organisational factors that impact on work ability. More specifically it also aimed to test the following hypothesis: Lack of fit or incongruence between individual and organisational values will significantly predict work ability and occupational stress.

2 METHOD 2.1 Participants A total of 109 participants completed the survey, 66 percent were women and 34 percent were men. The majority of participants were Australian (71%), with 25 percent identifying with European ethnicities, and another 4 percent identifying with South Asian ethnicities. The sample had an average chronological age of 47 years. However, most participants felt younger than their chronological age. To analyse differences between younger and older workers, the sample was split into two groups: participants 44 and younger and those participants 45 and older. This age grouping was chosen because it is consistent with most other research. Participants in the younger age group comprised 36 percent of the total sample, with 64 percent being in the older age group. The mean tenure for participants was 8 years of service with the Council, and 10 years of service in their current job. Whilst on average, participants indicated that they were planning to leave in 7.7 years, there were no significant differences between older and younger workers on their plans to leave the Council. 48

2.2 Measures A seven-page questionnaire was designed in consultation with the industry partner to measure self-rated work ability, health and well-being. The questionnaire comprised six main sections as described below. 1. Work ability, health and emotional well-being: comprised self report items from the Work Ability Index (WAI) (adapted from Finish Institute of Occupational Health 1998). A work ability score was derived by aggregating scores on responses about participants’ current work ability to do physical and mental demands of their work and projected work ability in two years time. A combined score comprised of: mean responses to Q1 of the WAI (1 = workability at it’s worst; 10 = workability at its best) weighted according to whether their job was physically demanding, mentally demanding, or both; and mean responses obtained from the question ‘in terms of your current health will you be able to you current job (1 = unlikely, 2 = not certain, 3 = relatively certain). Health and emotional well-being was assessed using five of the seven items on the SF-12 (Medical Outcomes Trust, 2002). This survey is a standard health survey uses to assess mental and physical health over a four-week recall period. It has been extensively validated across a number of research studies. Self Efficacy was measured using items the General Self Efficacy Scale Sherer, et al., 1982), based on Bandura’s (1977) theory was used to assess general expectancies of self-efficacy. The scale consisted of 15 items rated on a 7-point likert scale where high scores indicate high selfefficacy. Sample items included: ‘when I set important goals for myself, I achieve them’ and ‘when I decide to do something I go right to work on it’. High internal consistency (α = .86) and criterion validity has been reported (Sherer et al., 1982; Long, 1989). The scale had similar internal consistent in this study (α = .86): 2. Work satisfaction. Participants were asked a series of questions relating to satisfaction with their work and job. These included items about satisfaction with: opportunities for career advancement, supervisors and co-workers, work tasks and job role, and level of autonomy and appreciation received. After analyses that showed that these items were highly correlated, a Work Satisfaction score was calculated with every participant receiving a score that was the average across all items. One item, “the way my job provides steady employment” was not included in the scale because scale reliability statistics showed it was not reliable. 3. Occupational Stress. Sources of occupational stress were assessed using 19 items adapted from Occupational Stress Indicator (OSI) scales (Cooper, Sloan and Williams, 1988). Three scales were produced: • Workplace Culture: characterised by lack of encouragement and guidance from supervisors; Lack of social support by people at work; Favoritism and Covert discrimination; Inadequate or poor quality of training; Staff shortages and unsettling turnover rates; and Lack of power and influence. (Cronbach’s alpha = 0.91, 10 items) • Work/Life Balance: characterised by demands of work in private, social and family life; having too much work to do and mundane administrative tasks. (Cronbach’s alpha = 0.82, six items) • Role Anxiety: characterised by over promotion; Keeping up with new technologies, ideas, technology, innovations and difficulty coping with stress. (Cronbach’s alpha = 0.64, three items) 4. Personal and organizational values. Participants were asked to rate their personal values and those of their respective organisation (items were adapted from Palermo 2005). The individual values were examined to identify any clusters. Analyses revealed that three clusters emerged (respect and Support, collective achievement and independent achievement). • Value Scale 1: Respect and Support (Cronbach’s alpha = 0.91, four items): characterised by being treated respectfully, fair treatment of staff, valuing employees; and support when needed. • Value Scale 2: Collective Achievement (Cronbach’s alpha = 0.88, seven items):characterised by values such as honesty, accountability, compassion, taking responsibility, and integrity. 49

• Value Scale 3: Independent Achievement (Cronbach’s alpha = 0.89, four items):characterised by creativity, ambition, independence, and recognition of good performance. Importance ratings for personal values and organisation values were added to produce personal and organisational scales. The scales were then rescored to a 0–100 scale to enable ease of interpretation. A difference score was then computed (P-E: Personal scale score value – Organisational value scale score) 5. Reasons to stay with current employer. Participants’ were asked to rate the characteristics they believed would facilitate staying with their current employer. Individual items that appeared to correlate were clustered into the following scales. • Reason to Stay 1: Career Opportunities (Cronbach’s alpha = 0.81, five items):characterised by opportunities for employment in another section and pay increases; skills training and career advancement opportunities; and more demanding work. • Reason to Stay 2: Employee assistance and flexibility (Cronbach’s alpha = 0.75, six items): characterised by a work environments with positive attitudes to older workers; health activities at work; retirement planning; part-time work and other flexible options; and less demanding work. • Reason to Stay 3: Management (Cronbach’s alpha = 0.90, five items):characterised by having employer’s respect; a good job match; positive supervisor – employee relations and personal support from management. 6. Demographic information. Participants provided some personal details (e.g., gender, age, and ethnicity). They were also asked about their plans for retirement and what kinds of activities they would like their organisation to provide that would enhance their general health and well-being (e.g., massage). 2.3 Procedure Over 600 participants were invited to complete the questionnaire from a municipal city council with 18 percent participation rate achieved. Each participant was provided with a paper-based copy of the questionnaire, a plain language statement and a reply paid envelope for return to Deakin University. Participants were requested to complete the questionnaire within two weeks from the date of receipt. Participation in the study was voluntary and anonymous.

3 RESULTS 3.1 Work ability Table 1 shows work ability scores for males and females and younger and older workers. There were no significant differences according to sex or age group.

Table 1. Work ability mean scores for males and females, younger and older workers and the total sample. Age Group

Sex

Mean

Std. Deviation

N

44 years and under

Male Female Total Male Female Total

11.50 11.73 11.68 11.07 11.33 11.22

1.69 1.48 1.51 2.09 1.49 1.76

8 30 38 28 39 67

45 years and older

50

Participants were asked to indicate the level of certainty (on a scale from 1 = uncertain to 3 = very certain) they had in maintaining their work ability two, five and 10 years from now. Projections of work ability were very high within a two year time frame, but tended to decline as the time frame increased (Mean Work ability: two yrs from now = 2.87, SD = 0.33; five yrs from now = 2.68, SD = 0.61; 10 yrs from now = 2.43, SD = 0.74). Older participants appeared to be less certain about their ability to work at current levels in 5 years from now (F(1,97) = 9.67, MSE = 0.35, p < .05) and 10 years from now (F(1,97) = 24.55, MSE = 0.46, p < .05). There were also significant differences between males and females on estimates of work ability 10 years from now, with males being less certain than females about their ability to maintain work at an optimum after 10 years (F(1,101) = 7.86, MSE = 0.52, p < .05). 3.2 Individual factors Participants answered a number of questions related to their physical and mental health, and self efficacy. Independent t-tests showed that there were no differences in health related measures between younger and older workers, with the exception of physical activity. Older workers indicated that on average they were involved in twice as many hours of physical activity in one week than younger workers. 3.3 Organisational factors Participants were asked to indicate which factors would influence intention to stay on in their workplace. Overall, most of the factors appeared to resonate with participants as important considerations in relation to their intention to stay with the Council. Employees’ top five reasons to stay included: positive relationships with supervisor, a good work environment, employer’s respect, a good job match, good physical health and a positive organisational cultural attitude to older workers. In contrast, a change in the type of work (more demanding or less demanding) and a change in manager did not appear to influence decision to stay. There were no differences between men and women in relation to factors that influence their intention to stay for most items. However there were differences between younger and older workers on some items. Table 2 reveals that older workers tended to be less influenced by the provision of less demanding work (F(1,94) = 7.31, MSE = 1.22, p < .02), and more by positive attitudes towards older workers than younger workers (F(1,94) = 13.39, MSE = .75, p < .02. Younger workers tended to be more influenced by skills training opportunities (F(1,94) = 4.96, MSE = 1.01, p < .05) and career advancement opportunities (F(1,94) = 5.63, MSE = 1.09, p < .05) than older workers. 3.4 Work satisfaction and occupational stress Mean scores of older and younger workers on work satisfaction were investigated. Overall, all employees agreed they were satisfied. There were no differences between different work areas.

Table 2.

Participants’ responses on health related measures.

Age Group

Skill training opportunities

Career advancement opportunities

Positive attitude to older workers

Less demanding work

44 yrs and under (N = 36)

Mean SD

4.25 0.97

4.14 1.05

3.69 0.98

2.33 1.01

45 yrs and older (N = 65)

Mean SD

3.77 1.03

3.64 1.04

4.35 0.80

2.95 1.15

51

There were no differences between males and females or between younger and older adults on job satisfaction. Analyses were conducted to ascertain whether there were any differences between older and younger male and female workers on occupational stress scores. There were no differences between males and females or between younger and older adults on any of the occupational stress scales (Stress – Workplace Culture: M = 2.33, SD = .84; Work/Life Balance: M = 2.43, SD = .85; Role Anxiety: M = 2.30, SD = .79). 3.5 Personal and organisational values congruence Participants were asked to rate the importance of a list of values, for themselves and for their organisation. There were significant differences between personal and organisational value scale scores found. Paired sample t-tests displayed in Table 3 showed that participants experienced some level of values incongruence, in that they tended to ascribe importance to a greater extent to themselves than to their organisation. Significant differences on value incongruence scales were also found between older and younger males and females (F(sex)(3, 99) = 6.27, η2 = 0.16, p < .05; F(agegroup)(3, 99) = 2.92, η2 = 0.08, p < .05;). Univariate statistics showed that significant differences were evident for incongruence related to Independent Achievement between older and younger workers (F(1,101) = 4.95, MSE = 0.19, p < .05), and between males and females (F(1,101) = 4.38, MSE = 0.17, p < .05). Means are displayed in Table 4. Younger workers, especially males indicated more incongruence. These results suggest that for younger males in particular, personal values associated with creativity, ambition, independence and recognition of good performance were rated as more important for self than for the organisation. 3.6 Factors that influence work ability A correlation matrix was produced to explore the relationships between work ability and factors related to individual antecedents and organisational outcomes. Table 5 shows that workability was Table 3. Paired sample t-tests between personal and organisational value scales.

Values Scales

Mean (N = 109)

Std. Dev

Std. Error Mean

Pair 1

Personal: Independent Achievement Org: Independent Achievement

83.57 67.78

13.07 21.60

1.25 2.07

8.08

Pair 2

Personal: Collective Achievement Org: Collective Achievement

89.16 79.18

12.67 16.92

1.21 1.62

6.56

Pair 3

Personal: Respect and Support Org: Respect and Support

90.07 76.93

13.34 21.07

1.28 2.02

6.44

Table 4.

t (df = 108)

Mean scores on Independent Achievement value incongruence (P-E) for older and younger males and females.

Age Group

Sex

Mean

Std. Dev.

N

44 years and under

Male Female Total

1.68 1.54 1.57

0.16 0.20 0.20

8 30 38

45 years and older

Male Female Total

1.53 1.48 1.50

0.19 0.20 0.20

28 39 67

52

53

1.00 −0.27∗ 0.36∗ −0.45∗ 0.33∗ −0.43∗ 0.29∗ 0.25∗ 0.14 0.28∗ 0.17 0.26∗ 0.17 −0.11 −0.18∗ −0.16

1.00 −0.09 −0.24∗ −0.20∗ 0.10 −0.05 −0.05 −0.21∗ −0.09 −0.28∗ −0.01 −0.27∗ −0.88 −0.02 0.02

B

Note: ∗ p < .05 A Work Ability B Psychological Age C Self Efficacy D Physical Health E Mental Health

A B C D E F H I J K L M N 0 P Q

A

F H I J K L

1.00 0.00 0.25∗ −0.32∗ 0.39∗ 0.25∗ 0.31∗ 0.24∗ 0.33∗ 0.25∗ 0.18 −0.12 0.01 −0.06

C

1.00 −0.30∗ 0.36∗ 0.21∗ 0.11 0.22 0.17 0.22∗ 0.14 −0.20∗ −0.16 −0.21

E

1.00 −0.55∗ −0.42∗ −0.11 −0.43∗ −0.06 −0.49∗ −0.10 0.44∗ 0.45∗ 0.41∗

F

1.00 0.52∗ 0.21∗ 0.50∗ 0.21∗ 0.47∗ 0.18 −0.43∗ −0.43∗ −0.55∗

H

1.00 0.39∗ 0.75∗ 0.20∗ 0.66∗ 0.10 −0.59∗ −0.62∗ −0.78∗

I

J

1.00 0.40∗ 0.64∗ 0.26∗ 0.66∗ 0.04 0.03 0.08

Stress – Workplace Practice and Culture Work Satisfaction Organisational Values Ratings: Independent Achievement Personal Values Ratings: Independent Achievement Organisational Values Ratings: Collective Achievement Personal Values Ratings: Collective Achievement

1.00 −0.15 −0.12 0.10 0.15 0.12 0.14 0.09 0.14 0.13 −0.10 −0.07 −0.13

D

Table 5. Correlation Matrix of workability, occupational stress, and individual and organisational factors.

N M O P Q

1.00 0.45∗ 0.85∗ 0.38∗ −0.67∗ −0.74∗ −0.63∗

K

1.00 0.30∗ −0.85∗ −0.67∗ −0.58∗

M

1.00 0.12 0.10 0.13

N

1.00 0.77∗ 0.73∗

O

1.00 0.75∗

P

1.00

Q

Personal Values Ratings: Respect and Support Organisational Values Ratings: Respect and Support Value Incongruence: Respect and Support (P-E) Value Incongruence: Collective Achievement (P-E) Value Incongruence: Indep. Achievement, (P-E)

1.00 0.34∗ 0.80∗ 0.00 0.17 0.03

L

Table 6.

Hierarchical multiple regression resultant model: Self efficacy, health, value incongruence, work satisfaction and occupational stress as predictors of work ability.

Psychological Age Self Efficacy Physical Health Mental Health Organisational Values Ratings: Collective Achievement Stress – Workplace Practice and Culture Work Satisfaction

B

St. Beta

sr

−0.00 −0.01 0.00 0.00 0.00

−0.06 −0.23∗ −0.45∗ −0.28∗ −0.02

0.21 −0.42 −0.25 −0.01 −0.20

0.01 −0.00

0.25∗ −0.09

0.07 −0.06

Note: Workability score was transformed to address negative skew and increase normality by reversing scores and taking the log10.

negatively and significantly correlated with psychological age rather than chronological age. It was also positively correlated with self efficacy, health measures, occupational stress related to workplace culture and work satisfaction. Ratings of organisational values were significantly and positively correlated with work ability rather than personal values. Value incongruence related to collective achievement was also significantly correlated with workability. Organisational variables rather than personal factors appeared to be significantly correlated with occupational stress related to work culture. In particular work satisfaction, organisational values and value incongruence measures were moderately to highly correlated with occupational stress. Variables that were correlated with dependent variables of interest were included in hierarchal regression analyses conducted. However limitations in sample size limited the amount of predictor variables able to be included. In addition the possibility of multicollinerarity was high due to high correlations between value incongruence measures and organisational ratings. Therefore only the significant organisational values ratings were included in the analyses. Given that participants had uniformly rated personal values as more important to themselves than to their organisation, organisational values ratings appeared to be proxies of value incongruence in this study. In order to investigate how factors of work ability may work together to predict work ability, a hierarchical regression equation was produced. Demographic characteristics were entered on the first step, organisational factors were entered on the second step and individual health factors on the last step. Table 6 displays significant predictors of workability at each step of building the model. In step 1, psychological age, self efficacy and individual health measures were all significant predictors of work ability (F(4,108) = 20.03, MSE = .00, p < .05). When organisational factors were included in the model, the shared variance attributable to psychological age was diluted with occupational stress related to Workplace Culture appearing to be a significant predictor of work ability (F(5,108) = 16.12, MSE = .00, p < .05). Overall the resultant model displayed in Table 6 explained 48 percent of the variance in work ability. It suggested that work ability is increased with (in order of importance): increased physical and mental health, decreased occupational stress related to workplace culture, and increased self efficacy (F(7,108) = 13.30, MSE = .00, p < .05). Due to the importance of occupational stress as an organisational predictor of work ability, a hierarchical regression model was conducted to investigate predictors of occupational stress related to workplace culture. In the first step self efficacy and mental health were significant negative predictors of occupational stress (F(2,108) = 9.47, MSE = .61, p < .05). In the second step mental health and self efficacy were no longer significant predictors, with Organisational values (Respect and Support) appearing to moderate their relationship with occupational stress (F(3,108) = 15.14, MSE = .51, p < .05). The resultant model displayed in Table 7 explained 46 percent of the variance in occupational stress (F(6,108) = 14.83, MSE = .40, p < .05) and suggested that decreased occupational stress is likely to be predicted by (in order of important): increased work satisfaction, increased work ability, 54

Table 7.

Hierarchical multiple regression resultant model: Value incongruence, work satisfaction, work ability and reasons to stay as predictors of occupational stress related to work culture.

Mental Health Self Efficacy Organisational Values: Respect and Support Work Satisfaction Work Ability Stay – Employee Asst Flex

B

Beta

s2

0.00 −0.11 −0.01

0.00 −0.07 −0.25∗

0.00 −0.06 −0.22

−0.36 −0.11 0.24

−0.32∗ −0.23∗ 0.20∗

−0.26 −0.20 0.20

Note: ∗ p < .05.

increased organisational values related to Respect and Support (i.e. less value incongruence) and decreased Reasons to Stay related to Employee Assistance and Flexibility. Taken together these regression analyses suggest that self efficacy and health along with occupational stress related to work culture appear to predict work ability for these Council workers. However whilst self efficacy remains a significant predictor of occupational stress, organisational factors rather than individual factors appear to better explain the variance attributable to occupational stress. More specifically, work satisfaction, work ability, organisational values related to respect and Support and employee assistance mechanisms appear to predict nearly 50 percent of the variance found in occupational stress.

4 DISCUSSION This exploration study aimed to examine the impact of individual and organisational values on work ability and occupational stress. Council staff who participated in this study indicated a reasonably high level of work ability; that is, on average they judged their ability to do their work was good. Whilst there were no differences between older and younger grouped workers on levels of work ability, these levels tended to decline as workers attempted to predict their work ability into the future that comprised a longer timeframe. Regression analyses revealed that individual factors, namely self efficacy and physical health, appeared to be important predictors of work ability. Findings suggested that as these resources increase, so too will work ability. This finding is consistent with previous studies that have shown similar relationships between psychological resources, mental health, physical health and workability (Chan et al., 2000; Pohjonen, 2001; Hopsu et al., 2005; Seitsamo and Ilmarinen, 1997). In relation to self efficacy and health measures, there were no differences between older and younger workers suggesting that this finding is equally important of workers across the lifespan. Two of the most important support provisions for self-efficacy are information that an individual is valued and accepted, and guidance and information that assist an individual in dealing with different situations. Perceptions of having relationships where competencies, skills, and value as persons are recognised (reassurances of worth) have been found to be conducive to building self-efficacy (Delongis, Folkman, and Lazarus, 1988). Whilst correlations suggested that increased work ability may be associated with more congruence between personal and organisational values (related to collective displays of honesty, compassion, integrity, taking responsibility etc.) this effect was mediated by occupational stress and work satisfaction in the equation. Findings suggested that increased work satisfaction increases and decreased occupational stress may be related to increased work ability, confirming previous research (e.g. Tuomi et al., 2001; Webber et al., 2006; Salonen et al., 2003). Given the importance of occupational stress as an organisational predictor of work ability, findings suggested that organisational factors rather than individual factors were important antecedents 55

of occupational stress for these workers. Again organisational values rather than personal values were important predictors, particularly those associated with values of respect and Support. In fact, reduced value incongruence with increased work satisfaction and decreased importance of employee assistance programs appeared to mediate the effects of self efficacy and mental health on decreased occupational stress. The finding that increased importance of employee assistance programs appeared to predict increased occupational stress at first appears unintuitive. It suggests that participants who indicated that employee assistance programs were more important to them, (including the presence of more positive attitudes to older workers), would also experience more occupational stress. It may be the case that these workers may have had unmet expectations in relation to these factors and therefore this may have been contributing to increased occupational stress. The presence of organisational values related to respect and Support as a negative predictor of occupational stress also seems to support this proposition. It indicates that unmet expectations may be working together with value incongruence to contribute to increased stress. This may be exacerbated by people using double standards (Richardson et al., 2006) and age discrimination (Adams, 2002; McGregor, 2002; Henkens, 2000; Boerlijst, 1994) when assessing older workers. This finding may have specific implications for older workers. They indicated that positive attitude towards older workers work was more important for them than did younger workers as a reason to prolong work. These results should be interpreted cautiously. Given the specific organisational sample used, it may be difficult to generalise results to populations of workers in general, particularly those in different sectors or industries. In addition, sample size restricted more complex analysis that may have further explored the relationships between value incongruence, occupational stress and work ability. Further research is warranted that includes these variables as potential mediators of individual level factors as direct and indirect predictors of workability. Within the constraints of these limitations, the results of this study suggest that organisational and individual factors are important when considering interventions that may address organisational problems relating to an ageing workforce. This study, whilst exploratory, also suggests that organisational value incongruence may indirectly affect work ability through a direct effect on occupational stress related to work culture. That is, as organisational values (especially those related to respect and support) work together with work satisfaction and work ability they reduce occupational stress related to work culture. In turn, as occupational stress decreases, health and work ability increases. Implications of these findings for the participating organisation are in recommending strategies for promoting workability of all workers, and not just ageing workers. Findings suggest that older and younger workers may require different strategies to prevent early exit from work with programs focussed on: career advancement and skill development for younger workers; work flexibility and cultural change to attitudes to older workers for older workers; and strategies related to decreasing occupational stress related to workplace culture such as reducing value incongruence related to respect and support for the individual, and a sense of collective honesty, integrity and compassion for all workers. Throughout this report we investigated differences between older and younger workers on work ability and measures that relate to work ability. Overwhelmingly there were more similarities between these groups than differences. Therefore findings also suggest that strategies would be best geared to increasing and strengthening work ability for all staff across the lifespan, and not just older workers.

REFERENCES ABS, (2004). Mature age persons statistical profile: Labour force. CAT 4905.0.55.001, Canberra, ACT: Australian Bureau of Statistics Adams, S.J., (2002). Passed over for promotion because of age: An empirical analysis of the consequences. Journal of Labour Research, vol. 23 (3): pp. 447–461

56

Aittomaki, A., Lahelma & Roos, E., (2002). Working conditions and socio-economic inequalities in work ability. European Journal of Public Health, vol. 12 (4): pp. 35 Barnes-Farrell, JL., Rumery, SM., and Swody, CA., (2002). How do concepts of age relate to work and off-thejob stresses and strains? A field study of health care workers in five nations. Experimental Aging Research, vol. 28 (1): pp. 87–98 Boerlijst, J.G., (1994). The neglect of growth and development of employees aged over 40: A managerial and training problem. In J Snel & R Kremer (Eds), Work and ageing: A European perspective. Taylor and Francis: London Brooke, L., (2002). Human resource costs and benefits of maintaining a mature-age workforce. International Journal of Manpower, vol. 24 (3): pp. 260–283 Cant, R., O’Loughlin, K., and Legge, V., (2001). Sick leave – Cushion or entitlement? A study of age cohorts’ attitudes and practices in two Australian workplaces. Work: Journal of Prevention Assessment and Rehabilitation, vol. 17 (1): pp. 39–48 Chan, G, Tan, V, and Koh, D., (2000). Ageing and fitness to work. Occupational Medicine-Oxford, vol. 50 (7): pp. 483–491 Chiu, W.C.K., Chan, A.W., Snape, E., and Redman, T., (2001). Age stereotypes and discriminatory attitudes towards older workers: An East-West comparison. Human Relations, vol. 54 (5): pp. 629–661 Code, S., and Langan-Fox, J., (2001). Motivation, cognitions and traits: predicting occupational health, well-being and performance. Stress and Health, vol. 17: pp. 159–174 Cotton, P., (1996). Impediments to achieving best practice in managing work stress. Paper presented at the National Occupational Stress Conference: Health and Well-being in a Changing Work Environment, Brisbane de Boer, A.G.E.M., van Beek, J.C., Durinck, J., Verbeek, J.H.A.M., and van Dijk, F.J.H., (2004). An occupational health intervention programme for workers at risk for early retirement; a randomised controlled trial. Occup. & Environmental Medicine, vol. 61 (11): pp. 924–929 Delongis, A., Folkman, S., and Lazarus, R. S., (1988). The impact of daily stressors on health and mood: Psychological and social measures as mediators. J. of Personality & Social Psychology, 54: pp. 486–495 Edwards, J. R., (1996). An examination of competing versions of the person-environment fir approach to stress. Academy of Management Journal, vol. 39 (2): pp. 292–339 Finish Institute of Occupational Health, 1998, Work Ability Index. (2nd ed.) Fischer, F.M., Bellusci, S.M., Teixeira, L.R., Borges, F.N.S., Ferreira, R.M., Goncalves, M.B.L., Martins, S.E., and Christoffolete, M.A., (2002). Unveiling factors that contribute to functional aging among health care shiftworkers in Sao Paulo, Brazil. Exp. Aging Research, vol. 28 (1): pp. 73–86 Hart, P. M., Wearing, A. J., and Griffin, M. A., (1996). Personality, coping and organisational climate: Where should the intervention dollar be spent. Paper presented at the National Occupational Stress Conference: Health and Well-being in a Changing Work Environment. Brisbane: Australian Academic Press Henkens, K. and van Solinge, H., (2002). Spousal influences on the decisions to retire. International Journal of Sociology, vol. 32 (2): pp. 55–74 Hopsu, L., Leppanen, A., Ranta, R., and Louhevaara, V., (2005). Perceived work ability and individual characteristics as predictors for early exit from working life in professional cleaners. In G. Costa, W.J.A. Goedhart and J. Ilmarinen (Eds.), International Congress Series (vol. 1280), (pp. 84–88). Amsterdam: Elsevier Ilmarinen. J., (1999). Ageing workers in the European Union-Status and promotion of work ability, employability and employment. Helsinki: Finnish Institute of Occupational Health, Ministry of Social Affairs and health, Ministry of Labour Ilmarinen J., (2001). Ageing Workers in Finland and in the European Union: Their Situation and the Promotion of their Working Ability, Employability and Employment. The Geneva Papers, vol. 26 (4): pp. 623–641. Blackwell Publishing Ilmarinen J., (2004). Assessment and promotion of work ability, health and well-being of ageing workers in Giovanni Costa, Willem J. A. Goedhard, and Juhani Ilmarinen, (Eds). Proceedings of the 2nd International Symposium on Work ability, Verona, Italy 18 and 20 October Ilmarinen, J., and Rantanen, J., (1999). Promotion of work ability during ageing. American Journal of industrial medicine supplement, 1: pp. 21–23 Kryger, T., (2005). Australia’s Ageing Workforce. www.aph.gov.au/library/pubs/rn/2004-05/05rn35.htm Lazarus, R. S., and S. Folkman, (1984). Stress, Appraisal and Coping. New York, Springer McGregor, J. and Gray, L., (2002). Stereotypes and older workers: The New Zealand experience. Social Policy Journal of New Zealand, issue 18: pp. 163–177, Accessed 10 March 2004, www.msd.govt.nz./publications/journal/18-june-2002/index.html

57

Medical OutcomesTrust, (2002). SF-12 Health Survey (Version 2). Lincoln, USA: Quality Metric Incorporated. Mitchell, J., (1996). Stressing organisational culture. Health and Well-Being in a Changing Work Environment. Brisbane, Australian Academic Press. National Occupational Stress Conference. Brisbane. Australian Academic Press Palermo, G., (2005). Gender and Organisational Culture: Relationships Between Marginality And Women’s Career Success. Unpublished PhD thesis, Victoria University, Melbourne Platman, K., (2003). The self-designed career in later life: A study of older portfolio workers in the United Kingdom. Ageing and Society, vol. 23 (3): pp. 281–302. Platman, K., and Tinker, A., (1998). Getting on in the BBC: A case study of older workers. Ageing and Society, vol. 18 (5): pp. 513–535 Pohjonen T., (2001). Perceived work ability of home care workers in relation to individual and work-related factors in different age groups. Occupational Medicine 51: pp. 209–217 Rappaport, A., Bancroft, E., and Okum, L., (2003). The aging workforce raises new talent management issues for employers. Journal of organizational excellence, vol. 23 (1): pp. 55–66 Reiso H, Nygard JF, Brage S, Gulbrandsen P, and Tellnes G., (2001). Work ability and duration of certified sickness absence. Scandinavian Journal of Public Health. Sep. 29 (3): pp. 218–25 Richardson, B., Webber, L., Smith, K., and Webb, J., (2006). Age Matters: the effect of job applicant age on hiring decisions. Proceedings (refereed) of the Joint Conference of the Australian Psychological Society and the New Zealand Psychological Society, Auckland, N.Z. (pp. 345–349). September 2006. Australian Psychological Society: Melbourne. Salonen, P, Arola, H, Nygard, CH, Huhtala, H, and Koivisto, M., (2003). Factors associated with premature departure from working life among ageing food industry employees. Occupational Medicine-Oxford, vol. 53 (1): pp. 65–68 Seitsamo, J. and Ilmarinen, J., (1997). Life-style, aging and work ability among active Finnish workers in 1981–1992. Scandinavian J. of Work Environment & Health, vol. 23 (1): pp. 20–26 Sheen, V., (2001). Challenging Convention Australia’s ageing workforce – the challenge for human resource management.http://www.cota.org.au/ahrispeech.htm#top Sherer, M., Maddoux, J. E., Mercandante, G., Prentice-Dunn, S., Jacobs, B., and Rogers, R. W., (1982). The Self-Efficacy Scale: Construction and validation. Psychological Reports, 51: pp. 663–671 Smith, K, and Webber, L., (2005). Promotion: Age matters. In G. Costa, W.J.A. Goedhart and J. Ilmarinen (Eds.), International Congress Series (Vol. 1280) Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers (pp. 403–408). Elsevier: Amsterdam Still, L., and Timms, W., (1998). Career barriers and the older woman manager. Women in Management Review, vol. 13 (4): pp. 43–155 Sutherland, L. F., G. J. Fogarty, and Puthers, (1995). Congruence as a predictor of occupational stress. Journal of Vocational Behavior vol. 46: pp. 292–309 Taylor, P. and Walker, A., (1998). Employers and older workers: Attitudes and employment practices. Ageing and Society, vol. 18 (6): pp. 641–658 The Economist, (2006). The ageing workforce: Turning boomers in boomerangs. The Economist, vol. February 18th: pp. 61, 62 and 64 Tuomi, K, Ilmarinen, J, Klockars, M, Nygard, CH, Seitsamo, J, Huuhtanen, P, and Martikainen, R., (1997). Finnish research project on aging workers in 1981–1992. Scandinavian Journal of Work Environment & Health, vol. 23 (1): pp. 7–11 Tuomi, K., Huuhtanen, P., Nykyri, E., and Ilmarinen, J., (2001). Promotion of work ability, the quality of work and retirement. Occupational Medicine-Oxford, vol. 51 (5): pp. 318–324 Webber, L., Smith, K., and Scott, K., (2006). Age, work ability and plans to leave work. Proceedings (refereed) of the Joint Conference of the Australian Psychological Society & the New Zealand Psychological Society, Auckland, N.Z. (pp. 479–483) September 2006. Australian Psychological Society: Melbourne.

58

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Effect of Japanese employment system on Work Ability Index Masaya Tokuhiro, Hiroyuki Izumi & Jean-Luc Malo Department of Ergonomics, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, Kitakyusyu, Japan

Naomi Uehara Hitachi Metals Wakamatsu, Ltd, Kitakyushu, Japan

Masaharu Kumashiro Department of Ergonomics, Institute of Industrial Ecological Sciences, University of Occupational and Environmental Health, Kitakyusyu, Japan

ABSTRACT: In the near future, Japanese companies will have to consider enforcing an older retirement age to satisfy their employment needs in light of a decrease in the workforce due to a rapidly aging population and extremely low birth-rates. The Work Ability Index (WAI) is a potential tool to facilitate this change. However, according to our previous preliminary studies, the WAI is influenced by the work environment of Japan, including the permanent employment system and the seniority-based system. This cross-sectional survey was conducted with 499 workers. Results from the WAI questionnaires were compared with results from the effort reward imbalance model, showing the importance of job stability on WAI scores. The WAI scores of Japanese workers are much lower than the European reference values and decrease less steeply with age. Our results suggest the need to analyze the WAI score differently in the Japanese population.

1 INTRODUCTION In the near future, innovative changes will have to be brought to the employment format of the Japanese society, mainly because of a decrease in the workforce caused by a rapidly aging population, one of the lowest birth-rate in the world (1.32 per couple according to the Japanese Ministry of Health, Labour and Welfare, 2006) and the retirement of a large number of baby boomers. In Japanese companies, the burdens of the labour shortage combined with the increasing cost of the social security system have taken their toll. Some solutions have been explored such as encouraging the employment of aged workers as well as an extension of the retirement age from 60 to 65 years old in 2025 (according to the Japanese Ministry of Health, Labour and Welfare, 2006). An increased participation of female workers has been suggested, but factories remain shy in their employment policies. Finally, procedures to integrate foreign labour to Japanese companies exist but remain marginal. Consequently, Japanese companies increasingly have to rely on an aging workforce, thus intensifying the need to develop management solutions to maintain and reemploy older workers. Such methods may include considerations for the effective assignment of elderly workers and appropriate reward based on their work ability. The Work Ability Index (WAI) is an effective method to document the balance between the working environment and the health conditions of the workers. However, results from our preliminary studies suggested that the response pattern of Japanese workers differs from that of Europeans and that adjustments might be required in order to make a correct interpretation of the results gathered 59

using this tool. Three factors are thought to influence results when compared to the European standard. The distinctive Japanese work culture is the first of these. It combines a lifelong employment system (employment is guaranteed until the retirement age as long as the company doesn’t face serious financial drawbacks), a seniority system (salary and rank increase with age and years of service rather than with efficiency or other factors) and the low migration of employees between companies. Other particularities that include dedication to the company and long working hours make up the second factor. As such, while Japanese labour laws limit the maximum amount of over-time, it is common for employees to exceed this amount using their personal time. Finally, Japanese “kenson” similar to the concept of modesty is expected to generate an underestimation of self-reported capacity or ability (Yamaguchi et al., 2007, and Furnham et al., 2002), affecting WAI results. This study aims to clarify the interpretation of the WAI in the light of Japanese culture and employment system. To attain this goal, a cross-sectional survey was conducted with 499 workers from a home electric appliances factory using the WAI and the Effort-Reward Imbalance (ERI) model questionnaire. 1.1 Effort-Reward Imbalance The Effort-Reward Imbalance model by Dr Siegrist (1996), a German sociologist, “posits that effort at work is spent as part of a socially organized exchange process to which society at large contributes in terms of occupational rewards. Rewards are distributed to employees by three transmitter systems: money, esteem and security/career opportunity. The ERI Model claims that lack of reciprocity between ‘cost’ and ‘gains’ may cause a state of emotional distress…” (from de Jonge et al., 2000) The psychometric qualities of the questionnaire were assessed with Japanese workers by Tsutsumi et al. (2001, 2002). 2 METHODS This study was conducted in a medium size steel manufacturing company. In the company, field workers are required to participate in the melting and processing of raw metal to produce different parts (including: pillars for bridges, big components for industrial cutting tool and rollers). The seniority system and the lifelong employment system were implemented before the 1960’s. The survey was distributed by the occupational health physician to supervisors and later distributed to all employees during a period of two weeks in June 2007. The survey included questions about personal characteristics (name, age, length of service, job title, etc.), the short version of the Work Ability Index (Ilmarinen et al., 1996) and the effort-reward imbalance questionnaire (Tsutsumi et al., 2001). Statistical analysis were made using SPSS2. 3 RESULTS 3.1 Participants The 499 respondents (471 men and 28 women) correspond to over 99% of the total work force from the Melting and Processing department of the studied company (biggest department in this company). Of these, 92% performed physical work and 8% performed other administrative or managerial functions. None of the surveyed workers refused participation although the survey was completed on a voluntary basis. Such a high response rate is typical of surveys distributed by occupational health doctors in Japanese companies. It is partially explained by the legal power of the occupational health doctor and the job stability provided by the permanent employment system. The participants’ average age was 43.2 years (SD: 12.5, Median: 40.2) and the average length of service 19.4 years (SD: 13.6, Median: 16.8) for men and 44.9 years (SD: 11.7, Median: 42.3) and 14.2 years (SD: 10.1, Median: 15.0) for women respectively. 60

45

WAI score

35

25

15

5 10

Figure 1. 100%

20

30

40 Age

50

60

70

Distribution of WAI according to the age of the respondents. 0 4.44

0 5.2

0 6.2

0 8.2

12.5 18.4 26.1

46.6

25

52.4

71.9

65.2 71.9

Excellent

51 80.8

80.8

50%

Good

43.9 37.5

Moderate Poor

40.9 41.3 21.7 26.5

30.4

30.6

22.9 6.3

1.6 0

0%

Company Reference value 20–29

Figure 2.

25 11.3

13

11

8.3

1.2

0 Company

Reference value

30–39

Company

Reference value

40–49

Company

Reference value

50–59

Company

Reference value

60–64

Percentage of worker in each WAI score categories in comparison to data from Ilmarinen et al. (1997).

3.2 Work Ability Index results In all age groups, the average WAI scores (30.68) were lower than values from European studies where the averages are almost exclusively contained between 35 and 43 depending on the age group and occupational sector (Costa et al., 2007, and Ilmarinen et al., 1997). Figure 1 shows the WAI score distribution according to the age of the respondents. The linear regression slope shows a soft descent (r = −0.11). European studies have usually gotten steeper regression slopes (r = −0.27, example from Tobia et al., 2005), and higher scores. Figure 2, allows us to appreciate the difference in the WAI score distribution amongst the typical four classifications compared to values reference values by Ilmarinen and collaborators (1997). The absence of results in the “Excellent” group and the very limited percentage in the “Good” 61

Table 1.

Contribution of each Item of the WAI to the total score for the studied population.

Items

Scale

Average (SD)

% of max score lost

1. Subjective estimation of present Work Ability compared with the lifetime best 2. Subjective Work Ability in relation to both physical and mental demands of work 3. Number of diagnosed diseases 4. Subjective estimation of work impairment due to disease 5. Sickness absence during part year 6. Own prognosis of Work Ability after two years 7. Psychological resources

1–10

7.30 (1.90)

27.0

2–10

7.16 (1.60)

28.4

1–7 1–6 1–5 1, 4, 7 1–4

5.53 (1.65) 5.60 (0.83) 4.42 (0.79) 5.56 (1.82) 1.42 (0.48)

21.1 6.6 11.6 20.6 64.5

category leads to two hypothesis. That the population of this particular study is in poor health or that the answering patterns differ due to cultural reasons. We cannot accept the first postulate as a valid explanation to this discrepancy as these low scores are consistent with our ongoing longitudinal study in another larger factory. Furthermore, this study’s population does not show any other symptom of ill health, with the company having a normal productivity, health and safety records, and health check records. The Japanese workforce also tends to remain highly employed until a later age. In 2004, the unemployment rate was of 8.6% for the 15–24 years old; 6.2% for the 25–34; 4.5% for the 55–64; 4.2% for the 35–44; 3.9% for the 45–54 and of only 2.7% for the 65 years old and over according to the Japanese Ministry of Health, Labour and Welfare in a report published in 2005. The second hypothesis where the reason lies in a difference in answering patterns related to cultural characteristics appears more probable. Japanese modesty, or Kenson, might be responsible for these low scores. Cross cultural studies have already reported a tendency of Japanese people to underestimate their self-esteem (Yamaguchi et al., 2007) and personal skills (Furnham et al., 2002; Akimoto et al., 1999). This tendency probably contributed strongly to the low WAI score, but more studies will be needed to clarify how to interpret Japanese WAI scores and to create potential new categories of WAI results. Finally, it is interesting to notice that the highest proportion of Good WAI Score is found in the 60+ group, supporting a possible healthy worker effect as reported in some recent WAI studies. It could alternatively be related to an age related discrepancy in answering patterns. The Table 1 describes the average WAI score divided among its seven items. Item 7, psychological resources, presented low scores (average 1.42 on a 1–4 scale), and a proportionally low standard deviation. This could either be related to specific characteristics of this study’s workplace, but could more probably, be related to Japanese culture. Understanding of this question by Japanese workers should be investigated. 3.3 Results from the Effort-Reward Imbalance model Average results by age groups of the ERI questionnaire are comparable to reference values from (Seigrist et al., 2004) and are coherent to studies with Japanese workers (Tsutsumi et al., 2001, 2002). The Table 2 describes the Correlations between ERI results and WAI categories. Results from the ERI showed few correlations with higher or lower than average WAI scores. The Factor 3 related to job and position stability was the only one presenting significant relationship with the WAI level. In this population, a higher WAI score was linked to stability of job and position. It suggests that even if a Japanese worker suffers from a disease, he maintains a strong motivation and/or desire to remain in the same workplace. This might be taken into account by the employer when managing duties. 62

Table 2.

Correlations between ERI results and WAI categories.

Age ERI score Effort Factor 3 (Job & position stability)

Age 20–44

Age 45-65 WAI below average

Age 45–65 WAI above average

−0.160** −0.159** −0.315** −0.484**

−0.012 −0.190** −0.277** −0.610*

−0.045 −0.210** −0.251** −0.551*

*p < 0.05, **p > 0.01

4 DISCUSSION This research examined results from the WAI questionnaire and the ERI questionnaire in the context of the seniority system, the lifelong employment system, and Japanese cultural characteristics. The WAI results presented in the current study will prove useful for comparative purposes with other Japanese studies, but at this moment there appears to be no calibration method available to compare these results with foreign studies. The Average WAI score shows limited changes according to different age groups (Figure 1 and Figure 2). The older workers did not obtain low WAI scores. Possible explanations include: • Healthy worker effect: unhealthy workers retire prematurely, keeping the average WAI score high. This, however does not seem to be the case for workers under 65 according to preliminary retirement statistics analysis, • Aged workers are in good shape or working in a facilitating environment, • Different answering patterns of older generations: possibly related to cultural differences or beliefs; Therefore, the WAI might not be a sensitive cross-sectional evaluation tool in the Japanese context. The cultural differences in WAI responses suggest that scores and categories cannot be compared directly, stressing the interest of comparing relationships between WAI scores and other factors rather than using WAI scores directly. This is especially problematic since cultural differences can exist even between companies in the same country. Confounding factors could include: local context, company policies and disproportional representation of subgroups such as generation, gender and cultural or ethnic backgrounds. These considerations are not as important during longitudinal or intervention studies as long as the population is relatively stable, between the pretest and post-test evaluation. Therefore, we surmise that the international potential of the WAI index could reside in its ability to evaluate the impact of changes in occupational settings, rather than as a cross-sectional health evaluation tool. However, our hypotheses need to be compared with other types of jobs and employment systems, to study their influence on WAI. Also, better understanding of the effect of modesty on survey responses should be investigated to facilitate the adaptation of foreign questionnaires to the Japanese context. Finally, in order for the discrepancy between real work capacity and reported WAI scores, if there is any, to be documented, objective measurement of a worker’s capacity could be explored to set a gold standard and eventually understand the cultural bias of Japanese culture on WAI. REFERENCES Costa, G., and Sartori, S., (2007). Ageing, working hours and work ability. Ergonomics, 50:11: pp. 1914–1930 de Jonge, J., Bosma, H., Peter, P., and Siegrist, J., (2000). Job Strain, Effort-reward Imbalance and Employee Wellbeing: a Large-scale Cross-sectional study. Social Science & Medecine, 50: pp. 1317–1327

63

Furnham, A., Hosoeb, T., and Li-Ping Tang, T., (2002). Male hubris and female humility? A crosscultural study of ratings of self, parental, and sibling multiple intelligence in America, Britain, and Japan, Intelligence, 30(1): pp. 101–115 Ilmarinen, J., Tuomi, K., and Klockars, M., (1997). Changes in the work ability of active employees over an 11-year period. Scan J Work Environ Health, 23(1): pp. 49–57 Ministry of Health, Labour and Welfare, 2006 Siegrist, J., (1996). Adverse health effects of high-effort/low-reward conditions. Journal of Occupational Health Psychology, 1(1): pp. 27–41 Siegrist, J., Starke, D., Chandola, T., Godin, I., Marmot, M., Niedhammer, I., and Peter, R., (2004). The Measurement of Effort-Reward Imbalance at Work: European Comparisons. Social Science & Medicine, 58: pp. 1483–1499 Tsutsumi, A., Ishitake, T., Peter, R., Siegrist, J., and Matoba, T., (2001). The Japanese version of the EffortReward Imbalance Questionnaire: a study in dental technicians. Work and Stress, 15(1): pp. 86–96 Tsutsumi, A., Kayaba, K., Nagami, M., Miki, A., Kawano,Y., Ohya,Y., Odagiri,Y., and Shimomitsu, T., (2002). The effort-reward imbalance model: experience in Japanese working population. Journal of Occupational Health, 44: pp. 398–407 Yamaguchi, S., Greenwald, A. G., Banaji, M.R., Murrakami, F., Chen, D., and Shiomura, K. et al., (2007). Apparent Universality of Positive Implicite Self-Esteem. Psychological Science, 18(6): pp. 498–500

64

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Study on work ability of Vietnamese workers in selected industries Duong Khanh Van National Institute of Occupational and Environmental Health, Hanoi, Vietnam

Nguyen Ngoc Nga Vietnam Association of Occupational Health, Hanoi, Vietnam

ABSTRACT: Objectives – This study was aimed at adapting the Work Ability Index (WAI) to the Vietnamese setting and to start an investigation of the work ability of Vietnamese workers from different industries and different age groups. Methods – This descriptive cross-sectional study investigated a population of 2,833 Vietnamese workers (1,527 men and 1,306 women) at different ages from five job groups (fishermen, foundry workers, health care workers, pharmaceutical workers, cement factory workers). The Work Ability Index (WAI) questionnaire was used to assess their work ability. The WAI scores were calculated according to instructions provided by the Finnish Institute of Occupational Health. Age, gender and type of industry were taken into account in order to establish their relationship with WAI scores. The data was analyzed with SPSS 13.0. Results – In average, the 20–30 age group scored highest (44 ± 5 in men and 45 ± 4) on the WAI scale. Workers above 41 years old reported significantly lower rates of excellent WAI scores. However, in the age group of 51–60, 53.3% of workers still reported good and excellent scores. We also found that evolution of WAI scores through age was dependant on workers job categories. In this study, throughout their working lives, the WAI score of men decreased by 18% and the WAI score of women decreased by 15%. At the age of 60 years and over, the mean WAI score of men and women were not different. Conclusions – The WAI can be applied for Vietnamese people with only slight modifications. Recommendations were made for further studies and for the application of the WAI in monitoring workers’ health.

1 INTRODUCTION One of the main targets of occupational health services is to maintain and to promote workers’ work ability; therefore it is necessary to possess a valid tool to measure work ability. So far, in Vietnam, PWCmax (Maximum Physical Work Capacity) and PWC170 (Physical Work Capacity at a heart rate of 170) were used in researching physical work. But these indexes are better adapted to laboratory research and smaller samples, rather than to workplaces, and show limits in measuring work ability in technical and mental work. After its introduction in Vietnam by Prof. Kumashiro and Dr. Kurrpa, the WAI was translated into Vietnamese. After a pilot study in 2001, the translation was adjusted for the first large scale work ability investigation study in Vietnam, aiming at studying Vietnamese workers’ work ability in some selected industries. 65

2 METHOD In this descriptive cross-sectional study, the sample was comprised of 2,833 Vietnamese workers (1,527 men and 1,306 women) at different ages from five different industries (agriculture, fishery, foundry, health service, cement manufacture). The WorkAbility Index (WAI) questionnaire was used to assess their work ability. The WAI scores were calculated according to the standard method provided by the Finnish Institute of Occupational Health (Tuomi et al., 1998). Age, gender and type of industry were taken into account in order to establish their relationship with WAI scores. The resulting data was analyzed with SPSS 13.0.

3 RESULTS Table 1 shows the mean WAI score and WAI category of all the studied workers. The mean WAI score in this study was lower than that found by Goedhard (2004). In this study, 42.1% of the workers scored excellent in the WAI. Furthermore, the percentage of workers who scored excellent in this study was lower than that of Goedhard’s research but higher than that of Costa et al. research (2004). 5.8% of the workers had poor WAI score. This rate was much higher than that found in the research on Europeans. Table 2 compares the mean WAI score of workers by gender and age groups. The mean score decreased for men from 44 (in the 21–30 years age group) to 27 (in the 60 years and older age group). In women, it decreased from 45 to 27. In both men and women, the WAI score of the 20 years and younger age group was lower than that of the 21–30 years age group. This could be related to the experience and skills at work.

Table 1. WAI score and category. Parameters Average WAI score WAI category (%)

Excellent Good Moderate Poor

n

Results

2,833 1,193 960 516 164

40.5 ± 7 42.1 33.9 18.2 5.8

Table 2. WAI score of workers by gender and age groups. Men

Women

Age group

n

WAI score

n

WAI score

≤20 21–30 31–40 41–50 51–60 51–55 56–60 >60

35 252 453 486 218 149 69 72

43 ± 5 44 ± 5 42 ± 6 40 ± 7 37 ± 8 37 ± 8 37 ± 7 27 ± 8

31 255 312 475 166 123 43 56

44 ± 4 45 ± 4 42 ± 5 41 ± 6 37 ± 7 38 ± 7 35 ± 7 27 ± 8

66

In Vietnam, the age for retirement is of 60 years old for men and 55 years old for women. In the course of their working life, men’s WAI scores decreased by 18% while the WAI scores of women decreased by only 15%. AS was also reported in some other studies on WAI, we did not find a later increase of WAI score for older workers. At the age of 60 years and older, the mean WAI scores did not differ between men and women. The standard deviation (SD) of the WAI score increased from 5 (in men) and 4 (in women) in the youngest group to 8 (in men) and 7 (in women) in the oldest group. This indicated an increase in individual differences related to aging and work ability. The inter-individual differences of WAI score in women was smaller than in men. Table 3 shows the percentage of workers’ WAI category in different age groups. The rate of workers with excellent and good WAI decreased as they aged, while it was noted that the rate of workers with poor WAI increased, especially after 50 years old. Under 60 years old the rate of women with poor WAI tended to be lower than that of men, but 50% of women aged 60 years and older had poor WAI, while the rate in men was of only 44.5%. Table 4 details the WAI scores of workers in different job groups. The highest WAI score was found in pharmaceutical workers followed by health workers. The fishers and farmers respectively had the lowest WAI score. However, in this study, the population was not distributed evenly among jobs. The number of fishers and farmers involved in this study was greater than that of other jobs, and also the number of elderly workers in fishery and farming was also greater than that of other job categories. This may have somewhat influenced the results. 55 50 45

Men Women

40 35 30 25 20
20

21–30

31–40

41–50

51–60

51–55

56–60

 60

Figure 1. Age trend of WAI score. Table 3. WAI category of workers in different age groups. Excellent

Good

Moderate

Poor

Age group

Men

Women

Men

Women

Men

Women

Men

Women

n ≤20 21–30 31–40 41–50 51–55 55–60 51–60 >60 All age

605 42.9 62.2 48.6 32.3 20.8 20.6 20.7 1.4 39.6

587 64.5 74.2 46.2 39.4 25.4 14.0 22.4 3.6 45.0

542 45.7 27.5 32.8 42.2 36.6 33.8 33.6 8.3 35.5

417 32.3 22.7 37.5 37.3 34.1 23.3 30.9 5.4 32.0

278 11.4 8.4 11.3 19.8 32.2 35.3 33.3 45.8 18.2

238 3.2 2.0 15.4 21.5 32.5 44.2 35.8 41.1 18.2

101 0 2.0 2.0 5.8 13.4 10.3 12.4 44.5 6.6

62 0 1.2 1.0 1.9 8.1 18.6 10.9 50.0 4.8

67

100% 80%

Poor Moderate

60%

Good 40%

Excellent

20% 0%
20

21–30

31–40

41–50

51–55

55–60

51–60

 60

Figure 2. WAI categories of male workers according to age groups. 100% 80%

Poor Moderate

60%

Good 40%

Excellent

20% 0%
20

20–30

31–40

41–50

51–55

55–60

51–60

 60

Figure 3. WAI categories of female workers according to age groups.

Table 4. Average WAI score of aging workers in different jobs. WAI all ages

51–55

55–60

>60

Job

n

X ± SD

n

X ± SD

n

X ± SD

n

X ± SD

Foundry workers Health workers Farmers Cement workers Fishers Mechanicals Pharmaceutical workers Others

108 386 881 182 834 162 199 75

42 ± 6 45 ± 4 38 ± 8 44 ± 3 38 ± 7 42 ± 4 47 ± 3 43 ± 5

3 25 113 3 95 9 20 5

37 ± 6 42 ± 5 36 ± 7 40 ± 11 36 ± 7 40 ± 4 44 ± 3 36 ± 6

– – 44 – 48 – 6 –

– – 34 ± 8 – 36 ± 7 – 43 ± 5 –

6 – 75 – 45 – – –

21 ± 8 – 26 ± 8 – 29 ± 9 – – –

Table 5 shows the WAI category of workers with different jobs. The highest rate of excellent and good WAI was found in pharmaceutical workers, followed by health workers while the lowest scores were found in fishers and farmers. This may be somewhat explained by their working and living conditions. Education and employee selection could also be an explanation for this difference. Farming and fishery were two jobs with low education requirement compared to the other jobs. Farmers and fishermen were usually work free (self-employed) and therefore had fewer chances of having a good health care and health education than the other professions in this list. 68

Table 5. WAI category in different jobs. Job

n

Excellent

Good

Moderate

Poor

Foundry workers Health workers Farmers Cement workers Fishermen Mechanicals Pharmaceutical workers Others

108 386 881 182 834 162 199 75

43.5 75.4 24.7 72.0 25.5 37.7 91.0 62.7

43.5 18.9 39.0 25.8 39.1 56.2 7.5 21.3

8.3 5.2 25.4 1.6 28.2 5.6 1.5 16.0

4.6 0.5 10.8 0.5 7.2 0.6 0 0

Table 6. WAI score and its item. Item

Range

Excellent

Good

Moderate

Poor

1. Present work 2. WA related to WD 3. Disease 4. Impairment 5. Sick absent 6. Prognosis WA 7. Resource

0–10 2–10 1–7 1–6 1–5 1–7 1–4

9±1 9±1 6.6 ± 1 6±1 5±0 7±0 4±1

8±1 7.5 ± 1 6±1 5±1 4±1 6.7 ± 1 3±1

6±1 6±1 5±1 4±1 4±1 5±1 3±1

5±2 4±2 4±1 3±1 2±1 2±1 2±1

The score of all items reduced from excellent WAI group to poor WAI group. We found a strong correlation between self-evaluation of work ability and the final score category. 4 CONCLUSIONS The average work ability index score was different across different occupational sectors, with lower scores for traditional sectors (fishing and farming) and higher scores in more modern work environments (pharmaceutical, health and cement sector). The general age trend of WAI scores is found to decrease with age, with the exception of younger workers (under 20 years old) having slightly lower scores than the 21 to 30 age group. This might raise some concerns, but this difference is probably related to the difficulties associated with learning and the lack of experience. In this study, the men’s WAI score decreased by 18% and by 15% for women when comparing the scores starting from the 21–30 age group to the 60+ age group. Across all age groups, the scores for men and women seemed similar and at the age of 60 years and older, the mean WAI score of men and female was not different The WAI can be applied for Vietnamese people with only slight modifications in wording (which will be discussed in another article). In order to have a better view of WA of Vietnamese workers, further studies with larger population samples and more different industries are needed. The focus might be placed on a more elderly population in informal sectors which accounts for a large proportion of the labour force in Vietnam. Besides, the set-up of a data bank on WAI could prove to be very useful in monitoring workers’ health. REFERENCES Abstracts 10-year activity of the National Institute of Occupational and Environmental Health, Hanoi

69

Costa G, Antonacci G, Olivato D, Bertoldo B, and Ciuffa V., (2004). Ageing and Work Ability Index in Italian workers. Proceedings of the 1st International Symposium on Work Ability. People and Work Research Report 65, Finnish Institute of Occupational Heath, Helsinki: pp. 33–40 Duong Khanh Van, N.N.Nga, L.G. Khai, and T.T. Binh, (1994). Reasearch on changing some physiological parameters of young men and women in experimental physical work. Goedhard WJA., (2004). WAI scores and its different items in relation to age: a study in two industrial companies in the Netherlands. Proceedings of the 1st International Symposium on Work Ability. People and Work Research Report 65, Finnish Institute of Occupational Heath, Helsinki: pp. 26–32 Tuomi K, Ilmarinen J, Jahkola A, Katajarinne L, and Tulkki A., (1998). Work Ability Index. Finnish Institute of Occupational Health, Helsinki, 2nd revised edition.

70

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

WAI among young employees in Brazil with new scores Inês Monteiro Faculty of Medical Sciences, State University of Campinas – UNICAMP, Campinas, Brazil

Juhani Ilmarinen, Kaija Tuomi & Jorma Seitsamo Finnish Institute of Occupational Health. Helsinki, Finland

Eliane P. Goes, Angela Puzzi Fernandes & Eloísa Petruci Hodge Work and Health Research Group, Faculty of Medical Sciences, State University Campinas – UNICAMP, Campinas, Brazil

ABSTRACT: The purpose of this study was to evaluate work ability, using the Work Ability Index – WAI – among young workers employed in different sectors of the economy, using the WAI scores proposed by Kujala et al. A cross-sectional study was conducted of workers under 30 years of age employed is a variety of sectors, and the Tuomi et al. and Kujala et al. criteria for cut-off points were utilized. It is important to consider the new cut-off points for young workers – less than 30 years old in developing countries – to prevent problems in relation to work ability. Keywords: Work Ability Index, young employees, reference values

1 INTRODUCTION In developing countries workers sometimes begin working at an early age in many sectors, such as agriculture and other sectors characterized by physically demanding jobs; and workers in such jobs often have low levels of schooling (less then nine years). The promotion of work ability of young workers, from the time they begin working is very important for the proper maintenance of their health, skills and performance at work until the retirement age in good conditions. The WAI was developed in Finland by Tuomi et al. (1997) based on a study of workers aged 45 to 58 years old. Kujala et al. (2005) proposed, in an article published in 2005, and before that in a presentation at the 2003 ICOH Congress in Iguassu Falls, new criteria for WAI cut-off points for young workers, that included the following categories: poor (7–36 points), moderate (37–40 points), good (41–44 points) and excellent (45–49 points). The criteria proposed by Tuomi et al. (1997) for WAI cut-off points were: poor (7–27 points, moderate (28–36 points), good (37–43 points) and excellent (44–49 points). As reported by Kujala et al. (2005) “for young employees having impaired job performance, however, the use of this reference limit may resulting in an overestimation of work ability”.

2 METHOD The study is a part of a large study about work ability in different sectors in Brazil that was initiated in 2000. 71

A cross-sectional study was carried out in companies from different sectors of the economy, such as like as pharmaceutical industries, an information technology and telecommunication company; poultry industry; wholesale vegetables, fruits and flower market. The data collection was carried out in 2003–2006. The companies are located in two Brazilian States: Paraná, in the south, and São Paulo State, in the southeastern region. A questionnaire with socio-demographic data and the Work Ability Index were applied. The random sample included 1881 young workers. The research was approved by the Ethics Committee of the Faculty of Medical Sciences – State University of Campinas. The criteria suggested by Kujala et al. (2005) in relation to WAI categories cut-off points was used in the Work and Health Research Group, at State University of Campinas – Brazil, for workers who were less then 30 years old, since 2004.

3 RESULTS AND DISCUSSION The male population represented 61.2% of the workers (range from 14 to 29 years old), and the mean age was 23.5 years (SD = 3.4). The majority of workers was single (63.3%), and without children (70.8%). In relation to work demand 54.7% engaged in physically; 24.8% mixed (physically and mentally) and 20.5% mentally demanding work. The WAI scores range from 17–49 points and the average was 42.4 points (SD = 4.5). The distribution categories using Kujala et al. criteria were: poor – 10.1%; moderate – 19.6%; good – 33.0% and excellent – 37.4%. The mean current work ability was 8.5 (SD = 1.5) points. The WAI distribution categories with Tuomi et al. criteria were: poor – 0.5%; moderate – 9.3%; good – 42.7% and excellent – 47.2%. Statistical analysis was performed in SAS 9.1. The figures below shows the WAI values with two different cut-off points (Figures 1 and 2).

WAI – Kujala et al.(2005) criteria Poor Moderate Good Excellent

Figure 1.

Distribution of WAI categories among young workers with Kujala et al. (1997) cut-off points criteria. WAI – Tuomi et al.(1997) criteria Poor Moderate Good Excellent

Figure 2.

Distribution of WAI categories among young workers with Tuomi et al. (1997) cut-off points criteria.

72

4 CONCLUSION In special conditions such as those found in some areas in Brazil, it is important to think about health promotion and prevention at work. Thus the use of the new criteria for WAI cut-off points is important to anticipate problems concerned with work ability of young workers. The results give support to analysis related to workplace health promotion and the importance of instruments that closely measure work ability. Based on the new criteria, 29.7% of the young workers need some sort of measure to restore or improve their work ability. REFERENCES Kujala, V., Remes, J., and Ek, E et al. (2005). Classification of Work Ability Index among young employees. Occupational Medicine, 55: pp. 399–401. Tuomi, K., Ilmarinen, J., and Jahakola, A. et al. (1997). Índice de capacidade para o trabalho. [Work Ability Index]. (Helsinki: Finnish Institute of Occupational Health).

73

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Verifying of the theoretical model of perceived work ability in the field of teaching Veikko Louhevaara University of Kuopio and Finnish Institute of Occupational Health, Kuopio, Finland

Susanna Järvelin Institute of Biomedicine, University of Kuopio, Kuopio, Finland

ABSTRACT: The objective of this study was to assess the work-related and individual predictors of perceived work ability in the field of teaching, and to consider the predictors in the framework of the Work ability house. The subjects were 44 teachers and superiors with the mean age of 50 years. They were compared in age- and gender-matched pairs. Several significant work-related and individual predictors were observed for perceived work ability assessed with the Work Ability Index (WAI). The most novel finding was the incomplete recovery of the subjects with the reduced WAI from stress during sleep. In the near future the main challenge will be to support the work ability, health and wellbeing of older workers by improving ergonomics and individual resources needed at work.

1 INTRODUCTION Work ability is associated with work capacity and work performance. The terms are often quite confusing, and there is no consensus on the definition. Work performance and work capacity mainly reflect individual factors such as a worker’s physical and mental fitness, attitudes and behaviour, and his or her satisfaction and motivation for work (Robertson and Tracy, 1998, and Shephard, 2000). The work ability is characterized by the balance between a worker’s individual resources and demands of the work incorporating environment and work organisation. The principal individual resources include health, capacities, competence and values. The theoretical framework of the work ability can be described as a construction with four floors applying so called Work ability house (Figure 1). In the framework, health and physical, psychological and social capacity are the basic first floor elements of individual work ability. The second floor of the Work ability house reflects professional competence referring to knowledge and skills. The competence has also the dimension for developing own work and acting in different work communities. Values, attitudes and motivation are in the third floor, and they regulate the balance between individual resources and work as well as between work and leisure time. The fourth floor is dedicated to work and its various environmental exposures and psycho-physiological and social load factors. The organization of the work, functioning of the work community and management are multidimensional issues and difficult to conceptualize and evaluate (Ilmarinen, 2006). Ergonomics or occupational ergonomics is close to the concept of work ability because ergonomics can be defined as a multidisciplinary scientific field that is based on physiology, psychology and technical sciences. It considers and develops the interaction between a worker and his or her work. In ergonomics, a worker’s individual capabilities, needs and limitations are compared with respect to the demands of technical and organisational work systems. Ergonomic design and improvements are carried out to accommodate human factors and work demands (Louhevaara, 1999, and Sillanpää, 2007). 75

Society

Family

Close community

Work ability

Values Attitudes

Motivation

Professional competence

Occupational safety

Occupational health care

Work Work conditions Work content and demands Work community and organization Supervisory work and management

Health, functional capacity

Figure 1. The Work ability house (Ilmarinen, 2006).

The Work Ability Index (WAI) is a prevalent method to assess perceived work ability (Ilmarinen et al., 1997, and Tuomi et al., 1998). The WAI is affected by several ergonomics (work-related) and individual factors (Ilmarinen et al., 1997). Recently Hopsu et al. (2005) reported that the reduced WAI and overweight were powerful predictors for the early exit from working life in female professional cleaners during the 12-year follow-up. Physical fitness decreases due to age which is one of the most significant predictors for the WAI (Ilmarinen, 2001, and Robertson and Tracy, 1998). In order to have good results in the promotion of work ability it is important to define all factors affecting work ability in different occupational contexts, especially, when the impact of the age can be controlled. The objective of this study was to assess the work-related and individual predictors of perceived work ability in the field of teaching, and to consider the predictors in the framework of the Work ability house.

2 METHODS 2.1 Subjects The subjects were 44 teachers and superiors (18 men and 26 women). Their mean age was 50 (range 35–63) years. The subjects were compared in pairs derived from two age- and gendermatched groups according to their WAI: the group A having the excellent WAI with the mean of 45.8 (SD = 1.9) points and the group B having the reduced WAI with the mean of 34.7 (SD = 2.0) points. 76

2.2 Questionnaires Perceived work ability was assessed with the WAI which covered the following seven items: (1) Subjective estimation of current work ability compared with lifetime best (0–10 points), (2) Subjective work ability in relation to both physical and mental demands of work (2–10 points), (3) Number of diagnosed diseases (1–7 points), (4) Subjective estimation of work impairment due to diseases (1–6 points), (5) Sickness absenteeism during the past year (1–5 points), (6) Own prognosis of work ability after two years (1, 4 or 7 points), and (7) Psychological resources (enjoying daily tasks, activity and life spirit, optimistic about the future) (1–4 points). The score of the WAI ranges from 7 to 49, and is divided into four categories as follows: poor (7–27 points), moderate (28–36 points), good (37–43 points), and excellent (44–49 points) (Tuomi et al., 1998). Subjective burn-out and work-related stress were assessed with the Bergen Burnout Indicator (BBI) and the Occupational Stress Questionnaire (SQ), respectively (Näätänen et al., 2003, and Elo et al., 1992). Work motivation, professional competence, organizing of the work and functionality of the work community were assessed using a questionnaire items with the scale from 0 to 10, having the endpoints of “totally disagree” and “totally agree”, respectively. The medication, drinking and smoking habits were also inquired by the questionnaire. 2.3 Resting values The resting values of cardiorespiratory system were recorded in a supine position in the laboratory. Heart rate (HR) was recorded using the Suunto Smart Belt equipment (www.suunto.com). Diastolic blood pressure (DBP) and systolic blood pressure (SBP) were measured with the Omron M4 automatic device (www.europe.omron.com). 2.4 Physical capacity The flexibility of the lumbar spine was assessed with the modified Schober test (Gill et al., 1988). The body balance was tested with the functional balance test (Punakallio et al., 1997). Muscle strength of the trunk flexors was measured with the repetitive sit-up test (Pollock and Willmore, 1990). The assessments of the anthropometrics and body composition preceded the tests. The measurements were carried out in the laboratory. The aerobic capacity was assessed with the 2-km walking test (Laukkanen et al., 2000) on an indoor track of 400 m. 2.5 Heart rate and heart rate variability HR and heart rate variability (HRV) were recorded with the Suunto Smart Belt equipment. The 24-hour recording period consisted of a work shift, leisure time and sleep. The subjects were instructed to work in their habitual manner and to maintain their normal leisure activities. The data were analysed using the Wellness Software developed by Firstbeat Technologies Ltd. (www.firstbeattechnologies.com). 2.6 Statistical methods Descriptive statistics included the calculation of frequencies, means, standard deviations and ranges. The group differences were tested with the χ2 test or the t-test with paired samples. The level of significance was set at p < 0.05.

3 RESULTS The perceived BBI and SQ scores of the group A with the excellent WAI were significantly lower than those of the group B with the reduced WAI (p = 0.013 and p = 0.003, respectively). In the 77

Table 1. The questionnaire based results of the study: Bergen Burnout Indicator (BBI), Occupational Stress Questionnaire (SQ), medication and drinking and smoking habits. Group A (n = 22)

Group B (n = 21–22)

n

%

n

BBI No burnout symptoms Mild burnout symptoms Fair burnout symptoms Severe burnout symptoms

19 0 2 1

86 0 9 5

6 6 8 2

27 27 36 9

SQ No stress Fairly little stress Somewhat stress Fairly much stress Very much stress

3 10 6 2 1

14 46 27 9 5

1 4 5 7 5

5 18 23 32 23

Medication No medication Regular medication

17 5

77 23

7 14

32 64

Drinking No drinking Weekly Occasionally

1 7 14

5 32 64

0 12 9

0 56 41

Smoking No smoking Daily Occasionally

20 0 2

91 0 9

19 1 1

86 5 5

%

pa 0.013

0.003

0.001

0.184

0.513

a χ2 -test

Table 2. The results of the questionnaires about work motivation, professional competence, organisation of the work and the functionality of the work community with the scale of 0–10 (m = mean, sd = standard deviation).

Motivation Professional competence Organisation of the work Functionality of the work community a t-test

Group A (n = 22) m ± sd (range)

Group B (n = 22) m ± sd (range)

pa

8.2 ± 1.0 (5.3–9.6) 7.7 ± 1.2 (5.2–9.6) 7.5 ± 1.3 (5.2–9.7) 7.9 ± 1.0 (5.3–9.6)

7.3 ± 1.3 (4.3–9.3) 6.5 ± 1.8 (3.0–9.5) 6.2 ± 2.0 (2.0–8.6) 7.0 ± 1.6 (2.8–9.4)

0.014 0.060 0.036 0.041

with paired samples

group A, 23% of the subjects used medication regularly whereas the corresponding value was 64% for the group B (p = 0.001) (Table 1). The group A had a better motivation to work (p = 0.014) and work organization (p = 0.036) and they were more satisfied with the functionality of the work community (p = 0.041) compared to the group B. There was also a strong trend that the group A perceived to have a better professional competence (p = 0.060) than that of the group B (Table 2). There was a significant difference between the groups in the Body mass index (p = 0.016) and the percentage of fat (p = 0.014). In the variables of physical capacity a significant difference 78

Table 3. The resting values of heart rate (HR), diastolic blood pressure (DBP), systolic blood pressure (SBP), anthropometrics (Body mass index = BMI) as well as the values characterising physical capacity (Lumbar spine mobility = L-spine mobility) in the group A with the excellent WAI and the group B with the reduced WAI (m = mean, sd = standard deviation). Group A

HR (beats/min) DBP (mmHg) SBP (mmHg) BMI (kg/m2 ) Percentage of fat (%) 2-km walking test (index) Sit up (max rep.) L-spine mobility (cm) Functional balance (s) a

Group B

n

m ± sd (range)

n

m ± sd (range)

pa

19 22 22 18 18 15 19 18 18

67 ± 12 (46–89) 88 ± 13 (72–127) 136 ± 21 (116–200) 25.1 ± 3.7 (20.4–34.7) 25.3 ± 6.5 (15–49) 104 ± 11 (81–120) 36 ± 14 (6–50) 6.7 ± 0.9 (4.5–8.0) 16.9 ± 4.5 (11.2–31.6)

19 22 22 18 18 15 19 18 18

74 ± 11 (57–94) 87 ± 11 (69–106) 137 ± 16 (110–166) 28.2 ± 4.7 (22.4–37.3) 30.3 ± 7.9 (17–47) 85 ± 24 (39–119) 28 ± 12 (4–50) 7.1 ± 0.9 (5.8–9.0) 17.5 ± 6.6 (10.4–38.5)

0.102 0.805 0.854 0.016 0.014 0.003 0.048 0.155 0.767

t-test with paired samples

between the groups was detected in the index of the walking test and strength of the trunk flexors (p = 0.003 and p = 0.048, respectively) (Table 3). During sleep the group A with the excellent WAI seemed to recover from stress more completely than the group B with the reduced WAI according to their HRV (p = 0.051). No significant differences were observed between the groups in stress at work or stress during leisure time. The variables with significant differences between the group A and B were observed in all floors of the Work ability house (Figure 2). The variables can be considered relevant predictors of the perceived work ability, and they also supported the validity of the theoretical framework of the Work ability house.

4 DISCUSSION In spite of the small number of the teachers and superiors in this study, several significant both ergonomics i.e., work-related and individual predictors of perceived work ability assessed with the WAI were observed. The incomplete recovery of the teachers and superiors with the reduced WAI from stress during sleep was a novel finding. The increasing demands of work emphasize the need of the adequate recovery because a worker should be productive, creative and innovative in the current 24-hour society. The long-term incomplete recovery may impair health, work ability and wellbeing. Interesting findings were also the significance of overweight and walking capability related to the reduced WAI. Recently the overweight related to the reduced WAI was indicated to be a powerful predictor for the early exit from the working life (Hopsu et al., 2005). Therefore, the importance of the overweight should not be underestimated, because it seems to affect work ability in multiple ways. The promotion of health, work ability and wellbeing should include the information on nutrition and physical activity in all phases of the work career. The present results supported the theoretical framework of work ability i.e., the Work ability house developed and reported by Ilmarinen (2006) although the WAI evaluates perceived individual work ability and focuses on health oriented items. Furthermore, the strict categorization of variables to ergonomics or work-related or individual ones is sometimes difficult regarding, for instance, to professional competence. The workforce is ageing rapidly. In Finland, in the next 15 years approximately 900,000 workers, about 40% of workforce, will exit from the working life (Ilmarinen, 2006). In the near future the main challenge will be to keep older people at work. This requires much effort to promote health, 79

Society Family

Close community Work Ability Index (WAI) Work Organizing of the work (p
0.036) Functionality of the work (p
0.041)

Motivation (p
0.14) Professional competence (p
0.060) Health and functional capacity Medication (p
0.001) 2-km walking test (index) (p
0.003) WSQ (p
0.010) BBI (p
0.013) Fat% (p
0.014) BMI (p
0.016) Muscle strength of trunk flexors (p
0.048) Stress during sleep (p
0.051)

Occupational safety

Occupational health care

Values

Figure 2. The significant predictors of the perceived work ability assessed with the Work ability index (WAI) in the framework of the Work ability house (Ilmarinen, 2006).

work ability and wellbeing by improving ergonomics and individual determinants of the work ability. More studies on work ability and, particularly, on recovery are necessary. REFERENCES Elo, A-L., Leppänen, A., Lindström, K., and Ropponen, T., (1992). Occupational Stress Questionnaire: User’s Instructions, Reviews, Vol. 19. Finnish Institute of Occupational Health, Press of Vammala, Helsinki, Finland. Gill, K., Krag, M.H., Johnson, G.B., Haugh, L.D., and Pope, M.H., (1988). Repeatability of four clinical methods for assessment of lumbar spinal motion. Spine 13(1): pp. 50–53. Hopsu, L., Leppänen, A., Ranta. R., and Louhevaara, V., (2005). Perceived work ability and individual characteristics as predictors for early exit from working life in professional cleaners. In: Costa, G., Goedhard, W.J.A., and Ilmarinen, J. (eds) Proceedings of the 2nd International Symposium on Work Ability. Verona, Italy, pp. 84–88. Ilmarinen, J., Tuomi, K., and Klockars, M., (1997). Changes in the work ability of active employees over an 11-year period. Scan. J. of Work, Environment & Health 23; Suppl 1: pp. 49–57. Ilmarinen, J., (2001). Aging workers. Occupational and Environmental Medicine 58: pp. 546–552. Ilmarinen, J., (2006). Towards a longer worklife. Ageing and the quality of worklife on the European Union. Jyväskylä. Finnish Institute oh Occupational Health. Gummerus, Helsinki, Finland.

80

Laukkanen, R.M., Kukkonen-Harjula, T.K., Oja, P., Pasanen, M.E., and Vuori, I.M., (2000). Prediction of change in maximal aerobic power by the 2-km walk test after walking training in middle-aged adults. International Journal of Sports Medicine 21(2): pp. 113–116. Louhevaara, V., (1999). Participatory ergonomics as a measure for maintaining work ability. In: Ilmarinen, I. and Louhevaara, V. (eds), FinnAge – respect for the aging: Action programme to promote health, work ability and well-being of aging workers in 1990–1996. People and Work. Research reports 26. Finnish Institute of Occupational Health, Helsinki, Finland. Näätänen, P., Aro, A., Matthiesen, S.B., and Salmela-Aro, K., (2003). Bergen burnout indicator 15. Edita, Helsinki. Finland. Pollock, M.L., and Willmore, J.H., (1990). Exercise in health and disease. Evaluation and prescription for prevention and rehabilitation. WB Sauders Co. Philadelphia, USA. Punakallio, A., (2004). Trial-to-trial reproducibility and test-retest stability of two dynamic balance tests among male firefighters. International Journal of Sports Medicine 25: pp. 163–169. Robertson, A. and Tracy, C.S., (1998). Health and productivity of older workers. Scandinavian Journal of Work, Environment & Health 24 (2): pp. 85–97. Shephard, R., (2000). Aging and productivity: some physiological issues. Int. J. of Industrial Ergonomics 25: pp. 535–545. Sillanpää, J., (2007). Electromyography for assessing muscular strain in the workplace. People and Work Research Reports 79. Finnish Institute of Occupational Health, University Press of Tampere, Helsinki, Finland. Tuomi, K., Ilmarinen, J., Jahkola, A., Katajarinne, L., and Tulkki, A., (1998). Work ability index. Finnish Institute of Occupational Health, Press of Vammala, Helsinki, Finland. www.europe.omron.com (15 June 2007). www.firstbeattechnologies.com (15 June 2007). www.suunto.com (15 June 2007).

81

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

An assessment of the Work Ability Index (WAI) and its usefulness in predicting and promoting continued work in staff employed by a major UK Charity Tara Reilly, A. Rees & M. Tipton Department of Sport & Exercise Science, Institute of Biomedical & Bio-molecular Sciences, University of Portsmouth, Hampshire, United Kingdom

ABSTRACT: Objective – The aim of this project was to evaluate the work ability index (WAI) through feedback from the users, and compare the WAI score with job satisfaction and retirement intentions. Methods – All questionnaires were completed online by participants (N = 98) employed at the Headquarters of a major national charity in the South of England. Relationships between measured variables were examined by correlation coefficients (P < 0.05). Results – A mean WAI score of 40.6 reflected good workability, though it was noted that workability relating to the mental demands of work was lower than that of physical demands. No significant associations were found between workability and age, level of educational attainment, or occupational task. Employees were asked to rate their job satisfaction, to report a “desirable” retirement age and to estimate ‘likely’ retirement age. The WAI score did not correlate with their responses. The user’s evaluation of the WAI led to the conclusion that the instrument was not too technical, personal, irrelevant or vague. However the participants felt that a measurement of “work stress” should be included in the WAI. Conclusion – A more explicit question on job satisfaction should be incorporated into the WAI for similar working groups. The major factor determining the poor/moderate workability in this cohort was health. The WAI was not able to predict intention to retire in this population. Keywords:

workability, workability index (WAI), occupation, job satisfaction, health, retirement

1 INTRODUCTION “Workability”, as the name suggests, is about the capability of an individual to do work. It is a concept that has already gained favor in Scandinavia and elsewhere in Europe, but to date has not been extensively used or critically evaluated in the UK. Indeed, the only group to use the WAI in the UK and publish in the open literature is the National Health Service (Nachiappan and Harrison, 2005). In the past decade there has been a trend towards a lower average retirement age in the Western world, and this has been shown to result in a higher cost to society in order to support these retirees (Kilbom, 1999). Attempts to prolong working life have included a number of interventions notably a reduction in early retirement benefits, changes in normal retirement age, voluntary delayed retirement, and the introduction of policies to prevent age discrimination in employment (Kilbom, 1999). It has been estimated that 70% of people in the UK retire before reaching the age of 60 (Mein et al., 2000). 83

The changing demographic towards an ageing population, the financial implications of premature exit from the workforce, and the ability and desire to continue working, have provoked the introduction of EU legislation against ageism. Attention is now focused on initiatives which aim to enable and encourage older workers to continue working if they desire. These have included studies into the physiology of ageing, and the socio-economic benefits of extending working life. The present study was concerned with the measurement of workability and in particular, focused on the utility of the questionnaire-based Workability Index (WAI). It has been argued that the WAI may be beneficial as an aid to maintain workability; a tool to inform occupational health care; and a predictor of work-related disability and retirement intention (Tuomi et al., 1998). Thus, the WAI may be a potentially useful tool in the overall assessment of work-related health and performance. The WAI is a questionnaire-based tool designed to present an overall picture of “workability”. The WAI has been defined as “an instrument to be used in occupational health care. It reveals how well a worker is able to perform his or her work” (Tuomi et al., 1998). The authors of the WAI argue that the instrument should be deployed as one of the methods for assessing workability in health examinations and workplace surveys. Ilmarinen et al. (1991) state that the WAI can be used as a tool to determine which workers need the support of occupational health care. They claim that the WAI can predict the threat of disability and retirement, and help in the prescription of appropriate occupational health interventions aimed at extending working life. This in itself suggests that the WAI alone is not sufficient to fully assess workability and that supplementary health and work place surveys must be performed. The aims of the present study were to employ the WAI with a working group and obtain feedback on the Index in order to identify its limitations and strengths. The WAI was also administered simultaneously with a job satisfaction question and questions regarding the intention to retire. The relationship between these measures was examined. It was hypothesized that the WAI was capable of predicting the future work intentions and job satisfaction of current employees working for a major UK Charity.

2 METHODS All participants were volunteers and the protocol received ethical approval from the University of Portsmouth. The WAI was initially piloted with employees in the Department of Sport and Exercise Science at Portsmouth University to determine initial opinions of the questionnaire. The pilot study resulted in the administration of 26 WAI questionnaires distributed as a hard copy and returned in a confidential manner. Of these, 18 were returned (69% response rate). The respondents consisted of 12 academics, four technicians, and two administrative staff. The mean WAI score (46) for this sample was positioned within the good-excellent (40–48) range (Tuomi et al., 1998). Individual structured interviews with respondents in the pilot study were undertaken with the aim of further evaluating the usability of the WAI index. Most individuals expressed concern regarding anonymity with hard copy, paper questionnaires. They felt these may be traceable and therefore could not be totally honest in their responses. Consequently, an online version of the WAI was developed to protect anonymity and to facilitate completion. The finalized e-questionnaire consisted of four elements: informed consent – without this the volunteer could not proceed to the questionnaire; a background to the study – adapted from the workability questionnaire; the WAI questionnaire itself; and an evaluation of the WAI questionnaire. The evaluation component included items designed to determine whether the WAI had overlooked any critical elements of working life. It also included two questions regarding current retirement intentions (Moline, 2005). Finally, there was a question on job satisfaction (Dolbier et al., 2005) “how do you feel about your job as a whole” rated on a 1–7 scale (extremely dissatisfied – extremely satisfied). The second phase of the study was to administer the e-questionnaire version of the WAI to employees working for at the Head Quarters of a major UK Charity based in the South of England. 84

WAI (7 poor–49 excellent)

55

y
0.518x  38.569 R2
0.0259

50 45 40 35 30 25 0

1

2

Extremely Dis-satisfied

3

4

5

Job satisfaction

6

7

8

Extremely satisfied

Figure 1. The relationship between job satisfaction and WAI scores (N = 83).

This company was largely administration based with job roles ranging from operations, funding, communication, technical, finance, corporate, and human resources. The website was available for approximately 400 employees to complete the questionnaires at their discretion between July and September 2006. Employees were informed of the questionnaire website, password and the aim of the project by an “all-staff ” email from the company. Descriptive and correlation statistics were then used in the analyses of the data. Significance was accepted at P < 0.05.

3 RESULTS In total, 122 questionnaires were returned with 98 completed WAI questionnaires (30% response rate). The group (N = 98) produced a mean WAI result of 40.6 and 85% were primarily engaged in mental work. Mean workability in this cohort appears to be low for 50–55 year olds but higher for those aged 55–60 compared to normative data. However, there was no significant relationship between age (19–57 years) and WAI results. There did not appear to be any relationship between role at work or level of educational attainment (secondary-doctorate) and measured work ability index scores. The WAI was also compared with job satisfaction. The relationship between the WAI results and job satisfaction resulted in an R = 0.16, R2 = 0.03 (NS, P = 0.146) (Figure 1). Finally, the study examined the ability of the WAI to predict retirement intentions. In particular, participants were asked when they would ideally like to retire (R = −0.084, NS; P = 0.456) (Figure 2) and when they actually saw themselves retiring in reality (R = 0.183, NS; P = 0.098) (Figure 3). The WAI did not reveal a significant correlation with either of these factors, and thus was not predictive of either desirable or likely age of retirement.

4 DISCUSSION In their recommendations for prolonging a healthy working life, Illmarinen et al. (1991) suggested that work demands should change with age and that workability should be regularly monitored by various mechanisms including the WAI, indicating that the WAI can predict the threat of disability and retirement. For practical application of the WAI, it is important to identify employees with low workability and explore the underlying reasons for these observations. Ideally it would have been attractive to identify these subjects and interview them in order to establish if they are actually functioning poorly at work, and therefore at risk of early exit from work. This opportunity was 85

52 48

WAI (7–49)

44 40 36 32 28 24 20 45–50

50–55

55–60

60–65

65–70

When would you like to retire (years)

Figure 2. The relationship between “when you would like to retire” and WAI scores, demonstrated as (N = 81). The vertical lines shown here represent the boundary for each of the five year categories. 55 50

WAI (7–49)

45 40 35 30 25 20 50–55

55–60

60–65

65–70 75

When do you see your self retiring (years)

Figure 3. The relationship between “when do you see yourself retiring” and WAI scores, demonstrated as NS (N = 83). The vertical lines shown here represent the boundary for each of the five year categories.

precluded by the need to maintain confidentiality. However, to further explore the problems of those with low workability, further analyses were conducted on the data. Individuals with poor/moderate WAI self rated their workability at approximately seven out of 10, which is still reasonably high. Mental resources were also rated high relative to job demands in this group and therefore, and it may be speculated that the major factor in determining workability in this cohort was the reporting of poor health. Self-rated workability accounts for 15% of the variation in the overall scoring (49 points) in the workability index. This indicates that the WAI tends to be disproportionately influenced as a result of perceived workability rather than by more objective measures. These results also suggest that individuals may not be capable of accurately perceiving their WAI score. 86

Workability did not significantly correlate with age, occupational type or level of educational attainment for this working group. The observation that self rated mental workability was lower than physical self-rated workability indicates that those with more physically demanding jobs were better matched, or at least perceived themselves as more capable at performing their daily occupational tasks. However, physical endeavors, as well as monotonous work and repetitive movements, have been shown to increase the risk of early retirement among those below 45 years of age (Kilbom, 1999), and it may therefore be expected that physical jobs result in lower WAI scores. This was not found in this study, although the physicality of the roles in this cohort may not be compatible with those studied elsewhere. Dolbier et al. (2005) measured job satisfaction with the question (“taking everything in to account how do you feel about your job as a whole”) in a cohort (N = 745) of public sector employees working for the Texas Department of Human Services. The jobs assessed were mostly professional (63%) but also included administrative, technical and administrative support, much like the work roles within the present study. The responses obtain from this working group indicated that this population was very satisfied in their work (Figure 1). These results suggest that for these employees, the WAI does not predict job satisfaction and therefore strengthens the argument that perhaps a more explicit question on job satisfaction should be incorporated into the WAI.

5 CONCLUSION The hypothesis that the WAI was capable of predicting the future work intentions of current employees working for a major UK Charity is not supported. The hypothesis that the WAI was a function of job satisfaction was also not supported. This may be unsurprising given that the decision to retire early has been found to be multi-factorial and depends on issues such as employment grade, state of heath, and level of job satisfaction (Mein et al., 2000). Not all of these factors are captured in the WAI. The authors recognise that the cross-sectional nature of this study confers limitations, particularly in relation to generalisation. Notwithstanding these limitations, it is felt that the study provides additional insight in to the utility of the WAI and confirms that the complexities of the concept of workability provides significant methodological and measurement challenges. REFERENCES Dolbier, C., Webster, C., McCalister, K., Mallon, M., and Steinhardt, M., (2005). Reliability and validity of a one-item measure of job satisfaction. American Journal of Health Promotion, 19: pp. 194–198 Ilmarinen, J., Tuomi, K., Eskelinen, L., Nygard, C., Huuhtanen, P., and Klockars, M., (1991). Summary and recommendations of a project involving cross-sectional and follow-up studies on the aging worker in Finnish municipal occupations (1981–1985). Scand Journal of Work Environmental Health, 17(suppl 1): pp. 135–141 Kilbom, A., (1999). Evidence-based programs for the prevention of early exit from work. Experimental Aging Research, 25: pp. 291–199 Mein, G., Martikainen, P., Stansfeld, S., Brunner, E., Fuhrer, R., and Marmot, M., (2000). Predictors of early retirement in British civil servants. Age and Ageing, 29: pp. 529–536 Molinie, A., (2005). Feeling capable of remaining in the same job until retirement. International Congress series on work ability, 1280: pp. 112–117 Nachiappan, N., and Harrison, J., (2005). Work ability among health care workers in the United Kingdom. International Congress series on work ability, 1280: pp. 286–291 Tuomi, K., Ilmarinen, J., Jahkola, A., Katajarinne, L., and Tulkki, A., (1998). Work Ability Index. 2nd revised edition, edited by Rautoja, S and Pietilainen R (Finland, FIOH).

87

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Individual and work related determinants of work ability in white-collar workers T.I.J. van den Berg & S.M. Alavinia Department of Public Health, Erasmus MC, University Medical Center Rotterdam, The Netherlands

F.J. Bredt & D. Lindeboom Lifeguard, Utrecht, The Netherlands

L.A.M. Elders & A. Burdorf Department of Public Health, Erasmus MC, University Medical Center Rotterdam, The Netherlands

ABSTRACT: Objectives – The purpose of this article is to explore the associations of psychosocial factors at work, life style, and stressful life events on work ability among white-collar workers in commercial services. Methods – A questionnaire was used to assess psychosocial factors at work and stressful life events and in an examination the physical condition was measured among 1,141 workers. Linear regression analysis was conducted on the work ability index score (WAI). Results – Work ability was strongly associated with psychosocial factors at work such as teamwork, stress handling, and self-development and, to a lesser extent, with stressful life events, and lack of physical activity in leisure time. Conclusion – The strong associations between psychosocial factors at work and work ability suggest that in this study population health promotion should address working conditions rather than individual life style factors.

1 INTRODUCTION With the aging of the European work population, the prevalence of work-relevant symptoms and diseases will increase. Therefore, the role of (functional) health in working life becomes more important. Within this framework, the concept of work ability has been developed as an important tool. The work ability concept is based on the assumption that work ability is determined by an individual’s perception of the demands at work and the ability to cope with them. Former research focused mainly on physically demanding occupations. Therefore, it remains unclear whether in mentally demanding jobs the relative importance of personal and work-related factors is similar to their well-known contribution in physically demanding jobs. The purpose of this study is to explore the associations of psychosocial factors at work, stressful life events, and life style on work ability among white-collar workers. 2 METHODS Subjects – In the period between 2003-2007, a total of 2,666 white-collar workers in commercial services were invited for a health examination. Complete data sets were obtained for 1,141 (42.8%) subjects. 89

Table 1.

Results regression analysis: univariate and multivariate effects of individual and work related factors on work ability in commercial services (n = 1,141). WA Univariate model

Individual characteristics Age Male Psychosocial factors at work High teamwork Intermediate teamwork Low teamwork High stress-handling Intermediate stress-handling Low stress-handling High self-development Intermediate self-development Low self-development Stressful life events Low life change score Intermediate life change score High life change score Life style Lack of moderate physical activity Lack of vigorous physical activity Current smoker Problematic alcohol use Physical examination Normal body weight (BMI <25) Overweight (BMI 25–30) Obese (BMI ≥30) High VO2 max Intermediate VO2 max Low VO2 max High Biceps strength Intermediate Biceps strength Low Biceps strength

WA Multivariate model

β

SE

β

SE

−0.07* 2.08*

0.02 0.32

−0.10* 2.01*

0.02 0.30

0 −1.63* 4.02* 0 −1.52* −4.39* 0 −1.67* −4.21*

0.35 0.36

0 −0.28 −1.24* 0 −0.90* −2.72* 0 −0.94* −2.38*

0.36 0.41

0 −0.86* −1.25*

0.37 0.37

0 −1.07* −1.91*

0.34 0.34

0.44 −0.76* −0.64 −0.47

0.33 0.36 0.45 0.77

n.s −0.73* n.s n.s

0.33 0.70 0.37 0.37

0 n.s n.s n.s n.s

0.38 0.37

n.s n.s

0 −0.43 −1.95* 0 0.25 −0.56 0 −0.56 −0.76*

0.35 0.34 0.36 0.35

0.36 0.41 0.37 0.39

0.32

n.s = not significant, *p < 0.05

Measurement – Work ability was measured with the Work Ability Index (WAI) (Tuomi et al., 1998). Psychosocial factors at work were measured by the Stress monitor (Petri et al., 2001). The multi-item dimensions teamwork, stress handling, and self-development were measured on 5-point scales varying from ‘totally disagree’ to ‘totally agree’. The occurrence of stressful life events in the past 12 months was measured using a shortened (25 life events) Social Readjustment Rating Questionnaire (SRRQ) (Holmes and Rahe, 1967). Life style factors were measured with the Dutch version of the Stanford Wellness Inventory (Sallis et al., 1985). In accordance with the protocol of the American College of Sports Medicine (ACSM, 1975) weight, height, biceps strength, and cardio respiratory fitness were measured. Cardio respiratory fitness was assessed by a 12-minute sub maximal bicycle ergometer test. Statistics – The effects of individual and work-related characteristics on the outcome variable work ability was investigated with linear regression analysis. Variables with a p-value of 0.10 or less in the univariate model were selected for multivariate analysis. In the multivariate regression analysis independent variables with a p-value of 0.05 or less were retained in the final model. 90

3 RESULTS The study population included 769 men (67%) and 372 (33%) women. The mean age was 37.2 years (sd 9.3). The distribution of excellent, good, moderate and poor work ability was 42.8%, 45.4%, 9.7%, and 2.1%, respectively. In the univariate analyses (Table 1) work ability was statistically significantly influenced by psychosocial factors at work, stressful life events, lack of vigorous physical activity, and obesity. The multivariate model explained 27% of the variance in work ability. The influence of stressful life events increased in the multivariate model. No significant interaction was observed for age, sex, and psychosocial factors at work.

4 DISCUSSION This study showed that work ability of white-collar workers in commercial services industry was strongly associated with psychosocial factors at work, such as teamwork, stress handling, selfdevelopment, and, to a lesser extent, with stressful life events, and lack of vigorous physical activity. These factors together explained 27% of the total variance in work ability. An increase in age of 40 years decreased work ability with 4 points, which means 8% of the maximum score. The combined effect of the psychosocial factors at work is approximately 1.5 fold the effect of 40 years of ageing, which indicates a rather modest influence of age on work ability. Lack of vigorous physical activity decreases the WAI score with only 0.73 points, which is no more than 1.5% of the maximum score. The lack of significant results for maximum oxygen uptake and biceps strength in the multivariate model supports the hypothesis that in mentally demanding jobs a good physical condition is usually not required to meet the work demands and, thus, will have a limited influence on work ability. The results of the current study outline the importance of work-related factors in white-collar workers, with regard to work ability. The combined impact of psychosocial factors was much stronger than individual factors, and these factors are amendable to change. Health promotion in this population should address primarily working conditions rather than individual life style factors.

REFERENCES American College of Sports and Medicine. 1975 ACSM’s guidelines for exercise testing and prescription. cop. 2000, 6th edition. Senior ed. Franklin, B.A.; ass.ed. Whaley, M.H. et al. (Lippincott Williams & Wilkins, Philadelphia, PA). Holmes, T.H. and Rahe, R.H., (1967). The social readjustment rating scale. Journal of Psychosomatic Research, 11: pp. 213–218. Petri, C., van der Velden, P.G. and Kleber, R.J., (2001). Hoe gaat het met u? – een praktische Stress monitor [How are you? – A practical stress monitor]. (Thema, Zaltbommel). Sallis, J.F., Haskell, W.L., Wood, P.D., Fortmann, S.P., Rogers, T., Blair, S.N. et al., (1985). Physical activity assessment methodology in the Stanford five-city project. American Journal of Epidemiolgy 121: pp. 91–106. Tuomi, K., Ilmarinen, J., Jahkola, A., Katajarinne, L. and Tulkki, A., (1998). Work ability Index, 2nd revised edn, Occupational Health Care 19. (Finnish Institute of Occupational Health, Helsinki).

91

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Workplace trauma exposure, emotional imbalance and work ability Giuseppe Paolo Fichera, L. Neri, S. Sartori & G. Costa Department of Occupational and Environmental Health, University of Milan; IRCCS Maggiore Hospital, Mangiagalli and Regina Elena Foundation, Milan, Italy

ABSTRACT: Introduction – Exposure to workplace robbery is a very common and traumatic event among workers dealing with the public. Data is scarce about the association between exposure to robbery, robbery-related Post-traumatic Stress Disorder (PTSD) and work ability. Objective – To assess the association between workplace robbery, robbery-related PTSD, emotional imbalance and work ability in a sample of pharmacy workers. Methods – 136 pharmacy workers – 90 robbery victims and 46 non victims – were recruited from the Milan metropolitan area. They completed a self-report questionnaire including sociodemographic characteristics, history of robberies, a self-report version of the CAPS-I, General Health Questionnaire, Beck Depression Inventory, Work Ability Index. Results – No differences were found between victims and non victims for GHQ and BDI; WAI scores of victims were significantly lower than non-victims. Exposure to robberies was associated with lower WAI in a multivariate analysis. Ten victims reported Full or Partial PTSD following the robbery. Victims with PTSD reported a lower WAI and a higher GHQ and BDI than non-PTSD victims. Conclusions – Workplace robbery may have a mild but long-lasting effect on workers’ work ability. This effect is higher for those developing PTSD following robbery. Further research and early intervention programs are needed to promote recovery of psychophysical health, well-being and work ability after a traumatic event in the workplace.

1 INTRODUCTION Robbery-related assaults in the workplaces are very common both in developed and developing countries (Chappel and Di Martino, 2000). Italy is one of the European countries at medium-high prevalence (TRANSCRIME, 2007), with chemist shop owners and employees being among the workers particularly at risk. Exposure to workplace robbery is a traumatic event that could affect emotional well-being of victims. The literature on this topic is not large as researchers have directed their attention mostly toward victims of major traumatic events, such as war, deportation, kidnapping, catastrophes and disasters. Less research emphasis has been put on victims of apparently less dramatic events such as armed robbery in the workplace (Kamphius et al., 1998). However, the few existing studies are consistent in proving robbery – such as other kind of violent and threatening events occurring in the workplace – as an experience favouring the onset of Posttraumatic Stress Disorder (PTSD) symptoms, and a consequent reduction of global functioning of the worker in terms of well-being, quality of life and work ability (Nyberg et al., 2003; MacDonald et al., 2003; Admundson et al., 1998). 93

The specific association between work-related Posttraumatic Stress Disorder and work ability is an underinvestigated topic, although psychiatric disorders are among the most commonly certified diagnoses in the cases of sickness absence and disability pension (Helsing and Wahlström, 2004).

2 OBJECTIVES The main objectives of this study were to test the following hypotheses: • Work ability of workers victims of workplace robbery is significantly lower than that of their colleagues not exposed to such traumatic workplace experience. • Post-traumatic Stress Disorder (PTSD) and emotional imbalance are relatively common in workplace robbery victims. • Workers with work-related PTSD report significantly lower work ability and higher emotional imbalance than workers without PTSD.

3 METHODS 3.1 Sample and procedure All chemist shops located within the Milan metropolitan area were contacted by e-mail, thanks to the collaboration of the “Associazione Chimica Farmaceutica Lombarda Fra Titolari di Farmacia” 1 . In the e-mail, aims and procedure of the study were reported and all workers – irrespective of their status of victims or non victims of robbery – were invited to participate in the survey. In the e-mail it was emphasized that anonymity would be guaranteed and collected data would be treated only at a collective level. As a whole, 136 chemist shop workers took part in the study. Ninety were victims of at least one robbery (60 female and 30 male), while the remaining 46 resulted as never exposed (40 female and six male). All 136 subjects have been met personally at their workplace by the first author who delivered the questionnaire, further explained survey purposes and arranged the date for withdrawal of the compiled instrument. All subjects signed the informed consent before questionnaire completion. All 136 questionnaires were correctly completed and collected. 3.2 Measures Background section: socio-demographic variables such as sex, age, education work seniority and job position. Robbery and robbery-related PTSD section: this section was limited to subjects victims of robbery. It included open-ended questions on the number of robberies experienced during working life and on the characteristics of the more recent events, and also a self-report version of the Clinician Administered Post-traumatic Stress Disorder Scale (CAPS, Blake et al., 1986) for assessment of PTSD symptoms severity and PTSD related to robbery. The CAPS self-report version, created on purpose for the present study, was previously tested on 10 patients referred to the outpatient service at the “IRCCS Maggiore Hospital”2 for suspected PTSD. A satisfactory agreement between the clinical evaluation and the self-report assessment was observed. General Health Questionnaire (GHQ-12, Goldberg, 1972): A self-report measure of stress and emotional imbalance symptoms. The higher the score (range 0–36) the higher the emotional imbalance. The 12-item version was used in this study. 1 Regional 2 IRCCS

association between pharmacy owners of Northern Italy Maggiore Hospital Mangiagalli and Regina Elena Foundation, Milan, Italy

94

Beck Depression Inventory (BDI, A. Beck et al., 1961): A self-report 21-item measure of depression and depressive symptoms. The higher the score (range 0–63) the higher the severity of depression. Work Ability Index (WAI, Tuomi et al., 1998): A measure of health-related work ability. A higher score (range 7–49) indicates a better work ability. The score is usually classified in four levels of work ability, i.e. Excellent (44–49), Good (37–43), Moderate (28–36) and Poor (7–27). 3.3 Statistical analysis The statistical packages STATA 9.2 and SPSS 14.0 for windows were used for data analysis. T-test or Mann-Whitney test for continuous variables and chi-square test (with Yates’ continuity correction for 2 × 2 tables) or Fischer’s two-way Exact Test were used where appropriate. In order to test the association between WAI and workplace robbery exposure a multiple linear regression analysis was also conducted. We adjusted for sex, gender, education and emotional imbalance (BDI score). 4 RESULTS Table 1 reports the main socio-demographic characteristics of the two groups. The victims were older (t = 2.04; p < .05), were more likely male (χ2 = 6.44; p < .05) and chemist shop owners (χ2 = 9.81; p < .01), and had a longer work experience (t = 3.14; p < .01) compare to non victims. No differences were found for marital status and education. The mean number of robbery during working life for the victims was 4.44 (SD = 4.30), and the mean time since the last robbery was 40 months (SD = 50.10). The mean WAI score of the overall sample was 42.15 (SD = 4.62). According to the WAI categorical classification, 63 subjects (47%) reported Excellent Work Ability, 59 (44%) Good Work Ability, 12 (9%) Moderate Work Ability while none reported Poor Work Ability.

Table 1.

Sample characteristics.

Sex Male Female Job Position Owners Employees Education University degree No degree Marital Status Single Married Divorced/Widow

Overall sample (n = 136)

Victims (n = 90)

N

%

N

%

N

%

χ2

36 100

26.47 73.53

30 60

33.33 66.67

6 40

13.04 86.96

6.44∗

26 110

19.12 80.90

24 66

26.67 73.30

2 44

7.69 92.30

9.81∗∗

116 20

85.29 14.71

75 15

83.73 16.27

41 5

89.47 10.53

ns.

46 76 14

33.82 55.88 10.29

23 57 10

25.56 63.33 11.09

23 19 4

50.00 41.30 8.70

ns.

SD

Mean

SD

Mean

SD

t

11.65 12.11 – –

44.37 19.20 4.40 40.00

11.06 11.22 4.30 50.10

39.96 12.52 – –

12.38 12.67 – –

2.04* 3.14** – –

Mean Age Work seniority No. of robberies Mean time from last robbery (mo)

42.88 16.94 – –

*p < .05 **p < .01 ***p < .001

95

Non victims (n = 46)

Table 2.

Differences between victims and non victims in WAI, GHQ and BDI scores. Victims (n = 90)

WAI categories Moderate Good Excellent WAI score GHQ score BDI score

Non victims (n = 46)

N

%

N

%

χ2

11 40 38

12 45 43

1 19 25

2 42 46

ns.

Mean

SD

Mean

SD

t

42.15 11.25 6.29

4.62 4.62 6.53

44.21 10.67 5.74

3.23 3.67 5.45

−2.67** ns. ns.

*p < .05 **p < .01 ***p < .001 Table 3.

Differences between victims with Full/Partial PTSD and victims without PTSD in WAI, GHQ and BDI scores.

WAI categories Moderate Good Excellent WAI score GHQ score BDI score

Full/Partilal PTSD (n = 10)

No PTSD (n = 80)

N

%

N

%

χ2

5 5 0

50 50 0

6 35 38

8 44 48

∗

Median

IQR

Median

IQR

U

35.3 21.0 12.5

7.0 5.0 13.5

43.0 10.0 3.5

6.5 7.0 7.0

17.79∗∗

107*** 53.5*** 67.0***

*p < .05 **p < .01 ***p < .001

WAI score was not related to gender (t = −.44; ns.) and age (Pearsons’ correlation coefficient = −.08; ns.). Table 2 reports differences in WAI, GHQ and BDI scores between victims and non-victims tested by means of bivariate analyses. No differences were found between the two groups for GHQ and BDI scores, and also for categorical WAI; however, mean WAI score of robbery victims was significantly lower than that of non-victims (t = −2.67; p < .01). The association between WAI and exposure to robberies was further tested with a multivariate linear regression analysis. Exposure to workplace robberies was independently associated with lower WAI (β = −0.19, p < .05). Age, gender and education were not associated with WAI. BDI was independently and negatively associated with WAI (β = −0.52, p < .001).The full model explained 33% of variance in WAI. Inclusion of exposure in the model accounted for an additional 3% of total WAI variance (significance of change in R2 : p < .05). BDI, GHQ and WAI scores of the victims group were confronted in five sub-groups formed on the basis of time elapsed since the last robbery: <3 mo; 3–6 mo; 6–12 mo; 12–24 mo; >24 mo. A significant score decrease over time since the last robbery was found for GHQ (F(4,89) = 2.63; p < .05) and BDI (F(4,89) = 2.91; p < .05); no significant differences were found in WAI scores among the five sub-groups (F(4,89) = 1.40; ns.). 96

Table 4.

Correlations between number of PTSD symptoms, WAI, GHQ and BDI scores in robbery victims. 1

1. No. of DPTS symptoms 2. GHQ 3. BDI 4. WAI

2

3

4

– −.56***

–

– .59*** .52*** −.41***

– .70*** −.54***

*p < .05 **p < .01 ***p < .001

Six victims (7%) met the criteria for a full PTSD diagnosis and four (4%) for a partial PTSD diagnosis. Table 3 shows the difference in WAI, GHQ and BDI scores between victims with full/partial PTSD and victims without such diagnosis. Victims with Full/Partial PTSD reported a lower WAI as far as both the mean score (U = 107; p < .001) and the categorical score (χ2 = 17.79; p < .001) were concerned, and a higher GHQ (U = 53.5; p < .001) and BDI (U = 67.0; p < .001). Table 4 reports correlations between the number of PTSD symptoms, BDI, GHQ and WAI scores. The number of PTSD symptoms was positively correlated with GHQ (r = .59; p < .001) and BDI (r = .52; p < .001) and negatively correlated with WAI (r = −.41; p < .001). GHQ and BDI were positively correlated (r = .70; p < .001) and were both negatively correlated with WAI (r = −.54; p < .001 and r = −.56; p < .001 respectively). 5 DISCUSSION Results of this study highlight a high work ability status in a sample of Italian chemist shop workers across all ages, and suggest that exposure to robbery might have only a limited effect on their work ability. However, despite its mildness, the effect may be long-lasting. Indeed, while emotional imbalance of robbery victims (as measured by means of the GHQ and the BDI) seems to recover long after the last robbery, work ability did not show such a recovery pattern. Hence, we could hypothesize that while as a whole a chemist shop worker maintains an excellent work ability over the years, a robbery-related event may cause an impairment that lasts for a long time. Since our results can be affected by recall bias and the effect observed is quite small a longitudinal study of incident robbery victims should be implemented to test this hypothesis. For a significant share of victims (11%), exposure to workplace robbery was followed by the onset of a Full or Partial Post-traumatic Stress disorder. Such prevalence is lower than that found in previous studies on groups of victims of severe trauma such as rape, war, concentration camp deportation, hostage-taking, war imprisonment and kidnappings (Kilpatrick et al., 1987; Bremner et al., 1992; Bisson et al., 1998; Carlson et al., 1991; Favaro et al., 1999; Favaro et al., 2000; Shalev et al., 1998; Solomon et al., 1994). However, 11% prevalence is much higher compared to that in the general population (1.3%; Perkonigg et al., 2000) and is even more significant if one considers the high prevalence of robbery-related assaults, both in developed and developing countries. Post-traumatic Stress disorder – Partial diagnosis included – seems to bear important negative effects, besides emotional imbalance, also on work ability of chemist shop workers. Indeed, victims with Full/Partial PTSD reported a moderate work ability level. This is consistent with literature reporting that psychiatric disorders affect functions of importance for work ability especially in jobs requiring high cognitive and social competence (Helsing and Wahlström, 2004). Moreover, these findings indicate the need of prompt rehabilitative interventions aimed at restoring work ability among victims of robberies, also considering that an eleven-year follow-up study demonstrated that low work ability is predictive of mortality and disability pension (Costa, 2003). Finally, the association between the number of PTSD symptoms and the levels of emotional imbalance and work ability indicates that PTSD symptoms may exert a negative effect on work 97

ability and emotional imbalance even before reaching the threshold for clinical diagnosis; in fact, the more the PTSD symptoms the higher the emotional imbalance and the lower the work ability. This is consistent with other few studies conducted on the relationship between health and work ability status following workplace robbery. Exposure to robbery in the workplace could bring about negative consequences at different levels – from psychophysical health to quality of life to work ability – and these levels of global functioning of the person are commonly related one to each other (Miller-Burke et al., 1999).

6 CONCLUSION Exposure to robbery in the workplace may have a mild but long-lasting effect on workers’ work ability. Moreover, for a significant share of victims, exposure to robbery is associated with the onset of Post-traumatic Stress Disorder, with these subjects resulting at an increased risk for severe impairment in work ability and emotional well-being. Therefore, wider epidemiologic studies are needed to assess the natural history and predictors of psychological distress and long-lasting workability impairments among victims of traumatic events at their workplaces. This study also provide the rationale for the development of early intervention programs in the workplaces aimed at reducing the incidence, severity and duration of psychiatric disorders following traumatic events, so to enable a faster recovery of psychophysical health, well-being, and work ability. REFERENCES Admundson, G.J., Norton, G.R., Allerdings, M.D., Norton, P.J., and Larsen, D.K., (1998). Posttraumatic stress disorder and work-related injury. Journal of Anxiety Disorders, 12(1): pp. 57–69 Beck, A.T., Ward, C., and Mendelson, M., (1961). Beck Depression Inventory (BDI). Arch of General Psychiatry, 4: pp. 561–571 Bisson, J.I., Searle, M.M., and Srinivasan, M., (1998). Follow-up study of British military hostage and their family held in Kuwait during the Gulf War. British J. Med. Psychology, 71: pp. 247–252 Blake, F., Weathers, L., Nagy, D., Kaloupek, G., Klauminzer, D., Charney, T., and Keane, S., (1990). A clinician rating scale for assessing current and lifetime PTSD: the CAPS-I. The Behaviour Therapist, 18: pp. 187–188 Bremner, D., Southwich, S., Brett, E., Fontana, A., Rosenheck, R., and Charney, D.S., (1992). Dissociation and posttraumatic stress disorder inVietnam combat veterans. American Journal of Psychiatry, 149: pp. 328–332 Carlson, B., and Rosser-Hogan, R., (1991). Trauma experience, post traumatic stress, dissociation and depression in refugees. American Journal of Psychiatry, 148: pp. 1548–1551 Chappell D., and Di Martino V., (2000). Violence at work. (Geneva: International Labour Office) Costa, G., (2003). Lavoro a Turni e Notturno. (Firenze: SEE) Favaro A., Degortes, D., Colombo, G., and Santonastaso, P., (2000). The effects of trauma among kidnap victims in Sardinia, Italy. Psychological Medicine, 30: pp. 975–980 Favaro, A., Rodella, F.C., Colombo, G., and Santonastaso, P., (1999). Post-traumatic stress disorder and major depression among Italian Nazi camp survivors: A controlled study fifty years later. Psychological Medicine, 29: pp. 87–95 Goldberg, D.P., (1972). The Detection of Psychiatric Illness by Questionnaire. (Oxford: Oxford University Press) Helsing, G., and Wahlström, R., (2004). Sickness absence and psychiatric disorders. Scandinavian Journal of Public Health, 32 (63): pp. 152–180 Kamphuis, J.H., and Emmelkamp, P.M., (1998). Crime-related trauma: Psychological distress in victims of bankrobbery. Journal of Anxiety Disorders, 12: pp. 199–208 MacDonald., H.A., Coltola, V., Flamer, S., and Karlinsky, H., (2003). Posttraumatic stress disorder in the workplace: a descriptive study of workers experiencing PTSD resulting from work injury. Journal of Occupational Rehabilitation, 13(2): pp. 63–77 Miller-Burke, J., Attridge, M., and Fass, P.M., (1999). Impact of traumatic events and organizational response: A study of bank robberies. J. of Occup. & Env. Med., 41(2): pp. 73–83

98

Nyberg, E., Stieglitz, R.D., Frommberger, U., and Berger, M., (2003). Psychological disorders after severe occupational accidents. Versicherungsmedizin, 55(2): pp. 76–81 Perkonigg, A., Kessler, R.C., Storz, S., and Wittchen, H.U., (2000). Traumatic events and post-traumatic stress disorder in the community: Prevalence, risk factors and comorbidity. Acta Psychiatrica Scandinavica, 101: pp. 46–59 Shalev, A., Freedman, S., Peri, T., Brandes, D., Sahar, T., Orr, S., and Pitman, R.K., (1998). Prospective study of Post traumatic Stress Disorder and Depression following trauma. American Journal of Psychiatry, 155: pp. 630–637 Solomon, Z., Neria, Y., Ohrya, A., Waysman, M., and Ginsburg, K., (1994). PTSD among Israeli former prisoners of war and soldiers with combat stress reactions: A longitudinal study. American Journal of Psychiatry, 151: pp. 554–559 TRANSCRIME, (2007). L’andamento delle Rapine in Europa dal 1995 al 2005. In: Ottavo Rapporto sulla Sicurezza del Trentino 2006/2007. (Trento: Giunta della Provincia Autonoma) Tuomi, K., Ilmarinen, J., Jankola, A., Katajarinne, L., and Tulkki, A., (1998). “Work Ability Index”. (Helsinki: Finnish Institute of Occupational Health)

99

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Work ability and all cause mortality: A 25-year longitudinal study among Finnish municipal workers Jorma Seitsamo & Rami Martikainen∗ Finnish Institute of Occupational Health, Helsinki, Finland

ABSTRACT: The main objective of this study was to examine longitudinally the associations between the work ability of middle-aged employees and all cause mortality. The study material is comprised of a 25-year longitudinal study of Finnish municipal workers which was conducted at the Finnish Institute of Occupational Health in 1981–1997. The dependent variable was the time from the baseline to the year of death or to the end of the follow-up. Work ability was measured by Work Ability Index (WAI). Health behaviour, characteristics of work, marital status, and occupational group were also used in the analysis. Kaplan-Mayer method was used to estimate the survivorship functions for the four levels of work ability index (excellent, good, moderate, poor). The hazard ratios of all cause mortality were estimated using Cox proportional hazard model with time varying covariates. The results showed that work ability index was an independent predictor of mortality even when other factors were included in the model. It seems evident that work ability has longterm effects which reach out to the retirement years. This indicates that besides lengthening of work career, promotion of work ability might also give extra years to life.

1 INTRODUCTION Work ability, measured by Work Ability Index (WAI) has proved to be a reliable predictor of morbidity and disability in studies concerning ageing workers (Ilmarinen and Tuomi, 2004, and Tuomi et al., 1997b). Good work ability also predicts well-being and functional status far beyond the retirement transition (Tuomi and Huuhtanen, 2001). There is also some evidence that work ability may predict mortality (Tuomi et al., 1997b). More studies are still needed to get information from the long-term effects of work ability to later-life health and mortality. The main objective of this study was to examine longitudinally the associations between the work ability of middle-aged employees and all cause mortality. The specific study questions were: 1. Does the level of Work Ability Index (WAI) with the categories excellent, good, moderate, and poor have an impact on mortality? 2. Is WAI associated with mortality even if other factors are taken into account? 2 MATERIAL AND METHODS 2.1 The study sample The study sample constitutes a follow-up questionnaire study of Finnish municipal workers which was conducted at the Finnish Institute of Occupational Health from 1981 to 1997 (Tuomi et al., ∗ This

paper is dedicated to the memory of the brilliant statistician Rami Martikainen

101

1997a). In 1981, a postal questionnaire was sent to 7,344 municipal workers in different areas of Finland. The respondents were born between 1923 and 1937. The entire age range in this study was from 45 to 74 years of age. A total of 6,257 persons responded to the first questionnaire (the response rate was 85.2%). Over the follow-up, 715 persons deceased and 1,725 did not complete the questionnaire at each of the time points required. In 1997, a total of 3,815 persons had responded to all four (1981, 1985, 1992, 1997) cross-sectional questionnaires (the response rate was 69% of the living participants who responded to the first questionnaire). In 2006, the vital status and dates of death were added to the data. By then, 1,644 respondents had died. 2.2 Methods Statistical analysis. The Kaplan-Mayer method was used to estimate the survivorship functions for the four levels of work ability index (excellent, good, moderate, poor). The hazard ratios of all cause mortality were estimated using Cox proportional hazard model with time varying covariates. 2.3 Study variables The dependent variable was the time from the beginning of work career to the year of death or to the end of the follow-up. The independent variables were work ability, measured by Work Ability Index (WAI), health behaviour (i.e. alcohol consumption, smoking, physical exercise), and characteristics of work (responsibility for others, physical workload). Gender, marital status (unmarried, married, unmarried but co-habitating, separated, divorced, widow/widower), socio-economic group (bluecollar, lower-level white-collar, upper-level white-collar workers), and the age at which the person began his/her work career were also used in the analysis. Information regarding the main causes of death was acquired from Statistics Finland. In this study, the main interest was on all cause mortality. The dates and causes of death (ICD 8 – ICD 10) were acquired from Statistics Finland. The questions covering the characteristics of work and work organization have been reported elsewhere (Tuomi and Huuhtanen, 2001). The variables included here were responsibility for others (mean = 7.4, range 0–10), system of working hours (mean = 3.3, range 0–10), and the opportunity to develop and influence things at work (mean = 7.3, range 0–10). 3 RESULTS The survival curves for men and women are presented in Figures 1 and 2. Slight differences between the work ability index groups began to emerge roughly after 25 years’work career. Especially among men, the survival of poor work ability group decreased to about 0.5 indicating that about half of these men had died by the end of the follow-up. In the excellent work ability group, on the other hand, about 80% were still alive at the end of the study. Among women, no great differences were found based on the classification of work ability (Figure 2). When all the other independent factors were included in the analysis, level of Work Ability Index still remained an important predictor of mortality (Table 1). Besides good work ability, frequent physical exercise and work that involved responsibility for other people reduced death risk among men and women. Smoking caused a two-fold death risk for both genders. Among women, being married or living with a partner and the use of alcohol (compared to those who did not use alcohol at all) also significantly reduced mortality.

4 DISCUSSION Our study suggests that work ability, measured by Work Ability Index is a strong predictor of mortality. Especially among men, excellent WAI during occupationally active years was associated 102

1,1 1 0,9 0,8 0,7 0,6 0,5 0,4 0

11 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 Years in the same occupation Poor

Figure 1.

Moderate

Good

Excellent

Men’s Kaplan-Mayer survivorship functions for the four levels of work ability index (excellent, good, moderate, poor).

1,1 1 0,9 0,8 0,7 0,6 0,5 0,4 0

12

15

17

19

22

24

26

28

30

32

34

36

38

40

42

44

46

48

Years in the same occupation Poor

Moderate

Good

Excellent

Figure 2. Women’s Kaplan-Mayer survivorship functions for the four levels of work ability index (excellent, good, moderate, poor).

with over 50% higher survivorship compared to poor work ability. Good work ability reduced death risk even after adjustments for living habits and socio-economic factors. This result confirms the findings from a previous study of our study group (Tuomi et al., 1997b), but more studies are needed to clarify the relations between WAI and mortality. Nevertheless it seems 103

Table 1.

Cox proportional hazard ratios for 25-year all cause mortality among men and women. Hazard Ratio

Factor Socio-economic group Blue-collar workers Lower-level white-collar workers Upper-level white-collar workers Marital status Unmarried Married/co-habitating Divorced/separated Widow/widower Smoking Alcohol consumption Physical exercise Responsibility for other people at work Work Ability Index

Men

Women

1.05 1.06 1

1.20 1.05 1

1.04 0.79 0.79 1 2.26 *** 0.94 0.74 *** 0.95 ** 0.97 ***

1.27 0.78 *** 0.91 1 1.97 *** 0.75 *** 0.80 * 0.95 ** 0.99 a

***: p < 0.001, **: p < 0.01; *: p < 0.05; a: p < 0.10

evident that promotion of work ability has far-reaching effects that continue into retirement years. REFERENCES Ilmarinen J. and Tuomi K., (2004). Past, present and future of work ability. People and Work Research Reports, 65: pp. 1–25 Tuomi K. and Huuhtanen P., (2001). Promotion of work ability, the quality of work and retirement. Occupational Medicine, 51: pp. 318–324 Tuomi K., Ilmarinen J., Klockars M., Nygård C-H., Seitsamo J., Huuhtanen P., Martikainen R. and Aalto L., (1997a). Finnish research project on aging workers in 1981–1992. Scandinavian Journal of Work, Environment and Health, 23 suppl 1: pp. 7–11 Tuomi K., Ilmarinen J., Seitsamo .J, Huuhtanen P., Martikainen R., Nygård C-H., and Klockars M., (1997b). Summary of the Finnish research project (1981–1992) to promote the health and work ability of aging workers. Scan. J. of Work, Environment and Health, 23 suppl. 1: pp. 66–71.

104

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Impacts from occupational risk factors on self reported reduced work ability among Danish wage earners Lea Sell, Anne Faber & Karen Søgaard National Research Centre for the Working Environment, Copenhagen, Denmark

ABSTRACT: The aim of this paper is to identify work environmental risk factors for developing reduced work ability. The population studied is a representative cohort of Danish wage earners during the period 1990 to 2000 (Burr et al., 2003). The study is based on questionnaires and all information on the physical job demands, the psychosocial work environment, lifestyle, age, gender, education as well as work ability are self reported, and validated, measures. The design of the study is to use information on work environment, lifestyle as well as demographic factors in 1990 to predict the outcome reduced work ability in 2000. Logistic regression is used for the analyses. The analyses showed that high physical job demands was the most dominant risk factor for developing reduced work ability. Working with the body twisted was a significant risk factor for all groups of manual workers, care workers included. Furthermore, physical job demands of working with the hands lifted and working standing or squatting, as well as being exposed to high noise, were significant risk factors for blue collar workers and non skilled workers. For non skilled workers also lifting burdens above 20 kg and being exposed to teasing or threats of violence were highly significant risk factors for the development of reduced work ability. White collar workers had the lowest level of reported reductions in work ability over the ten year period and for this job group the largest risk factors seemed to be smoking and high body mass index. For care workers, also being exposed to teasing or threats of violence at the workplace seemed of major significance for the development of reduced work ability.

1 INTRODUCTION Reduced work ability is an important topic because it often links to early retirement from the labour market (Hopsu et al., 2005); (Liira et al., 2000) as well as it can be seen as a sign of decreased capabilities and life quality in the older age. While still at the labour market, the workers with reduced work ability have a significantly increased risk of longer periods of sickness absence (Reiso et al., 2001; Nygard et al., 2005). Thus, costs are high for the society, and for the individuals with reduced work ability. A high work ability of employees is also a requisite for good company performance (Tuomi et al., 2005) and for vital branches like the health care sector, retaining the ability to work is becoming of increasing current interest (Hasselhorn et al., 2007). Work ability is most frequently assessed using the Work Ability Index (WAI) (Tuomi et al., 1998) which apart from being basis for research also is used in health examinations and work place surveys performed by health professionals. This measure comprises three different dimensions; the demands of the job, the workers’ health status, and the workers’ resources. Included in the questionnaire on which the index is based is a question where the employees are asked to evaluate their own work ability. The question is a rating of current work ability compared to lifetime best, and has been applied as an individual indicator of work ability in a recent Finnish study, with good 105

results (Nygard et al., 2005). In the analyses in the present paper, work ability is assessed by using a single item measure based on a self-evaluation of to what degree the respondent considers own work ability as reduced due to disease, accident or toil. The study design is to use questionnaire data in 1990 on job exposures, life style as well as socioeconomic details to predict the outcome, reduced work ability 10 years after, as reported in 2000.

2 THEORETICAL BACKGROUND In earlier studies, work ability has been linked to age (Ilmarinen 2001), gender (Lin et al., 2006), mental well-being (Tuomi et al., 2005), stress and burnout (Hasselhorn et al., 2007). Some studies have used sickness absence as a way to measure work ability (Lindberg et al., 2006a; Lindberg et al., 2006b) and because of the evidence that work ability is significantly related to sickness absence (Reiso et al., 2001; Nygard et al., 2005) the majority of the factors selected for the analyses in this paper are also based on research in causes of sickness absence. A study from 2000 (Lund et al., 2005) identified a number of psychosocial factors in the work environment as risk factors for long term sickness absence. These included low decision authority, low skill discretion, high quantitative or emotional demands, job insecurity, low social support and low management quality, role conflicts, low predictability and being exposed to unpleasant teasing, unwanted sexual attention or threats of violence or violence at work. Unfortunately, the data on work environment exposures in 1990 which we use here, do not contain all of these scales of the psychosocial work environment. Only opportunity to learn new things and qualify on the job, level of support and encouragement from colleagues, the degree of predictability, measured by a question on to what degree the employee is receiving information on decisions concerning the work place, and finally if the employee is exposed to unpleasant teasing or unwanted sexual attention, are testable dimensions in this analysis. Uncertainty relating to decisions concerning the work place has evidenced to be an invisible stressor that raises the risk of heart disease (Iversen et al., 1989). Exposure to teasing, unwanted sexual attention or threats of violence provoke stress, anxiety and, in the long run, fatigue, in the victims (Hoegh, 2002). On the other hand opportunities to learn new things and qualify on the job as well as social support are documented preventive factors for job stress (Johnson and Hall, 1988). Also noise is included in the model as it has proven to have significant effect on the development of mental diseases. Noise is measured on a dichotomous scale, according to exposure to noise so high that one more than 75% of the working hours must raise one’s voice in order to be able to speak with others. For a review of the effects of noise on mental health see (Stansted et al., 2000). Physical job demands included in the present analyses are: Having to work with the back heavily bent forward with no support for hands or arms, working with twisted or bent body, working with hands lifted to shoulder height or higher or the neck heavily bent forward, and working squatting or kneeling (Lund et al., 2006). Also having a very passive job, as defined by sitting almost all the working time is included in the analysis of the white collar job group. Finally, life style factors, such as smoking, body mass index (BMI) and alcohol consumption are included (Christensen et al., 2007).

3 METHOD 3.1 Data and analytical strategy Data on work environment and health in the working population were obtained from the Danish Work Environment Cohort Study, DWECS (Burr et al., 2003). The cohort started out 106

with a random sample drawn in 1990 from the central population register, consisting of people aged 18–59 years per 1st October 1990. People in this panel were reinterviewed in 1995, 2000 as well as 2005. The 1990 sample consisted of 9,653 people living in Denmark, of which 8,664 participated (90%). Of these 6,067 were wage earners of which 4,670 also participated in the survey in 2000. The survey method was telephone interviews based on questionnaires sent out by mail beforehand. Participants in the 1990 sample are selected for the present analyses if they meet the following four criteria: 1) wage earner in 1990, 2) participating in the survey in both 1900 and 2000 (n = 4,670), 3) no report of already having an illness or after-effect from an accident and at the same time having been absent for more than 10 working days in 1990 (n = 246) and finally 4) they must not report to have changed type of job in 1995 (n = 827). After having omitted observations with missing values on any of the analysed items the balanced dataset consisted of 3,111 individuals. The question on work ability in 2000 was posed in the following way: “Is your work ability reduced due to diseases, accidents or toil”? The answers fall in four categories: “Yes, indeed” (6.0%), “yes, to some extent” (10.7%), “no, not much” (8.7%), or “no” (74.6%). Having answered in the first two categories is in the following classified as reports of reduced work ability (16.7%). To maximize the probability that the job exposures reported in 1990 have had a certain duration those reporting to have changed type of job in 1995 were not selected for the analyses. Moreover, as the exposures which are in focus in this paper are those which have an impact on actual loss of work ability, those who already have reduced work ability in 1990 are also excluded. Reporting to have an illness or an after-effect from an accident and at the same time having been absent more than 10 working days during the last 12 months is used as a proxy of reduced work ability in 1990. The job exposures in 1990, along with information on life style and demographic details, also from 1990, now form the basis which is used for predicting reduced work ability in 2000. Higher risk of long term sickness absence has also been investigated in relation to job group (Lund et al., 2007). High risk occupations for long term sickness absence were found to be working as kindergarten teacher, being employed in health care or day care, cleaning work, doing food preparation and working as unskilled workers. The present paper uses a division of the total working population in five major job groups: White collar workers, teachers and shop workers, blue collar workers, non skilled workers and care workers. White collar workers include academicals, technicians, bankers and office workers. Blue collar workers include both skilled tradesmen and industrial workers. Non skilled workers represent a variety of jobs, ranging from postal workers to agricultural workers. Care workers and teachers and shop workers form their own job groups. The division into job groups is made in order to minimize the effect from unknown organizational factors which vary among the job groups. The job groups are still large enough to ensure the necessary variation in the known job factors. Moreover, it should add to the total credibility of results that the evidenced risk factors are known to be valid for also sub samples of the general population. Table 1 summarizes the key variables for the 3,111 individuals in the cohort meeting all selection criteria.

Table 1.

Summary of key demographic and economic variables in balanced panel (N = 3,111).

N = 3,111

White collar workers N = 683

Teachers/ shopworkers N = 362

Blue collar workers N = 577

Non wkilled workers N = 1,156

Care workers N = 333

Male Female Reduced work ability in 2000 No vocational education Mean age in years (1990)

41.0% 59.0% 9.1% 7.9% 38.5

47.5% 52.5% 14.1% 9.9% 39.1

73.5% 26.5% 25.1% 35.7% 38.6

63.7% 36.3% 16.4% 22.8% 37.3

6.9% 93.1% 21.3% 29.1% 37.2

107

3.2 Statistical analyses Ordinary logistic regressions were used for estimating the outcome reduced work ability in 2000. Explanatory variables were work environment factors, lifestyle and demographic details reported in 1990. Initially, correlation analyses were performed on all explanatory variables. The correlation coefficients were below 0.30 for all explanatory variables, except for the physical job demands which turned out to be highly mutually correlated. As this raises the risk of multicollinearity (Gujarati, 1999) individual models were estimated for those exposures. The correlation matrix for physical job demands is shown in Appendix 1. The estimation method is maximum likelihood and statistical computer program is SAS 8.2, the logit procedure. Results are presented in odds, expressing the increase in the odds of reporting reduced work ability for a one point increase in the explanatory variable.

4 RESULTS Different analyses were carried out for each of the five job groups created. Table 2 shows the results of the analyses for white collar workers, teachers and shop workers and care workers. Table 3 shows the results for skilled blue collar workers and unskilled workers.

Table 2.

Odds for reporting reduced work ability in the three job groups. White collar workers N = 683

Teachers/shop workers N = 362

Care workers N = 333

N = 1,378

Odds

CI 95

Odds

CI 95

Odds

Female a Agee No vocational educationa BMIe 20+ cigarettesa Daily alcohola Developmentb Social supportc Predictability b Conflictsa High noise a Passive joba Twisting body d Bending body d Lifts 20+ kg d

1.208 1.044* 2.196*

.656–2.224 1.015–1.074 1.016–4.742

2.211* 1.086* 1.543

1.025–4.770 1.047–1.126 .567–4.201

1.167–1.305 1.011–1.015 1.449–1.596

1.102* 5.777* .938 1.231 1.096 1.151 1.333 1.264 1.164 – – –

1.005–1.209 1.032–32.330 .536–1.642 .897–1.688 .837–1.434 .801–1.652 .408–4.352 .355–4.502 .656–2.066 – – –

1.121* 2.676 .830 1.174 1.179 1.314 .354 1.257 1.137 – – –

1.009–1.246 .244–29.399 .418–1.649 .823–1.673 .864–1.609 .861–2.007 .075–1.676 .382–4.135 .420–3.079 – – –

1.051–1.056 <0.001–<0.001 1.378–1.418 1.138–1.162 1.126–1.152 1.264–1.356 2.232–2.251(*) .956–1.101 – 1.248* 1.105 1.102

a Dichotomous

CI 95 .360–4.238 .980–1.047 .772–2.984 .969–1.145 <.001–>999. .742–2.633 .825–1.602 .878–1.477 .889–1.938 .978–5.131 .420–2.453 – 1.057–1.474 .885–1.379 .876–1.384

variable. rating scale with 4 categories. cVerbal rating scale with 5 categories. d Verbal rating scale with 6 categories. e Continuous variable. Significance levels: (*): 0.05 < P < 0.10, *: 0.0000 < P < 0.05. CI95: 95% Confidence Interval. NOTE: In cases of more models per job group, odds are given as the intervals between the lowest odds ratio which the models yielded and the highest odds ratio which the models yielded. The confidence intervals corresponding to the intervals likewise express the most extreme confidence limits. bVerbal

108

All lifestyle factors, demographic variables and psychosocial risk factors were tested for all five job groups. Physical risk factors of twisting or bending the body many times per hour, working with the back bent forward with no support for hands or arms, working with the hands lifted to shoulder height or higher, having to work squatting or kneeling, working standing and having to lift burdens above 20 kilos were used only in the estimation for skilled blue collar workers and for unskilled workers. An exception was for care workers where twisting the body, bending the body and frequency of lifting above 20 kg were tested, too. For white collar workers and teachers and shop workers the effect of having a passive job, defined as sitting almost all the working time, was tested as the only variable representing work posture. Table 2 shows that for white collar workers and teachers and shop workers no work environment factors seemed to be of significance for predicting reduced work ability. Lifestyle factors seemed to be of significance to white collar workers, especially smoking. For teachers and shop workers, being a woman more than doubles the risk of reporting reduced work ability ten years later. Two of the examined work environment factors turned out as significant risk factors for reduced work ability for care workers. These were being exposed to teasing or threats of violence at the workplace, in the table defined as conflicts, and working with the body twisted. Continuing to Table 3, the picture changes dramatically. For non skilled workers nearly all risk factors which are tested turn out significant. Only daily consumption of alcohol; being a woman;

Table 3.

Results for skilled blue collar workers and unskilled workers. Blue collar workers N = 577

Non skilled workers N = 1,156

N = 1,733

Odds

CI95

Odds

CI95

Femalea Agee No vocational educationa BMIe 20 + cigarettesa Daily alcohola Developmentb Social supportc Predictabilityb Exposure to conflictsa High noisea Twisting bodyd Bending bodyd Standingd Working squattingd Work with hands liftedd Lifting 20 + kgd

1.654–1.990* 1.049–1.051* .928–.909 1.078–1.083* 2.429–2.590* 1.155–1.193 1.073–1.091 .939–.954 1.041–1.071 .804–.899 1.697–1.995* 1.117* 1.047 1.153* 1.091 1.216* 1.084

.998–3.053 1.029–1.072 .576–1.555 1.015–1.150 1.148–5.493 .760–1.815 .874–1.339 .769–1.164 .841–1.323 .285–2.499 .993–3.369 1.006–1.241 .913–1.199 1.009–1.317 .912–1..305 1.033–1.431 .914–1.285

1.296–1.644 1.056–1.061* 1.885–2.126* 1.049–1.057* 2.149–2.390* 1.085–1.126 1.077–1.124 .890–.919 1.157–1.218 1.881–2.022* 2.197–2.620* 1.175* 1.197* 1.133* 1.421* 1.189* 1.563*

.893–2.402 1.039–1.079 1.270–3.351 .998–1.111 1.067–4.482 .761–1.600 .901–1.339 .753–1.086 .951–1.474 1.077–3.513 1.380–1.669 1.071–1.289 1.058–1.353 1.019–1.259 1.197–1.686 1.015–1.394 1.332–1.834

a Dichotom

variable. rating scale with 4 categories. cVerbal rating scale with 5 categories. d Verbal rating scale with 6 categories. e Continuous variable Significance levels: (*): 0.05 < P < 0.10, *: 0.0000 < P < 0.05. CI95: 95% Confidence Interval. NOTE: In cases of more models per job group, odds are given as the intervals between the lowest odds ratio which the models yielded and the highest odds ratio which the models yielded. The confidence intervals corresponding to the intervals likewise reflect the most extreme confidence limits. bVerbal

109

and the three psychosocial variables: opportunity to learn new things and qualify on the job, support and encouragement from colleagues, and the degree of predictability, do not turn out significant. For blue collar workers, as was the case for teachers and shop workers, being a woman raises the risk of reduced work ability significantly. Common risk factors for the two groups were: Being exposed to high noise, working with the body twisted, working standing and working with the hands lifted. However, a comparison of the actual sizes of the odds, indicate that the impacts of the risk factors are higher among the group of non skilled workers. The relative importance of risk factors is more or less clear: Starting out with age, for teachers and shop workers an additional year of age raises the risk of reduced work ability by 8.6% (corresponds to odds: 1.086), as compared to a maximum of 6.1% for all other job groups considered together. Also a high BMI seems to be of largest significance for this job group. For white collar workers the largest risk factor seems to be smoking which more than quadruple the risk of reduced work ability. White collar workers have the lowest reported level of reduced work ability in 2000, of only 9.1%, which corresponds well to the relatively few risk factors. Even though only two factors turned out to be significant work environmental risk factors for care workers, their relative importance seem larger for this group of employees than for other groups: Being exposed to conflicts raise the risk for care workers of reduced work ability by just about 125% and only by 102% for non skilled workers which was the only other job group for which this risk factor turned out significant. For work with the body twisted, each level of increased frequency of this work posture raises the risk of reduced work ability for care workers by 25%, where the impact for blue collar workers is only 12% and for non skilled workers only 17.5%. The overall most important risk factors seemed to be: 1) working with the body twisted, which seem to be of major significance for all three groups of manual workers, 2) working with the hands lifted for which the odds are large and significant for blue collar workers and non skilled worker, and finally 3) working standing or squatting, for which the odds are significant for both blue collar workers and non skilled workers but with relatively smaller sizes of the odds. For non skilled workers, lifting burdens above 20 kg seemed to be the most important risk factor when comparing the relative sizes of the odds of other factors. The group of non skilled workers seemed to be at risk of many different hazards in the work environment and also had a large prevalence of reduced work ability in 2000 (16.4%). It may be because this job group represents a very varied job structure and thus it is more difficult to establish work environment standards for these jobs. Also, small job groups may have more difficulty in getting heard and thus experiencing work environment improvements. Moreover, making a separate analysis, not controlling for education, yielded the results that the psychosocial work environment factors, opportunities to learn new things at the job and being informed about decisions that concern the work place, turned significant. The odds for developing reduced work ability when having the lowest possible level of opportunities for development were between 1.68 and 1.94 (0.004 < p < 0.025) and the odds for developing reduced work ability when having the lowest possible level of information were between 1.66 and 2.01 (0.024 < p < 0.11). Omitting education in the analyses of the other job groups did not change results.

5 DISCUSSION As the job group to a large extent decides which risk factors are present, separate analyses were made for distinct job groups which turned out to be of major importance. The main empirical results indicated that physical job factors constituted the most dominant risk factors for developing reduced work ability. Working with the body twisted was a significant risk factor for all groups of manual workers, care workers included. Physical job demands of working with the hands lifted and working standing or squatting, as well as being exposed to high noise, were significant risk factors for blue collar workers and non skilled workers. For non skilled workers also lifting burdens above 20 kg and being exposed to teasing or threats of violence were highly significant risk factors for the development of reduced work ability. White collar workers had the lowest level of reported 110

reductions in work ability over the ten year period and for this job group the largest risk factors seemed to be smoking and high body mass index. The impact of having a very passive job was tested for this job group but did not turn out to be significant. A very passive job however, may interact with lifestyle in terms of being overweight. Possible interactions between having a very passive job and being overweight were not examined. For care workers also being exposed to teasing or threats of violence at the workplace seemed of major significance for the development of reduced work ability. Care workers were the only job group for which no lifestyle factors and no demographic factors were risk factors for reduced work ability. What is especially noteworthy is that for this job group, work ability apparently does not decrease with rising age. Other studies support this result (Cloutier, David, Prevost and Teiger, 1999) and these studies suggest that with experience, care workers develop work strategies which are both time saving and protect against back injuries. However, this group has other types of risk factors than other job groups, relating to the notion of having emotional work, and these are not included in the present analyses (Zapf, 2002). The application of a single item, self reported work ability measure yielded results which to a large degree supported results from earlier studies (Lund et al., 2006). However, this work ability measure may have a bias towards measuring primary physical work ability because the respondents are asked whether their work ability is reduced due to disease, accident or toil which does not apply directly to reduced work ability due to, for example, high mental demands.

6 CONCLUSION The present analyses showed that a single-item, self-evaluated work ability measure is applicable in an analysis of possible causes of reduced work ability. Especially physical job factors turned out to be dominant risk factors for developing reduced work ability. Working with the body twisted was a significant risk factor for all groups of manual workers, care workers included. Working with the hands lifted and working standing or squatting, as well as being exposed to high noise, were significant risk factors for blue collar workers and non skilled workers. For non skilled workers also lifting burdens above 20 kg and being exposed to teasing or threats of violence were highly significant risk factors for the development of reduced work ability. White collar workers had the lowest level of reported reductions in work ability over the ten year period and for this job group the largest risk factors seemed to be smoking and high body mass index. For care workers, also being exposed to teasing or threats of violence at the workplace seemed of major significance for the development of reduced work ability. Not all job factors were tested in the present analyses and especially for dimensions of the psychosocial work environment information was not attainable. The results of the present analyses point towards that, when accounting for a limited range of factors in the work environment, the workers at highest risk for developing reduced work ability are workers with physically strenuous work.

Appendix 1.

Twisting Squatting Sitting Standing Hands lift Bending

Correlation analysis for physical job factors. Squatting

Sitting

Standing

Arms lift

Bending

Lift 20 kg

0.25520 – – – – –

−0.23440 −0.37990 – – – –

0.18913 0.27286 −0.66971 – – –

0.21931 0.34576 −0.29347 0.26646 – –

0.35250 0.37431 −0.31098 0.25919 0.31557 –

0.19512 0.34170 −0.34809 0.27794 0.27784 0.32559

111

REFERENCES Burr, H, Bjorner, JB, Kristensen, TS, Tuchsen, F, and Bach, E. Trends in the Danish work environment in 1990-2000 and their associations with labor-force changes. Scand J Work Environ Health 2003; 29: pp. 270–279 Christensen, KB, Lund, T, Labriola, M, Bültmann, U, and Villadsen, E. The Impact of Health Behaviour on Long Term Sickness Absence: Results from DWECS/DREAM. Industrial Health 45: pp. 348–351, 2007 Gujarati, DN. Essentials of Econometrics. McGraw-Hill. 1999 Hasselhorn, H-M, Tackenberg, P, and Müller, BH. Working Conditions and intent to leave the profession among nursing staff in Europe. 7:2003. 2007. University of Wuppertal Hoegh, A. Aggression at work. 1-1-2002. National Institute of Occupational Health. 2005. Thesis/Dissertation Hopsu, L, Leppanen, A, Ranta, R, and Louhevaara, V. Perceived work ability and individual characteristics as predictors for early exit from working life in professional cleaners. International Congress Series 2005; 1280: pp. 84–88 Ilmarinen, JE. Aging workers. Occup Environ Med 2001; 58: pp. 546–552 Iversen, L, Sabroe, S, and Damsgaard, MT. Hospital admissions before and after shipyard closure. 299 28, pp. 1073–1076, 1989 Johnson, JV, and Hall, EM. Job strain, work place social support, and cardiovascular disease: a cross-sectional study of a random sample of the Swedish working population. Am J Public Health 1988; 78: pp. 1336–1342 Liira, J, Matikainen, E, Leino-Arjas, P, Malmivaara, A, Mutanen, P, Rytkonen, H, and Juntunen, J. Work ability of middle-aged Finnish construction workers – a follow-up study in 1991–1995. International Journal of Industrial Ergonomics 2000; 25: pp. 477–481 Lin, S, Wang, Z, and Wang, M. Work ability of workers in western China: reference data. Occup Med (Lond) 2006; 56: pp. 89–93 Lindberg, P, Josephson, M, Alfredsson, L, and Vingard, E. Promoting excellent work ability and preventing poor work ability: the same determinants? Results from the Swedish HAKuL study 2. Occupational and Environmental Medicine 2006a; 63: pp. 113–120 Lindberg, P, Vingard, E, Josephson, M, Alfredsson, L. Retaining the ability to work – associated factors at work.1. European Journal of Public Health 2006b; 16: pp. 470–475 Lund, T, Labriola, M, Christensen, KB, Bultmann, U, and Villadsen, E. Physical work environment risk factors for long term sickness absence: prospective findings among a cohort of 5,357 employees in Denmark. BMJ 2006; 332: pp. 449–452 Lund, T, Labriola, M, Christensen, KB, Bultmann, U, Villadsen, E, and Burr, H. Psychosocial work environment Exposures as risk factors for long term sickness absence among Danish employees: Results from DWECS/DREAM. JOEM 2005; 47: pp. 449–452 Lund, T, Labriola, M, and Villadsen, E. Who is at risk for long-term sickness absence? A prospective cohort study of Danish employees. Work 2007; 28: pp. 225–230 Nygard, CH, Arola, H, Siukola, A, Savinainen, M, Luukkaala, T, Taskinen, H, and Virtanen, P. Perceived work ability and certified sickness absence among workers in a food industry. International Congress Series 2005; 1280: pp. 296–300 Reiso, H, Nygard, JF, Brage, S, Gulbrandsen, P, and Tellnes, G. Work ability and duration of certified sickness absence. Scand J Public Health 2001; 29: pp. 218–225. Stansted, SA, Haines, MM, Burr, M, Berry, B, and Lercher, P. A review of Environmental Noise and Mental Health. Noise and Health 2000; 8: pp. 1–8 Tuomi, K, Ilmarinen, JE, Jahkola, A, Katajarinne, L, and Tulkki, A. Work Ability Index, 2. Ed 1998 Tuomi, K, Vanhala, S, Janhonen, M, and Nykyri, E. Employees’ work ability and company performance: A follow-up study in the metal industry and in retail trade. International Congress Series 2005; 1280: pp. 234–237 Zapf, D. Emotion work and psychological well-being: A review of the literature and some conceptual considerations. Human Resource Management Review. (12): pp. 237–268, 2002

112

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Work ability and work quality as indicators for a longer and more productive working life Michele Meloni & Pierluigi Cocco Department of Public Health, Occupational Health Section, University of Cagliari, Italy

Giovanni Costa Department of Occupational Health, University of Milan, Milan, Italy

ABSTRACT: Objective – In order to cope with the new challenges posed by globalization, job tasks and mental and physical workloads have to be re-designed to fit with levels of productivity in an ageing workforce. Methods – We tested the work ability index (WAI) and the work quality index (WQI) in a group of 14 workers in a food manufacturing plant, before and after a kaizen intervention to promote total productive maintenance. Results – The average WAI was 43, and it was not affected by age (r = 0.086, p NS), nor did it vary by occupation, whether in supervisors or in production line workers. However, among supervisors, WAI showed a modest inverse correlation with age (r = −0.239, p NS), while it was higher among production line workers (r = 0.596, p NS). Work quality (WQI) also improved, due to a 10% reduction in the average heart rate; an 84% reduction product loss in the boxing area and a 3% increase in total production. Conclusion – Although inference is limited by the small population study size, our experience suggests that kaizen intervention positively affects WAI and WQI, thus representing a good tool to promote and maintain health and safety in an ageing working population. 1 INTRODUCTION In western countries, the progressive ageing of the working population is expected to lead to an impressively rapid change in the daily physical and mental workload within a few decades (Rantanen, 2003). Such a change challenges the ability of the Occupational Health and Ergonomics disciplines with restyling job design and work organization to ensure good psychophysical health for longer in ageing workers and to improve productivity. The keywords in the global economy, such as innovation and creativity, in the continuous challenge towards technology improvement and higher quality products, call for more skilled workers while, at the same time, dealing with an elderly working population. To cope with these new challenges, both job tasks and mental and physical workloads have to be redesigned to fill the missing link between technology and ageing. However, thus far, the new production technologies and the Japanese interpretation of lean production such as Total Productive Maintenance (TPM) have focussed on the hardware (machine shops), more than the soft ware (human creativity, health, workers). In this perspective, it is our experience that full involvement of the occupational physician in designing job tasks is of paramount importance: work ability index (WAI) and work quality index (WQI) are two helpful indicators in this regard. WAI, as first introduced by Illmarinen et al. (1992), is a tool indicating the level of a worker’s health compliance in a given workplace. Each of seven items addressing health in the work place 113

is scored and the responses grouped into four different categories (poor, modest, good, excellent). Using WAI to evaluate job design and to redesign job tasks results in greater workforce awareness with regard to their health status, as well as their ability to cope with work demands. Such strategy is therefore effective in promoting health in the working environment. On the other hand, WQI summarizes two parameters: one indicating work productivity by shift and another measuring workload via a simple biological indicator such as heart rate, in order to provide an estimate of energy expenditure on a certain task. Therefore, WQI gives a measure of energy efficiency in a given workplace, designed to promote a progressive improvement of working life by defining and reducing discomfort and fatigue, as well as identifying product waste and defects, and by promoting organizational and human efficiency. WQI definition includes seven progressive steps to tackling work discomfort: 1) definition of fatigue rate; 2) measurement of work load; 3) identification of the target; 4) development of countermeasures and estimation of fatigue rate and time needed for operations; 5) implementation of countermeasures; 6) reassessment and evaluation of workload fatigue rate and time for operations; 7) assessment of effects in terms of safety, health, and productivity. The Japanese term defining the TPM organizational approach is ‘Kaizen’, literally meaning “change for the better”, but more loosely translated as “continuous improvement” (Ohno T, 1988). The Kaizen approach implies that all the small, incremental changes deemed necessary should be applied continuously and sustained by strong commitment, thus resulting in relevant improvement. ‘To plan, to do, to check, and to act’ have been identified as the Kaizen steps to achieve efficiency, comfort, well being, as well as motivation, and creativity. In this study, we used WAI and WQI to assess the results of a work re-organization programme using kaizen intervention to promote total productive maintenance (TPM) in a food manufacturing plant, mainly based on workers involvement. 2 METHODS The kaizen intervention programme was carried out in a food manufacturing plant southern Sardinia (Italy) by the local occupational health physician in a subgroup of 14 male workers (10% of the total male workforce) randomly selected among those employed in 10 different production jobs. Workload, bad working postures, and environmental factors, such as noise, heat, cold, and light, were considered as occupational stress factors on the shop floor. Improvement was achieved thanks to a new work design eliminating awkward postures and long term standing postures on the production line. In accordance with the Japanese philosophy of work organization, the scope was to avoid “muda” (waste) of muscular energy, “muda” attributable to standing for long periods of time in the upright position, “muri” (fatigue) human fatigue, “mura” (irregularity) of logical sequences in work processes on the “gemba” (shop floor). By analysing the workload in a given work position, we elaborated an algorithm to assess the added value (ε) to production and safety, health maintenance and well being on the shop floor. ε=

nFPP − WP /HR nW

where: ε = production evaluation; nFPP = number of final pieces/products; WP = waste products; nW = number of workers involved; HR = heart rate ε = added value, calculated by the difference of the ratio of production performance indexes to biological parameters before and after Kaizen improvement intervention. Heart rate at rest and blood pressure were assessed by physical examination during periodical health surveillance and at the production line while performing the usual job task, using an ambulatory 114

blood pressure monitoring (ABPM) device (Pickering et al., 1994). Measurements were taken before initiating the kaizen intervention programme and again a few (2–6) months later. In the same circumstances, workers completed the WAI questionnaire scoring their health status in relation to their job, and their ability to cope with work demand. Information on total number of products in each production job was provided by the factory management for each work shift. Being roughly normally distributed, both WAI and WQI were treated as parametric variables to calculate summary statistics (arithmetic mean and standard deviation) as well as their linear regression with age in the overall study population and by work position, whether production supervisor or food processor. The difference between slopes in supervisors and food processors was tested with covariance analysis.

3 RESULTS The average age in the total study population was 46.1 (sd = 7.48), however, it was greater for supervisors (mean = 50.0, sd = 5.45) than workers in the production line (mean 42.1, sd 7.45) (p < 0.01). Overall, the average WAI was 43 (sd = 4.1), the upper cut point between the good and excellent range of score values (Illmarinen et. al., 1992), and it was not affected by age (r = 0.086, p NS), nor did it vary by occupation, whether supervisors or production line workers (Figure 1a). a. Overall study population 60 WAI

50

40 y
0,047x  40,693 R2
0,0074

30

20 20

30

40 Age

50

60

b. Production line workers and supervisors 60

Analysis of covariance Comparison of slopes F
1.02; p NS Comparison of intercepts F
0.17; p NS

WAI

50 40 30

Supervisors Production workers

20 20

30

40 Age

50

60

Figure 1. Work ability index (WAI) by age among the total study population (a), and in production line workers and supervisors (b). Results of the covariance analysis are embedded on graph b.

115

However, among supervisors a modest inverse correlation between WAI and age was observed (r = −0.239, p NS), while slope was upward among production line workers (r = 0.596, p NS) (Figure 1b, c). Due to the small size of the study population, the analysis of covariance failed to provide statistical support to the difference between slopes (F = 1.02; p NS). Work quality (WQI) was also measured in 10 different job positions, based on total number of final products, total number of wasted products, number of workers, average heart rate associated with work tasks. After kaizen intervention, the average heart rate showed a 10% reduction, loss of products in the packaging area was reduced to 84%, and and an increase of 3% in total production. Moreover, from 2002 an 18% reduction in the number of first-aid interventions and a reduction of 43% in absenteeism was observed.

4 DISCUSSION Although inference is limited by the small population study size, based on our herein described shopfloor experience, we suggest that kaizen intervention positively affects WAI and WQI, thus representing a good tool to promote and maintain health and safety in an ageing working population, particularly among workers on the production line. The ε calculated showed the impressive improvement in all the job positions, related to a sharp decrease in average HR and blood pressure after kaizen changes. Based on our results, WAI appears to be a good indicator of beneficial effects of the kaizen approach and particularly at shopfloor on the production line. Further work is required to assess in more detail the extent of such postitive effect. Work organization deals with the way work is managed, including job design, new technologies, automation, pace of production, work/rest schedules, shift work, working hours, and overtime, while job design examines work pace, skill and strain required, as well as the degree of the worker’s control over his job: substantive participation, e. g. “partcipatory ergonomics”. Teamwork and a participatory approach influence work organization. Growing consensus exists regarding workplace hazards which result from not pursuing the continuous improvement of workers’ education and knowledge. Workers whose tasks are strongly demanding, and require a heavy workload with low level opportunity of proposing alternative better coping options, have been shown to be at greater risk of suffering injuries and developing adverse health outcomes, such as stress, psychological strain, and cardiovascular diseases (Schnall, 2000). From the industrial management perspective, low quality products and a greater rate of waste frequently occur under such circumstances. Alternatively, we propose a long lasting, possibily more effective, strategy through educating workers to have their say in how their own task and safety can be improved, using their own competence and work experience. Our results using WAI and WQI show that such an approach not only improves workers’ health and safety, but also their perception and productivity. REFERENCES Illmarinen, J., Tuomi, K, and Klochars, M., (1997). Changes in the work ability of active employees over an 11 year period. Scandinavian Journal Work, Environment & Health, 23: Suppl 1: pp. 49–57 Ohno, T., (1988). Toyota Production System: Beyond Large-Scale Production. Portland, Or: Productivity Press Pickering, T.G., and James, G.D., (1994). ABPM blood pressure and prognosis. Journal of Hypertension, 12: pp. s29–s33 Rantanen, J., (2003). Perspectives of Occupational health in the Changing world of work. Giornale Italiano di Medicina del Lavoro e Ergonomia, 25: 3: pp. 269–270 Schnall, PL, Belkic, KL, Landsbergis, PA, Baker, D, EDS., (2000). The Workplace and Cardiovascular Disease. Occupational Medicine: State of the Art Reviews: 15 (1).

116

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Disabled people at work and work ability: A study in a Brazilian company∗ Eloisa Petruci Hodge Work and Health Research Group, Faculty of Medical Sciences, State University of Campinas – UNICAMP, Campinas, Brazil

Inês Monteiro Faculty of Medical Sciences, State University of Campinas – UNICAMP, Campinas, Brazil

ABSTRACT: The inclusion of disabled people in society is a current concern in many countries. The objective of this study was to evaluate the work ability of disabled workers in comparison to nondisabled workers. The sample for this cross-sectional study consisted of a total of 293 workers, from which 13 were disabled with either a physical, mental or hearing deficiency. A socio-demographic and life style questionnaire and the Work Ability Index were used. The subjects were young, well educated, and were engaged in physical, leisure and other activities outside of work. A lower WAI score was associated with being disabled and with the perception that the work caused fatigue and or stress. In line with this, the inclusion of disabled people in the work force must occur simultaneously with actions to improve overall work conditions. In this way, disabled workers may become effectively included and remain active until retirement age. Keywords:

disabled workers; Work Ability Index; social inclusion.

1 INTRODUCTION The inclusion of disabled people in society is a current concern and they are especially vulnerable to poverty and social exclusion according to the European Commission (2005). The Ethos Institute of Social Responsibility (2001) reported that about 610 million of people around the world have some disabling condition, with 386 million of those being part of the economically active population. In the European Union (2001) 14.5% of the population within the active age bracket reported some kind of disabling condition. For the 10 newest member countries, this percentage increases to 25%. It is estimated that the majority of the disabled people, around 80%, live in developing countries according to data from the European Union (2001). The Brazilian Institute of Statistics and Geography reported that, in 2002, there were 24.5 million people in Brazil presenting some disabling condition (Instituto, 2005). The aim of this study was to evaluate the work ability of disabled workers in comparison to a group of non-disabled workers.

∗ This

research is part of the Master Dissertation of Eloísa Petruci Hodge (20/02/2006) – “Riscos associados ao trabalho e capacidade para o trabalho entre trabalhadores de uma indústria farmacêutica” [Work associated risks and work ability among pharmaceutical industry workers] and received grant from CAPES and CNPq.

117

2 METHODS A cross-sectional descriptive study was carried out among workers of a private Brazilian pharmaceutical industry. The population under study was composed of 630 workers, from which 25 presented some kind of physical, mental or hearing disabling condition as defined by the Brazilian legislation, confirmed by a medical report. A convenience sample was used as described by Simon (2006). It consisted of 293 workers, from which 13 were disabled, randomly chosen by month of birth. The workers performed activities with physical, mixed, and mostly mental demands, as reported by Ilmarinen (2001). Most of the disabled workers (69.2%) performed their functions in the production area, and executed the same activities as the non-disabled workers. A questionnaire with socio-demographic and life style questions, elaborated by Monteiro (1997) and the Work Ability Index (WAI) developed by Tuomi et al. (1997) were used to collect the data. The results were statistically analyzed and a significance level of 5% (p = 0.05) was used as described by Conover (1971) and Hosmer and Lemeshow (1989). Regarding the ethical aspects of the study, all data was obtained with the agreement of the company and the workers, and was approved by the Ethics Committee.

3 RESULTS The majority of the disabled workers was male (61.5%), single (69.2%) and without children (84.6%), and the average age was 28.8 years (range 18–47). Coincidently, 84.6% had at least 11 years of formal education, 53.9% were still studying at the university, and 69.2% regularly carried out housekeeping tasks at home (average of 1.7 hours/day). None of the disabled workers reported tobacco use, 30.8% reported alcohol intake, 76.9% was regularly engaged in physical activities in agreement with levels recommended by the World Health Organization; and 69.2% had an adequate Body Mass Index (BMI) according to the World Health Organization. In the non-disabled group, 50% of the sample was male (n = 140), and the average age was 28.6 years (range 18–62). The majority was single (51.4%), without children (66.1%); 93.2% had at least 11 years of formal education; 33.2% still studied (technical, under graduate and graduate courses) and 61.4% regularly do housekeeping tasks at home (average of 2.1 hours/day). Tobacco use was reported by 8.2% (average of 7.1 cigarettes/day); 43.2% reported some alcohol intake. The majority (70.7%) was engaged in regular physical activities, in agreement with levels proposed by the World Health Organization (2006); had a normal BMI (73.9%), and all reported engaging in leisure activities. The distribution of the workers in different WAI categories is presented in Table 1. In the disabled group the occurrence of the excellent WAI score is significantly lower than in the non-disabled group. When questioned about their health compared to other people of the same age, 38.5% of the disabled workers considered themselves to be much better, 30.8% better, and 30.7% the same. In the non-disabled group, 18.2% considered themselves to be much better, 26.8% better, 52.1% the same, and 2.9% worse; 62% reported that the work caused fatigue and/or stress. The non-disabled workers had a 6.3 times greater chance of having an excellent WAI score than the disabled ones. Those who did not indicate that the work caused fatigue and/or stress presented a 1.9 times greater chance of an excellent WAI than the ones who did indicate it. From the multivariate logistic regression analysis the profile of the workers with an excellent WAI is composed of non-disabled workers and who do not feel fatigue and/or stress from work. The non-disabled had a 5.9 times greater chance of an excellent WAI and those who did not report fatigue and/or stress had a 1.8 times greater chance of excellent WAI than the opposite group (Table 2). 118

Table 1.

Distribution of the workers of a pharmaceutical industry with respect to WAI categories.* WAI poor/moderate and good#

WAI excellent

Disabled worker

N

%

N

%

Total

No Yes

129 11

46.1 84.6

151 2

53.9 15.4

280 13

Total

140

153

293

#

categories grouped due to the size of the sample. Chi-squared test: p = 0.007. * WAI cut-off point for workers under 30 years of age as proposed by Kujala et al. (2005).

Table 2.

Univariate logistic regression analysis for WAI categories of workers from a pharmaceutical industry.

Variable

Comparison levels*

p-value

OR**

CI 95%

Sex Health compared to others Disabled Current use of medication Work causes fatigue and/or stress

Female/Male 1–2/3–4 No/Yes No/Yes No/Yes

0.270 0.724 0.017 0.199 0.010

1.30 1.09 0.16# 0.73 0.52$

0.82 − 2.05 0.69 − 1.72 0.03 − 0.71 0.45 − 1.18 0.31 − 0.85

*Reference level/comparison level; WAI excellent (n = 153); WAI good/moderate (n = 140). **OR = Odd Ratio of excellent WAI; CI 95% = 95% Confidence Interval for the OR. # considering “Disabled = yes” as reference: OR = 6.44 (CI 95%: 1.40; 29.58). $ “Work causes fatigue and/or stress = yes” as reference: OR = 1.94 (CI 95%: 1.17; 3.20).

4 DISCUSSION AND CONCLUSION The results from this study show this is a young population, with good education, and that the majority is engaged in physical and leisure activities and in some sort of additional activity outside work – study or housekeeping tasks. Those additional activities (study or housekeeping tasks) can sometimes intensify the physical/mental weariness and increase the chance of developing diseases, especially mental and musculoskeletal diseases, as related by Seligman-Silva (1994). The fact that the disabled workers reported a better self-evaluation of their health condition is probably associated with the fact that they are inserted in the work environment instead of unemployed. However, the presence of a disabling condition is strongly associated with lower WAI score: when we combine this data with the fact that most of the disabled workers work in the production areas and reported that “the work caused fatigue and/or stress”, we also have to consider the fatigue and environment stress agents as probable causes of the decrease in WAI. Several studies have reported that workers exposed to fatigue generating factors present a reduction in the WAI, as related by Ilmarinen et al. (1991) and Tuomi et al. (1999). However, those studies were conducted among workers in the 45 to 58 year age range, what can be seen as a limiting factor to the results obtained. The evaluation of the disabled workers has shown an important statistical relationship with early functional ageing, first because those workers have at least one assumed base disease treated by a physician, and also because they have to be regularly absent from work to continue their medical 119

treatment, both factors that directly reduce the WAI score, as reported by Duran and Monteiro-Cocco (2004). In conclusion, the inclusion of the disabled population in the work environment must occur in a responsible manner. The inclusion policies must be accompanied by actions to improve work conditions and workplace health promotion. In this way, disabled workers may become more effectively included and remain active until retirement age. REFERENCES Conover, W. J., (1971). Practical Nonparametric Statistics. (New York: John Wiley & Sons). Duran, E. C. M, and Monteiro-Cocco, M. I., (2004). Capacidade para o trabalho entre trabalhadores de enfermagem do pronto socorro de um hospital universitário. Revista Latino-americana de Enfermagem. 1: pp. 43–49 [English abstract]. European Commission, (2004). Report on social inclusion. An analysis of the national Action Plans on social inclusion (2004–2005) submitted by the 10 new member states. (Luxembourg: European Commission). European Union. Oficial Gateway, (2005). [Accessed in 2005 Oct 11]. Available on: http://europa.eu.int/scadplus/ leg/en/s02311.htm. Hosmer, D. W., and Lemeshow, S. L., (1989). Applied Logistic Regression. (New York: John Wiley & Sons). Ilmarinen, J., (2001). Aging Workers. Occupational and Environmental Medicine, 8: pp. 546–52. Ilmarinen, J., Tuomi, K., and Eskelinen, L. et al., (1991). Summary and recommendations of a project involving cross-sectional and follow-up studies on the aging worker in Finnish municipal occupations (1981-1985). Scandinavian Journal of Work Environment and Health, 17(1 suppl): pp. 135–141. Instituto Brasileiro de Geografia e Estatística (IBGE), (2005). Censo Demográfico 2000. [Demographic Census] [Acessed in 2005 Fev 15]. Available on: http://ibge.gov.br/home/estatistica/populacao/censo2000/ tabulacao_avancada/tabela_ gr_uf_2.1.3.shtm. Instituto Ethos de Responsabilidade Social, (2002). Inclusão das pessoas com deficiência. [Disabled people inclusion]. (São Paulo: Instituto Ethos). Kujala, V., Remes, J., Ek, E., Tammelin, T., and Laitinen, J., (2005). Classification of Work Ability Index among young employees. Occupational Medicine, 55: pp. 399–401. Monteiro-Cocco, M. I., (2002). Capacidade para o trabalho entre trabalhadores de uma empresa de tecnologia da informação. [Work ability among information technology workers]. [free docence thesis]. Campinas (SP): Universidade Estadual de Campinas. Pan-American World Health, (2003). Doenças crônico-degenerativas e obesidade: estratégia mundial sobre alimentação saudável, atividade física e saúde. Brasília: OPAS; 2003. Seligmann-Silva, E., (1994). Desgaste mental no trabalho dominado. (Rio de Janeiro: Ed. UFRJ, São Paulo: Cortez). Simon S., (2006). Definitions of important terms. [Accessed in 2006 Feb 10]. USA. Available on: http://www.cmh.edu/stats/definitions/convenience.htm. Tuomi, K., Ilmarinen, J., and Jahakola, A. et al., (1997). Índice de capacidade para o trabalho. [Work Ability Index]. (Helsinki: Finnish Institute of Occupational Health). Tuomi, K., Ilmarinen, J., and Seitsamo, J. et al., (1999). Work, life-style, health and work ability among ageing municipal workers in 1981–1992. In: FinnAge. Action Programme to promote health, work ability and well-being of aging workers in 1990–96, edited by Ilmarinen, J., Louhevaara, V. (Helsinki: Finnish Institute of Occupational Health), pp. 220–233. World Health Organization, (2006). Move for health. Benefits of physical activity. [Accessed in 2006 Feb 12]. Available on: http://www.who.int/moveforhealth/advocacy/information_sheets/benefits/en/index. html.

120

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Work ability of a population of 40+ in Luxembourg Nicole Majery Service de Santé au Travail Multisectoriel, Luxembourg

ABSTRACT: Since 2005, the «Service de Santé au Travail Multisectoriel» (STM) in Luxembourg is collecting data about the work ability of workers of 40 years and older. This article summarizes data collected from 6,059 questionnaires from workers occupying a variety of jobs. We found the commonly reported decreasing trend of the work ability index until 40–55, followed by a later increase for the workers over 60 years old.

1 INTRODUCTION In 2005, with a financial support from the European Social Found a project was started to study work ability of the workers older than 40 years. In Luxembourg, the “Serveice de Santé au Travail Multisectoriel” (STM) is an occupational health service which has in charge 28,000 companies, representing 180,000 workers with different kind of works. The main activity sectors are transportation, construction activities, commercial activities, cleaning companies and offices.

2 METHOD Every worker older than 40 years coming for a physical examination was invited to answer the work ability index (WAI) questionnaire. It was used in different languages spoken in Luxembourg: German, French, English and Portuguese.

3 RESULT 6,059 questionnaires were completed. 71% of the participants were men. In Luxembourg early retirement is very common. Therefore only 2% of the participants have more than 60 years.

Table 1.

Percentage of the participants for different age groups.

40–44 years 45–49 years 50–54 years 55–59 years >60 years

40% 31% 19% 8% 2%

121

42 41 40 39 38 37 36 35 34 40–44

45–49

50–54

55–59

60

55–59

60

WAI

Figure 1. Average WAI results by age group (n = 6,059).

9 8.5 8 7.5 7 6.5 6 5.5 5 4.5 4 40–44

45–49

50–54

Current work ability compared with the lifetime best

Figure 2.

Own prognosis of work ability two years from now

Current WAI compared with lifetime best and own prognosis in two years from now.

122

45

43

41

39

37

35

33

31 40–44

45–49

50–54

55–59

60

Physical work Mental work Mental  physical work

Figure 3. Work ability in relation to the job demands.

Table 2. WAI for different sectors mainly represented among the participants. Kind of job

40–44 years

45–49 years

50–54 years

55–59 years

>60 years

Truck drivers

40.22 (b: 5.00) 312

40.83 (b: 6.61) 255

39.38 (b: 7.33) 158

39.82 (b: 7.34) 76

40.67 (b: 7.02) 14

39.60 (b: 7.38) 133

37.77 (b: 6.66) 132

34.15 (b: 9.78) 66

30.85 (b: 9.19) 47

31.81 (b: 13.07) 8

40.32 (b: 6.60) 117

38.58 (b: 7.16) 90

37.69 (b: 7.90) 48

33.76 (b: 8.43) 21

31.50

43.34 (b: 4.73) 302

41.31 (b: 7.16) 179

41.57 (b: 6.41) 659

39.77 (b: 7.55) 109

41.95 (b: 4.48) 10

Number Cleaning workers Number Construction workers Number Office workers Number

1

22% considered their work as physically demanding, 19% as mentally demanding and 59% as well physically and mentally demanding. For each age group the WAI is calculated. It is decreasing from 40 to 55 and than increasing again. The current work ability compared with the lifetime best has the same evolution as well as the prognosis of work ability two years from now. Physically demanding workers have the lowest work ability. 123

41.5

41

40.5

40

39.5

39

38.5 20

20–24

25–29

30

WAI

Figure 4. Work Ability Index and BMI.

The increasing score after 55 years can be considered as a healthy worker effect. As the following table shows, this healthy worker effect is not found for work with high physical demands (construction workers and cleaning staff). WAI in relation of BMI: Employees with an overweight or obesity have a lower WAI.

4 CONCLUSION We observed that the WAI decreases from 40–55 years and than increases again. But for high physical demanding jobs, the WAI is decreasing also after 55 years. However the small number of participants over 60 years does not allow confirming these results. Employees with mental demanding activities have a better work ability. For physical demanding work, the WAI scores were the lowest. Workers with a high BMI have a lower WAI than those with normal weight. REFERENCES Ilmarinen J., (1999). Ageing workers in the European Union-Status and promotion of work ability, employability and employment. Finnish Institute of Occupational Health, Ministry of Social Affairs and Health, Ministry of Labour, Helsinki Ilmarinen J. and Ratanen J., (1999). Promotion of Work Ability during Ageing. In: American Journal of Industrial Medicine Supplement 1: pp. 21–23 Tuomi K, Huuhtanen P., Nykyri E. and Ilmarinen J., (2001). Promotion of Work Ability, the quality of work and retirement. In: Occp. Med Vol 51 No5, pp. 318–324

124

Chapter 3 Staying at Work

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Older worker career plateau: Issues and remedies Pat McCarthy Graduate School of Business, RMIT University, Melbourne, Australia

Sharon Moore School of Management, University of Western Sydney, Sydney, Australia

ABSTRACT: The objective of this research was to understand older worker experiences of employment, and to examine the impact of those experiences on their motivation and effectiveness in today’s organisations. This was achieved through a study of motivation, organisational and job content plateau. The research involved a diverse group of older workers at the national headquarters of a large and highly regarded Australian Government Business Enterprise. Roles ranged from administration to executive leadership across a broad range of functions. Empirical data derived from the application of case study and pattern matching methodologies which sought to identify gross matches and mismatches. The research concluded that although older workers possess the motivation and attributes necessary for making a valuable contribution to today’s organisations they are likely to lack technical skills currency. The research also found that an optimal input by older workers is very difficult for organisations to achieve and sustain. The study observed that organisational plateau impacts most, if not all people, but this situation is generally accepted by older workers provided they are treated fairly and with respect. The key area of focus of organisations and individuals ought to be on the prevention of job content plateau.

1 DEFINITIONS Career has been defined as the unfolding sequence of a person’s work experiences over time (Arthur, Hall & Lawrence, 1989, in Arthur et al., 2005). Career Plateau, ‘the point in a career where the likelihood of additional hierarchical promotion is very low’ (Ference, Stoner and Warren in Duffy, 2000). There are three broadly accepted career plateau types: Organisational or Structural Plateau; Job Content Plateau; and Personal or Life Plateau. 2 RESEARCH ORGANISATION The research organisation is an Australian Government Business Enterprise with revenue of more than AUS $4.2 billion. It operates in the letters, parcels and logistics markets; and financial services. The organisation has approximately 35,000 employees and has enjoyed AAA Standard & Poor’s rating for many years. It is regularly rated Australia’s most trusted corporation. 3 RESEARCH METHOD The principal methodology was case study and pattern matching. 40 people volunteered to be interviewed. In addition, participants were asked to bring to the interview a chart depicting their 127

work-life experiences, its ups and downs. Many participants chose to use the term ‘plateau’ to describe their current work circumstances as depicted in their charts. There were three commissioned supplementary studies. 47 volunteers aged over 40 years old from the Information Technology Division participated in case studies. 40 volunteers aged under 30 working in the organisation’s Call Centre were engaged in story telling activities describing their work experiences. In addition, 8,525 current employees completed questionnaires about their retirement intentions and a further 1,750 recent retirees completed questionnaires about their reasons for retirement.

4 WHAT OLDER WORKERS WANT Participants in this research expressed strong desires for empowerment whilst simultaneously seeking a paternalistic approach from their employer with regard to employment security and interesting job opportunities. Older workers seek personal growth and stimulation. They sought jobs that were designed to provide challenge and a level of ambiguity that provided freedom from micro management. They seek variety, broad job scope and they want control over how they undertake their work through the opportunity to make decisions. They like working in teams and value interaction with others. The older workers in the sample sought procedural fairness, a work environment where the politics are well-managed, where they have the freedom to speak freely where there is a lack of bureaucracy and a lack of age based discrimination and stereotyping. They want a good relationship with their supervisor, they like being appreciated for their ideas and they welcome the opportunity to use their knowledge. Sound communication practices including feedback and performance review were valued and older workers sought rewards for contribution. They appreciate being sponsored for challenging projects. Post retirement intentions were altruistic in most cases-giving back to the community was found to be the second most important job feature in any post retirement work. The most important job feature, and a significant post retirement intention, was the opportunity for personal development. These findings about older worker motivation support research conducted by Latham and Pinder (2004) who conducted a comprehensive examination of the progress made in motivational research over the ten years between 1993 and 2003. Drawing on Maslow’s seminal work from the 1950s, Latham and Pinder cited a fifteen-country study reported by Ronen (2001), in Erez and Kleinbeck (2001). This study found widespread continuing support for Maslow’s theory that people have innate biological needs for acceptance and approval, status, power, control of resources and predictability and order. Further support for Maslow’s theory has been found by Ajila (1997) and Kamalanabhan et al. (1999), who concluded that there remains wide acceptance of the practical significance of Maslow’s theory. An interesting observation by Kanfer and Ackerman, 2004, pp. 15 is that there is ‘neither theoretical nor empirical evidence to support the notion of an inevitable and universal decline in work motivation with age’. This proposition was valid for these author’s research which found that older workers, like their younger colleagues, are highly motivated to work. Nevertheless, this capacity to be motivated is seriously tested in the workplace. Older workers considered that they provide valuable experience based skills that are consolidated over time such as being a loyal and calming influence in the workplace, possessing knowledge of organisational history, judgement and political nous in decision making and conflict situations; the so called ‘soft skills’ sought for the future by employer organisations such as the Business Council of Australia (2002). However, the lack of attention to technical skills currency may be insufficient for older workers to remain employable in the future given adverse stereotypes held about older workers by the community in general (Taylor, Encel and Oka, 2002). This is because preferred methods of learning by older workers involved learning on-the-job, learning through challenging assignments and talking with others. Formal learning was less favoured as 45% rarely undertook professional development and 62% last undertook to formal study greater than 10 years 128

ago. This lack of attention to formal skills currency is likely to represent an insufficient effort by older workers to maintain up to date technical skills.

5 WHAT OLDER WORKERS GET: A PASSIVE JOB A significant disconnect was found between what motivated older workers and their workplace realities. Lemire, Saba and Gagnon (1999) refer ‘to the feeling of failure or frustration that individuals may experience following a temporary or permanent halt in the progression of their career’. Typical of the descriptions of work experiences by study participants were: ‘I’m blocked’, ‘I’ve been in the same job for 15 years’, ‘My job has narrowed and I’m closing in on myself ’, ‘My ageing and retirement is starting to come up in my performance reviews’, ‘I’m not getting challenging and exciting work’, ‘The organisation does not appreciate that older workers have a fair bit to offer’. More than 93% of the study participants felt they were impacted by Organisational or Structural Plateau, supporting the research of Bardwick (1986) who refers to ‘the rule of ninety-nine percent – 99% of all workers will eventually reach this type of Plateau because of the scarcity of jobs as one moves higher up the organisational hierarchy. On average female employees in this study felt that they had arrived at an enduring Organisational or Structural Plateau at age forty, 10 years before males. Women’s careers were frequently disrupted by the birth of and caring for children. The study concluded that Organisational/Structural Plateau is impractical for organisations to eliminate but that this form of Plateau does not appear to be a significant concern for older workers provided they are treated fairly. In addition, the study concluded that although Personal or Life Plateau would likely impact on work performance, resolution is not the role of commercial or most not for profit organisations and requires expertise that is not possessed by most organisations. Significantly, Job Content Plateau is important for older workers. This form of Plateau occurs when the employee has mastered the job in terms of new learning opportunities. Bardwick (1986) concluded that Job Content Plateau occurs when an employee has occupied the same job for between three to five years. This finding was supported by this Australian research and by the research of others. The career plateau phenomenon was found to apply generally across professions, industries and countries (Lim, Thompson and Teo, 1998). These researchers researched the Singapore police and found that job tenure of greater than six years negatively impacted job satisfaction and organisational commitment and produced a higher incidence of career plateauing. Chau (1998) reported similar patterns amongst internal auditors. A study of USA bankers by Corzine, Buntzman & Busch (1999) concluded that the expectation of plateauing increased with age. It is interesting to note that the Australian Bureau of Statistics (2004): CAT No 6209, data shows that 41% of the working population of Australia have been in the same job for more than five years; this compared with 43% from this researcher’s study. A consequence of career plateau is often a passive job. It was found that older worker jobs could be described as passive in more than 90% of the cases examined. These older worker jobs were typically low demand with high control but with workers experiencing low self efficacy which they attributed to their perceptions of age based discrimination and stereotyping at work. Consequently, the older workers in this study were under-utilised and at risk of stress. Many expressed their situation with anger rather than disillusionment.

6 WHY IS IT SO DIFFICULT TO GIVE WORKERS WHAT THEY WANT? Consistent patterns have emerged from this research, and that undertaken by others, of what is important to workers. The obvious question is: why don’t employers create and maintain workplaces where these factors are optimised? There appear to be many reasons. Adverse stereotypes about older worker are embedded within the broader Australian and global community (Taylor, Encel and Oka, 2002). 129

There are also a multiplicity of problems facing today’s organisations and its management. Organisations operate in a more complex, turbulent and hostile environment than do individual workers, and with many stakeholders who seek to optimise their particular situation. The worker is often self-centred, and can be demanding on the manager’s time. The manager has not only worker interests, but his/her own as well. There are issues of personality difference and perceptions of organisation fit. Porter (1998, pp. 383–389) noted that these modern realities present many practical problems for organisations, ‘achieving interrelationships in practice has proven to be extraordinarily difficult for many firms’. Porter attributes this to a formidable array of organisational impediments that involve problems of cost-sharing, management behaviours that include ‘protection of turf ’, attribution of blame, conflicts, performance measurement bias, matters of organisation structure and culture. In this milieu, the organisation endeavours to juggle competing and often conflicting aspirations whilst achieving the primary organisational task of remaining viable. Employee participation has been widely reported as providing a vehicle for introducing most of the intrinsic motivators into a workplace and its jobs. The organisational benefits of employee participation have been found to include enhanced productivity, improved creativity, acceptance of change and increased worker satisfaction (Pettigrew, 1986; Hull & Read, 2003; Scott-Ladd and Marshall, 2004; Appelbaum, Adam, Javeri and Lessard, 2005; Cort, 2005; Ortiz and Arnborg, 2005; Woodruffe, 2005). Successful participative workplaces must survive in the difficult environment alluded to by Porter. It necessarily requires supervision that is sufficiently confident and competent to manage boundaries and interfaces with other areas of the organisation. These competencies are rare in supervisors and managers (Karpin, 1995). Participative workplaces require managers throughout the organisation to let go of some of their traditional control over the work; and this is very difficult for many managers, particularly those with a low tolerance of ambiguity, low personal confidence and with inadequate political skills. The short-term profit expectations of shareholders and investors (Business Council of Australia, 2004), and the behaviour and needs of other stakeholders such as suppliers, places additional constraints on the time and effort that can be applied to addressing the concerns and needs of older workers. Key complications to satisfying worker demands come from investor pressures for good returns, within short time frames. Interestingly, most individual Australian workers are now both investors, through superannuation fund membership, and wage/salary earners. The investor-worker seeks maximum return on investment, and typically with a time horizon that may be as short as three months. In contrast, the worker has longer term plans for themself, and along the way, seeks improved conditions and payment. All of these business realities constrict management’s capacity to attend to the needs of the individual, older workers, who frequently are seen by management to no longer possess cutting edge skills, or exhibit the enthusiasm for effective intervention in the work scenarios mentioned above. This makes it imperative that older workers proactively manage their employability rather than expect the employer to do this for them. Arguably, older workers need to pay much more attention to skills currency to combat adverse community wide stereotyping.

7 CONCLUSION This research found that there is often a disconnect between the needs of a dynamic and complicated business environment, and the personal interests of older workers. This situation is compounded by evidence that the community, at large, hold adverse perceptions that lead to stereotyping of older workers and age based discrimination. This situation can be seen to result in self fulfilling outcomes as research has also consistently identified that job tenure increases with age and that there is a tendency for older workers to avoid formal education suggesting that many older workers may be out of touch with the needs of modern business. 130

This permutation can encourage older workers to become frustrated, and passive at work, and consequentially not aligned to organisational success. The combination of these circumstances has led more than 90% of the participants in this study to describe there careers as having plateaued. They complained of underutilisation and of personal stress that are features of the passive job they generally occupy. The reality that organisational plateau will eventually prevent promotion means that older workers should focus on prevention of job content plateau by actively seeking opportunities for personal growth. Workers are at risk of job content plateauing when job tenure exceeds three to five years. However, the difficulty that people have in conceptualising beyond what they know restricts their ability to pro-actively manage their circumstances and conceive new opportunities (Simon, 1952, in Riddalls and Bennett, 2003, pp. 414). Within a global business world, organisations are increasingly forced into difficult compromises in which the worker, without skills currency, is at the mercy of the organisation. A fundamental shift in thinking by workers offers a more empowered employment alternative for older workers. Marketable skills offers the worker the opportunity to break out of the inevitable career plateau, to be stimulated, to learn new things, to control time and to blend work with other interests. However, the extent of the shift in thinking is significant, and the case study participants are thought unlikely to make the transition. Employers can do a lot more to encourage high performance from its older workers. Employers and governments should encourage workers to place a greater emphasis on skills maintenance throughout their entire career, from career beginning to post sixty years of age. Workers should be encouraged to think about careers in ways that defy boundaries and barriers as they search for whole new pathways to success; this is more likely to occur if this type of thinking is encouraged from a young age. Employers could assist older workers prevent job content plateauing, and employers should pro-actively identify and eliminate age-based discriminatory practices. Although this research concluded that older workers typically have the personal attributes and motivation to work, there are likely to be valid concerns about their technical skill currency for today’s organisations. In addition, there are many hurdles for older workers to overcome to achieve personal satisfaction whilst working in today’s organisational environments. Pro-active and wise employer leadership is needed to manage the tension between what is possible in the employment of older workers and the realities of the workplace. REFERENCES Appelbaum, S H, Adam, J, Javeri, N, and Lessard, M, (2005). A Case Study Analysis of the Impact of Satisfaction and Organisational Citizenship on Productivity. Management, Research, vol. 28, no. 5: pp. 1–26 Arthur, M B, Khapova, S N, and Wilderom, C P M, (2005). Career Success in a Boundryless Career World. Journal of Organisational Behaviour, vol. 26, no.: pp. 177–202 Australian Bureau of Statistics (2004). Measures of a Knowledge-based Economy and Society, AustraliaHuman Capital Indicators www.abs.gov.au (viewed 6 April 2005), Canberra Bardwick, J M, (1986). The Plateauing Trap: How to Avoid it inYour Career andYour Life. NewYork: American Management association Business Council of Australia, (2002). Employability Skills for the Future. In: Commissioned by Commonwealth of Australia-Director, T R S, Dept of Education, Science and Training (Ed.) Melbourne. 65 pages Business Council of Australia, (2004). Beyond the Horizon: Short-Termism in Australia: A Call to Think into the Future. IN B.C.A (Ed.) Melbourne, 36 pages Cort, A, (2005), Toeing the Line. Assembly, vol. 48, no. 9, August: pp. 40–47 Corzine, J B, Buntzman, G F, and Busch, E T, (1999). Machiavellianism in U.S. Bankers. International Journal of Organizational Analysis, vol. 7, no. 1, January: pp. 72–84 Erez, M, Kleinbeck, U and Theierry, H, (2001). Work Motivation in the Context of a Global-izing Economy, New Jersey, Erlbaum Ference, T P, Stoner. J, a, F & E., and Kirby, W, (1977). Managing the Career Plateau. Academy of Management Review, October: pp. 602–612

131

Hull, D, and Read, V, (2003). Simply the Best Workplaces in Australia. IN ACIRRT (Ed.) Sydney, University of Sydney, 41 Pages Kanfer, R, and Ackerman, P L, (2004). Aging, Adult Development, and Work Motivation. Academy of Management Review, vol. 29, no. 3, July, 22 pages Karpin, D, (1995). Enterprising Nation: Reviewing Australia’s Managers to meet the challenges of the AsiaPacific Century. Report of the Industry Task Force on Leadership and Management Skills, Commonwealth of Australia, 408 pages Latham, G P, and Pinder, C C, (2004). Work Motivation Theory and Research at the Dawn of the Twenty-First Century. Annual Review of Psychology, vol. 56, June: pp. 485–516 Lemire, L, Saba, T, and Gagnon, Y-C, (1999). Managing Career Plateauing in the Quebec Public Sector. Public Personnel Management, vol. 28, no. 3: pp. 375–392 Lim, V K G, Thompson, S H, and Teo, (1998). Effects of Individual Characteristics on Police Officer’s Work-related Attitudes. J. Managerial Psychology, vol. 13, no. 5/6: pp. 334–341. 8 9 Ortiz, J P, and Arnborg, L, (2005). Making High Performance Last: Reflections on Involvement, Culture, and Power in Organisations. Performance Improvement, vol. 44, no. 6, July: pp. 31–38 Pettigrew, A M, (1986). Is Corporate Culture Manageable? Sixth Annual Strategic Management Conference: Cultures and Competitive Strategies. Singapore 13–16, October Porter, M E, (1998). Competitive Advantage, New York, Free Press Riddalls, C E, and Bennett, S, (2003). Quantifying Bounded Rationality: Managerial Behaviour and the Smith Predictor. International J. of Science Systems, vol. 34, no 6, 15 May: pp. 413–26 Ronen, S, (2001). Self-actualization: Implications for Motivation Theories. In See Erez,Kleinbeck & Theierry, 2001 Scott-Ladd, B, and Marshall, V, (2004). Participation in Decision Making: A Matter of Context? Leadership and Organisation Development Journal, vol. 25, no. 7/8: pp. 646–658 Taylor, P, Encel, S, and Oka, M, (2002). Older Workers: Trends and Prospects. The Geneva Papers on Risk and Insurance, vol. 27, no. 4, October: pp. 512–533 Woodruffe, C, (2005). Employee Engagement: The Real Secret of Winning a Crucial Edge Over Your Rivals. The British Journal of Administrative Management, Dec 2005/ Jan 2006, pp. 1–5

132

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Can organisations influence employees’ intentions to retire? Jodi Oakman Centre for Ergonomics and Human Factors, La Trobe University, Bundoora, Vic, Australia

Yvonne Wells Lincoln Centre for Research on Ageing, La Trobe University, Bundoora, Vic, Australia

ABSTRACT: The ageing of Australia’s population is having a dramatic impact on the labour market. This study examines employees’ retirement intentions and the main influences on changing these intentions in a large public sector organisation interested in developing policies to assist with retention of their mature age workforce. In a multivariate model which included age, gender, work factor measures and the SWAI (Shortened Work Ability Index), age was found to be the strongest independent predictor of retirement intention. However, when age was controlled, the SWAI was the second strongest predictor. Measures of the work environment had no independent influence on retirement intention. These findings are of great significance to organisations wanting to improve retention of mature age workers. Firstly, organisations need to change perceptions about retirement. Secondly, because participants’ perception of their capacity to work effectively was a strong independent predictor of intention to remain at work, it is of utmost importance to ensure that employees are assured that they have the physical and emotional capacity to remain in their current employment.

1 INTRODUCTION Australia has an ageing demographic. This is effecting the labour market and is going to require a shift in corporate culture so that older workers are valued and retained for longer (Productivity Commission, 2005). This phenomenon is not unique to Australia but is being experienced throughout the OECD (Auer and Fortuny, 2000). Dependency ratios, that is the number of workers compared to the number of non workers or dependents in a given population, are decreasing rapidly, thereby increasing pressure on a smaller number of workers (Sheen, 2001). Australia presently has around 170,000 new entrants to the labour force each year; however, for the entire decade of 2020–2030, there will be a mere 125,000 new entrants (Access Economics Pty Ltd, 2001). One strategy to mitigate the inevitable labor shortfall is to delay the retirement of older workers. Despite opinions to the contrary, worker retention has many benefits for companies and is a cost effective method of maintaining workforce numbers (Brooke, 2003). There are many misconceptions about the capacities and liabilities of older workers, and a concerted re-education program is needed to ensure that the real value of older workers is understood and retention of this group of employees is perceived as an important part of an organisations’ business model (Walker, 1999). One American survey found mixed views about older workers (Munnell et al., 2006); older managers were more likely than younger ones to be receptive to both employing and retaining older workers, and there was general consensus that mature age employees were more productive that their younger peers. In the literature, several models have been used to predict retirement age or intended retirement age, which are constructed in a variety of ways (Taylor and Shore, 1995). Greater use of objective 133

measures of retirement intention is needed, so that more cohesive and constructive planning can be achieved. In this article a model of retirement intention, which included a wide range of predictive factors, was evaluated; these factors included age, ability to work, general health, influences from outside work, superannuation, psychosocial factors and job demands. The work ability concept was developed in 1981 (Ilmarinen, 1999; Ilmarinen, 2001). Subsequently, the work ability index (WAI), a self report measure of an individual’s work ability, was constructed (Ilmarinen et al., 2005). The WAI has been widely used in Europe, but its use in Australia has been limited. It has the potential as a practical tool for employers in managing the ageing of their workforce. The WAI allows employers to quantify the perceived workability of their workforce and therefore enables effective strategies to be developed that more accurately reflect the requirements of their employees. This is mutually beneficial, as when workers are able to perform optimally in the work place, productivity can be improved. This study aims to look at what factors predict intention to retire and whether organisations can influence these factors so that employees will delay retirement.

2 METHODS 2.1 Participants A large government organisation commissioned La Trobe University to examine its work force and assist with the development of strategies aimed at retaining workers who may otherwise be considering retirement. A questionnaire was developed and distributed via the company’s intranet site and 332 employees, most of whom worked full-time (88.5%), responded. Respondents comprised 130 (39%) women and 202 (61%) men. The organisation has equal numbers of men and women in its employment. The mean age of the survey respondents was 42.1 years, in comparison to the organisational average of 43.5 years. 69% of respondents had tertiary qualifications, and 39% indicated that they had 10 or more years of service with the organisation. Of this subgroup the average age was 52.7 years. 2.2 Survey To examine whether an organisation can influence an employee’s retirement intentions, a model was created which included potential predictors of retirement intention: age, ability to work, general health, influences from outside work, superannuation, psychosocial factors and job demands. The survey comprised several well established measures: the Work Ability Index (Ilmarinen, 2001), The Copenhagen Psychosocial Questionnaire (COPSOQ) (Kristensen et al., 2005) and the General Health Questionnaire (GHQ12) (Goldberg and Williams, 1988). The WAI index was shortened at the request of the employer organisation, and hence is referred to as SWAI (Shortened Workability Index): the question relating to the number of current diagnosed diseases was removed. Retirement intentions were measured through two items: anticipated timing of retirement and employment plans after leaving the organisation. 2.3 Analysis Preliminary analyses of intention to retire (used as a continuous variable) indicated that relationships with several potential predictors were non-linear. In particular, there was a sharp differentiation between respondents who intended to retire in less than two years or 2–5 years and those who intended to retire later (in 6–10 years, 11–15 years, or 16 or more years). For this reason, the dependent variable, intention to retire, was dichotomised at 5 years. Logistic regression analysis was used to test whether intention to retire was associated with organisational factors. Variables representing socio-demographic factors (age, gender, marital status and dependent children) and mental health (GHQ) were entered first, followed by four measures of work 134

factors: work satisfaction, work demands, social cohesion, and job control. The Short Workability Index (SWAI) was entered in the final step. 3 RESULTS Of the socio-demographic variables, only age significantly predicted intention to retire within the next five years. Older respondents were significantly more likely to intend to retire within the next five years than younger ones. When the four measures of work were entered, age remained a significant predictor of intention to retire, but was joined in the model by Work satisfaction and Social cohesion (see Table 1). Addition of the SWAI to the regression equation significantly improved intention to retire within the next five years. SWAI was a significant predictor of intention to retire, but acted as a mediator between the work factors and intention to retire. Respondents were more likely to retire within the next five years if they were older rather than younger and if their perceived ability to work was low. 4 DISCUSSION With a shrinking labour market the retention of employees is gaining greater importance. There is increasing interest in the psychosocial work environment as organizations strive to maximize the potential of their workers.

Table 1.

Results of logistic regression analyses.

Variable Step 1 Age Gender Partnered Dependent children GHQ12 Step 2 Age Gender Partnered Dependent children GHQ12 Work satisfaction Work demands Work social cohesion Work control Step 3 Age Gender Partnered Dependent children GHQ12 Work satisfaction Work demands Work social cohesion Work control Short Work Ability Index

B

Wald

Df

Sig

Odds ratio

.24 −.29 .56 −.26 .02

52.51 .59 1.67 .54 .20

1 1 1 1 1

.000 .441 .196 .464 .654

1.27 .75 1.75 .77 1.02

.25 −.14 .62 −.25 .02 −.53 −.03 .41 .13

53.11 .13 1.87 .48 .18 5.31 .02 4.68 .43

1 1 1 1 1 1 1 1 1

.000 .718 .172 .489 .670 .021 .890 .031 .514

1.29 .87 1.85 .78 1.02 .59 .98 1.50 1.14

.25 −.28 .59 −.33 −.04 −.33 −.08 .35 .20 −.14

49.23 .52 1.71 .77 .94 1.84 .16 3.07 .90 8.75

1 1 1 1 1 1 1 1 1 1

.000 .471 .192 .381 .332 .175 .689 .080 .343 .003

1.28 .75 1.81 .72 .96 .72 .92 1.41 1.22 .87

(in bold: significant odds-ratios at p < 0.05)

135

In this study we were interested in establishing which factors influence a person’s intention to retire. Intuitively we expected that organizational factors, (such as role choices in organizations, job satisfaction, length of tenure, and job performance) would influence decisions relating to employment and intention to retire. However, often the evidence to support these expectations is sparse. While the results from this study support the importance of workplace factors, the relationship is more complex than a simple additive relationship between the psychosocial work environment and work ability and their influence on retention of employees. SWAI acted as a mediator between the workplace factors and intention to retire. This result is of key importance to organisations as it enables them to have some influence over whether their employees might choose to remain at work or retire. Modification of the work environment may have significant impact over an individual’s intentions. Mein et al. (2000) found in the widely published Whitehall II study that self-rated health, employment grade and job satisfaction are all independent predictors of early retirement. In contrast, the current study found that intention to retire was not independently associated with any of the work factors, including work satisfaction. Instead, work factors were mediated by the SWAI and it was through this pathway that these factors influenced intention to retire. Other studies (Cotrim et al., 2005; Hopsu et al., 2005; Seitsamo, 2005) have investigated the used of WAI as a predictor for retirement and also found it a powerful predictive tool in identifying employees who are likely to retire early. This suggests that the WAI is a useful means to assist managers to understand perhaps, realizing greater potential from their employees. Workplace factors were found to be an important contributor in the determination of an individual’s intention to retire. It is imperative for workplaces to ensure that they maintain a positive psychosocial environment, and this should be an important and strategic goal for organisations that are attempting to retain older workers for longer. Such a goal is particularly important in organisations with a high median age, as these workplaces are most vulnerable to the impacts of the ageing workforce. More work, both at an organisational and at a much broader societal level, is needed to change perceptions about when retirement should happen. Retirement needs to be seen not as a definitive point, but as a transitional state. Support is needed from all levels – from governments, organisations and individual employees – for this change to occur. A cultural shift is also needed so that it becomes acceptable to work as long as a person is both willing and able to, rather than retiring because they reach a certain age.

5 CONCLUSIONS Longitudinal evaluations of workplace interventions that target retention of older workers are required, so that their effectiveness can be evaluated. This will ensure that there is information available for organisations to develop good quality strategies that accurately address the requirements of ageing workers. REFERENCES Access Economics Pty Ltd. (2001). Population Ageing and the Economy: Commonwealth of Australia. Canberra. Auer, P., and Fortuny, M., (2000). Ageing of the Labour Force in OECD Countries: Economic and Social Consequences: Employment Sector International Labour Office. Geneva. Brooke, L., (2003). Human resource costs and benefits of maintaining a mature-age workforce. International Journal of Manpower, 24(3): pp. 260–283. Cotrim, T., Simoes, A., Ramalho, F., and Paes Duarte, A., (2005). Why healthcare workers ask for early retirement at a central Portugese hospital? Workability preliminary results. In: International Congress Series 1280: (pp. 258–263). Amsterdam: Elsevier.

136

Goldberg, D., and Williams, P. A., (1988). User’s Guide to the General Health Questionnaire. NFER-NELSON. Hopsu, L., Leppanen, A., Ranta, R., and Louhevaara, V., (2005). Perceived work ability and individual characteristics as predictors for early exit from working life in professional cleaners. In: International Congress Series 1280: (pp. 84–88), Amsterdam: Elsevier. Ilmarinen, J., (1999). Ageing workers in the European Union – Status and promotion of work ability, employability and employment. Helsinki: Finnish Institute of Occupational Health. Ilmarinen, J. E., (2001). Ageing workers. Occupational & Environmental Medicine, 58(8): pp. 546–552. Ilmarinen, J. E., Tuomi, K., and Seitsamo, J., (2005). New dimensions of Work ability. In: International Congress series 1280: (pp. 3–7), Amsterdam: Elsevier. Kristensen, T. S., Hannerz, H., Hogh, A., and Borg, V., (2005). The Copenhagen Psychosocial Questionnaire-a tool for the assessment and improvement of the psychosocial work environment. Scandinavian Journal of Work, Environment & Health, 31 (6): pp. 438–449. Mein, G., Martikainen, P., Stansfeld, S. A., Brunner, E. J., Fuhrer, R., and Marmot, M. G., (2000). Predictors of early retirement in British civil servants. Age and Ageing, 29 (6): pp. 529–536. Munnell, A. H., Sass, S. A., and Soto, M., (2006). Employer Attitudes towards older workers: survey results: Centre for retirement research, Boston College. Productivity Commission, (2005). Economic Implications of an Ageing Australia: Australian Government: Productivity Commission, Canberra. Seitsamo, J., (2005). Qualities of work, functioning and early retirement. A longitudinal study among Finnish ageing workers in 1981–1997. In: International Congress Series 1280: pp. 136–141, Amsterdam: Elsevier. Sheen, V., (2001). Challenging Convention: Australia’s ageing workforce – the challenge for human resource management. Paper presented at the Australian Human Resources Institute Conference, Melbourne Convention Centre, Melbourne. Taylor, M. A., and Shore, L. M., (1995). Predictors of Planned Retirement Age: An Application of Beehr’s Model. Psychology and Aging 10 (1): pp. 76–83. Walker, A., (1999). Combating Age Discrimination at the Workplace. Experimental Aging Research, 25 (4): pp. 367–376.

137

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Extending the working life Robert G. Goedhard Social Security Agency UWV, Utrecht, The Netherlands

Willem J.A. Goedhard The Netherlands Foundation of Occupational Health and Aging, Middelburg, The Netherlands

1 INTRODUCTION In 2006 an interesting book was published entitled: “Towards a longer work life, aging and the quality of work life in the European Union” (Ilmarinen, 2006). This title inspired us to explore the question: how much longer could the active work life be extended in the future? Also: “would it be feasible to extrapolate the findings from cross-sectional WAI studies to higher ages and can these results be used to determine the onset of retirement?” The WAI (Work Ability Index) is an instrument to monitor the work ability of aging workers. Early retirement policies for workers below the age of 65 are rapidly changing in the EU (RedayMulvay, 2005). Present day policy of the EU is to increase average retirement age by five years in the year 2010; i.e. the average retirement age in the European Union should become 65 years of age in 2010. During the last decade the participation rate of workers of 55–64 years is increasing rapidly (CBS, 2007). Ilmarinen (2006) raised the question “Is it possible to continue to work in the same job at 60 years of age?”

2 METHODS For this study we made use of regression models that were obtained in three different studies with WAI in the Netherlands. These studies were performed among workers in the metal industry, the electro-technical industry, and white collar civil servants of the Defense department. Table 1 shows the summarized findings of these studies. For the present study we used the results of a simple linear regression analysis upon the data concerning WAI and age using Statgraphics and extrapolated the data until a specific cutoff point was reached. The cutoff point is in this study defined as the age at which an average WAI-score of 37 is reached (see Figure 1).

Table 1. The studied populations.

Studied population

N

mean age (range)

Mean WAI ± s.d.

WAI < 37 (poor-moderate)

Metal industry Electro-technical industry Civil servants (Defense dept.)

89 194 126

41.5 (19–59) 39.9 (23–60) 43.0 (21–61)

40.9 ± 4.86 42.7 ± 5.18 41.9 ± 4.25

20.2% 14.1% 13.5%

All

409

41.2 (19–61)

41.2 ± 4.89

14.0%

139

49 45

WAI

41 37 33 WAI vs. AGE WAI
37 AGE
68.89

29 25 15

25

35

45

55

65

75

85

AGE

Figure 1.

Table 2.

Example of extrapolating the regression line and cutoff at WAI = 37.

Results of the regression analysis.

Studied population

N

Model WAI vs. AGE

corr coeff

Metal industry Electro-technical industry Civil servants (Defense dept.)

89 194 126

WAI = 0.14 ∗ AGE + 46.72 WAI = 0.16 ∗ AGE + 48.93 WAI = 0.10 ∗ AGE + 46.11

−0.30 −0.28 −0.25

All

409

WAI = 0.14 ∗ AGE + 47.68

Age vs. WAI R2 (exp.variance)

Upper age limit WAI = 37

0.091 (9.1%) 0.077 (7.7%) 0.062 (6.2%)

69 76 90

0.078 (7.8%)

78

3 RESULTS The studied populations don’t differ much concerning mean age and mean WAI scores. They do differ in the percentage of workers with a poor or moderate workability. As is shown in Table 1 the workers in the metal industry have the highest risk of a poor-moderate WAI score. Table 2 describes the models and the corresponding data. The explained variances vary considerably between the different work groups as is also expressed by the correlation coefficients. Using the results of the linear regression analysis the ages at which the cutoff points were reached also differ considerably (see also Figures 2–4). The lowest age of the intercept of the regression model with the WAI score of 37 is observed in the group workers at the metal industry (blue collar workers). The highest age intercept is found in the white collar workers group of civil servants. The 95% confidence intervals were calculated by the statistical program (Statgraphics).

4 DISCUSSION From the above mentioned findings the preliminary conclusion can be reached that continuation of employed work above 65 years is feasible, provided that WAI score is within the category good, i.e. higher than 37. Is this a reasonable assumption? We have to realize that there is a lack of information about the decrease of work ability of individual workers over 65 years of age. 140

WAI

49

37

25 15

25

35

45

55

65

75

85

AGE

Figure 2.

Data analysis of the metal industry population (n = 89). WAI = 46.72 − 0.14 ∗ Age (95% C.I.).

WAI

49

37

25 15

Figure 3.

25

35

45 55 AGE

65

75

85

Data analysis of the electro-technical industry population (n = 19). WAI = 48.9 − 0.16 (95% C.I.).

∗

Age

Is it possible to ask workers to extend their working life to higher age limits than so far acceptable? It is well known that health, one of the key elements of work ability, is deteriorating in many older individuals. In the Gothenburg longitudinal study in Sweden, however, it was demonstrated that in the age interval 70–81 a considerable proportion of people do not suffer from any definable disorder (Svanborg, 1988). Continued working can also be regarded as a positive development (Butler, 1988). Productive aging can possibly become the start of new initiatives in Occupational health. The work ability Index is proven to be a useful instrument in occupational health. To our knowledge the instrument has not been used to determine the onset of retirement. Studies are presently carried out concerning flexible retirement policies (Maltby et al., 2004). This would imply that older workers extend the working life beyond the age of 60. Preliminary conclusions of the present study are: • Older people should be encouraged to extend their working life. • The nature of work exposure is an important factor in determining the age of retirement (blue collar workers should be dealt with in a gently way). • The WAI can probably be used to determine the individual age of (gradual) retirement. 141

WAI

49

37

25 20

Figure 4.

30

40

50

60 AGE

70

80

90

Data analysis of the civil servants population, working for the Defense Department (n = 126). WAI = 46.1 − 0.10 ∗ Age (95% C.I.).

• The active working life can possibly be extended beyond 65–70 years (WAI surveillance warranted). • Age is explaining only a small percentage of the WAI variance; studies with other variables [e.g. stress, VO2-max] are necessary. • Data need to be collected of studies among older workers over the age of 65. REFERENCES Butler, R.N., (1988). Aspects of aging and health in the age interval 70–85. In: Schroots, J.J.F. et al. (editors): Health and Aging, Springer-Swets, Lisse, The Netherlands. pp. 143–151. CBS (central Office statistics, Heerlen), (2007). Ilmarinen, J., (2005). Towards a longer work life. Ageing and the quality of work life in the European Union., FIOH, Helsinki Maltby, T. et al., (2004) Ageing and the transition to retirement. Ashgate, Aldershot, England. Reday-Mulvey, (2005). Working beyond 60. Palgrave MacMillan, New York. Svanborg, A., (1988). Aspects of aging and health in the age interval 70–85. In: Schroots, J.J.F. et al. (editors): Health and Aging, Springer-Swets, Lisse, The Netherlands. pp. 133–141.

142

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Pension preferences and work environment Mikael Stattin Department of Sociology, Umeå University, Umeå, Sweden

ABSTRACT: In order to meet the challenges posed by population ageing many contemporary countries are striving for a prolonged working life among elderly employees. However despite implemented policy measures, average retirement age are in most OECD countries well below the official retirement age and also in relation to the levels of the 1960s and the 1970s. Policy reforms that modify financial incentives in social security programs might not be enough to encourage later retirement. Therefore, it is important to perform research that might improve our understanding of factors that affects people’s preferences towards the timing of retirement. The aim of this study is to (a) map preferred retirement age among different groups of employees aged 55–64 of age in Sweden and (b) to analyse the relationship between pension preferences and exposure to various working conditions, such as physical and psychosocial work environment. The study is based on data from the PSAE study (Panel Survey of Ageing and the Elderly), which is a large-scale survey of living conditions among the elderly that was launched in 2002–2003 in Sweden. A basic feature of the PSAE is its integration into Statistic’s Sweden’s annual Survey of Living Conditions, which among other things means that the PSAE contains longitudinal data that in some cases goes as far back as to 1979. By means of the longitudinal power in the data material the analysis address the question to what extent different work environment factors are causally associated with pension preferences later in life.

1 INTRODUCTION How long should someone be expected to work during their lifetime? At what age is it reasonable to stop working? In recent decades, questions such as these have attracted growing interest as research questions, as a policy area, and as an important issue among the public as a whole (SOU, 2002, 2003, OECD, 2000, 2003). The reason for this, of course, is the somewhat paradoxical trend of increasing numbers of people retiring from work prior to the statutory retirement age, despite increasing life expectancy, improved standards of living, and improved health. In most OECD countries, the average retirement age is well below the statutory pension age (OECD, 2006). This so-called early exit trend is a phenomenon that is present in most parts of the industrialised world, though the scale of early departure from work varies∗ . It is now a widely accepted fact that the early exit trend represents a substantial structural change in the labour market. Most scholars view the phenomenon as problematic. First of all because reason most early retirees are supported by different public social insurance programmes, meaning that an increase in early retirement represents a growing economic burden for the state. In combination with population ageing, the early exit trend implies an obvious risk that the pressure on public transfer systems will increase, since older people tend to consume more welfare services than younger people. Another negative ∗ This

development has during recent years been halted in some countries. The average retirement age is not decerasesing as previously. (OECD 2006)

143

anticipated consequence of the early exit trend is labour shortage. As the decreasing number of young people entering the labour market is paralleled by increased early exit, the labour force will stop growing or even begin to shrink in some countries, and this in turn might have consequences for economic growth (OECD, 2006). As a response to this development, many countries have taken initiatives in order to postpone the transition from work to retirement and to increase labour market participation among older people (OECD, 2003, Ilmarinen 1999, SOU, 2002:29, Ds 2002:10, SOU, 2003:91). These initiatives have mainly involved strengthening the economic work incentives in pension programmes, and also by allowing more flexibility in these programmes. Much research on early retirement has focused on the causes of premature withdrawal from work. However, an equally important research task is to study the part of the elderly work force that is still in work. Of vital importance here is information concerning the relationship between the ambitions of recent political reforms and retirement preferences in the elderly work force. This is also the point of departure of this paper, the overall aims of which are (a) to study the pension preferences of the elderly work force in Sweden; that is, to discover the age at which they wish to retire, and to uncover the reasons for this, and (b) to analyse the relationship between pension preferences and working conditions. As background to the analysis, the next section presents a general discussion of the determinants of early exit. 1.1 Why do people retire early? Although there has been a great deal of research into the predictors of premature departure from work, no clear and simple explanation has been found. Instead, research has shown that the causes are embedded in a complex interaction of different factors. In relation to the present ambition to increase labour market participation among the elderly, it is noteworthy that the early exit trend started as a reaction to high unemployment levels and the need for structural change in the aftermath of the oil crises in the 1970s. High unemployment rates led governments in many countries to reduce the supply of labour, in order to improve the labour market chances of the younger members of the workforce. A common way of doing this was to implement various social policy measures and programmes that facilitated early exit from work (Wilensky, 2002). A common explanation for the large outflow of older workers relates to how the incentive structures in these programmes were designed. Here it is argued that early exit is mainly an effect of individual behaviour and generosity in social policy programmes. There is scientific support for such an effect, not least through comparative studies that have shown that countries with generous programmes have proportionally more early retirees than countries with more restrictive rules (OEDC, 2003c, Gruber and Wise, 2002). However, when studying pension preferences this relationship seems to be less straightforward. Esser (2006) has demonstrated that generous welfare programmes may also have a positive effect on the motivation to work. In Esser’s study, the preferred pension age was higher in countries with generous early retirement programmes. The explanation for this is that these countries often have highly regulated labour markets with comparatively good working conditions, and an explicit ambition to maximise the integration of working-age people in paid work. Thus, relying solely on incentive structures in social policy schemes seems far too simplistic to be of much use in the attempt to understand the reasons behind pension preferences and early exit from work. The institutional construction of early exit opportunities has long been embedded in a specific pattern of values regarding the relationship between old age and work – that early exit first and foremost is a good thing, and that it represents an improvement of the welfare state. It has been seen as reasonable that older workers who have been working for a long time have earned the right to be unchained from demanding and unhealthy working conditions. In addition, early exit has been viewed as an act of solidarity in relation to younger workers – as someone older leaves the workforce, someone younger enters. Around this value pattern there has been an obvious consensus between different actors such as the state, employers, and employees, especially in times of economic recession. 144

De Vroom and Guillemard (2002) argue that these externalisation practices have shaped and fostered an early exit culture in many contemporary societies. Early retirement is to an increasing extent perceived by the public as an acquired social right, and the social norm that stipulates a duty to work has been replaced by a new norm – the right to leave. The massive outflow of older workers has legitimised and institutionalised this norm, which in turn has made it difficult to “shift the vicious circle of early exit to the virtuous circle of active aging” (de Vroom and Guillemard, 2002). In most countries today, the use of early exit as a way to systematically reduce the supply of labour is viewed as a reprehensible political strategy. Large premature withdrawals are incompatible with demographic changes, and represent a waste of both manpower and competence. However, it has proved difficult to decrease the outflow of older workers as well as to find universal solutions to the problem (Ebbinghaus, 2000). Many scholars state that a reduction of early exit rates requires interventions in several societal areas, and also measures that take into account non-financial incentives. One such area of great importance is the labour market, along with the environmental and structural constraints that face older workers, and that might push people out from work. It has been shown that early exit from work is often an involuntary process, and a more or less forced one. As an effect of structural change in the labour market, technological change, or increased international competition, the opportunities for older workers to keep or to get a job may be jeopardised. Exposure to working conditions that negatively affect health, motivation, and job chances has proven to be of great importance for both early exit and pension preferences (Stattin, 1998, RFV, 2001a, Soidre, 2005, Esser, 2006, and Siegrist, 2007). Other important aspects include attitudes towards elderly workers and age discrimination, since stereotypes and generalising assumptions about the work ability and productivity of elderly workers are quite common in working life (RFV, 2001, and Ilmarinen, 1999). Researchers today are in strong agreement that these kinds of push factors are important causes of early exit (Ebbinghaus, 2000). Many of the efforts to reduce early exit rates have involved modifications of existing social policy programmes, while less attention has been paid to quality of working life. At the same time, there is a great deal of data indicating that working life is characterised by increasing demands, leaner production, more precarious contracts, and so on (Marklund et al., 2005). This development hardly improves the possibilities for older workers at the labour market. Therefore, there are several good reasons to more closely study the relationship between working conditions and the pension preferences of older workers. In this paper, two main questions are in focus. The first question concerns pension preferences per se among elderly workers in Sweden, and the causes that people state for their preferences. This question relates to the above discussion about whether there are signs of an early exit culture in Sweden. The second question relates to a push perspective, and involves analysis of the relationship between working conditions and pension preferences. To answer with question, we will initially present a conventional cross-sectional analysis of the associations between different indicators of work conditions and pension preferences. However, such analyses are often troubled with various kinds of selection bias. Respondents with poor health and reduced wellbeing might be more inclined to report poor working conditions, and respondents might rationalise their pensions preferences by attributing them to poor working conditions. Therefore, we have also used longitudinal data in order to improve the quality of the analysis.

2 DATA The results presented in this paper are based on data from a large scale survey (The Panel Survey of Ageing and the Elderly – PSAE) of living conditions among the elderly in Sweden, launched in 2002-2003. The survey was coordinated with the Statistics Sweden annual survey on living conditions (ULF). ULF is based on a representative sample of 6,000 Swedes aged from 16 to 84, and is organised around twelve different arenas of living conditions; these arenas were therefore 145

also covered by the PSAE. ULF normally has no specific focus on elderly people, and so a number of adjustments were made in the 2002–2003 ULF. The questionnaire was redesigned and augmented in order to fit the purpose of the study. The sample size was increased by 2,000 individuals in the age category 55+. The upper age limit was abolished, and a substantial extension of the questionnaire for age category 55+ was included. The purpose was to facilitate analysis of different phases of ageing with the aim of studying work conditions, work ability, and pension preferences in the elderly work force; the transition from working life to retirement; and social integration, health development, and care needs, especially among the retired population, including the oldest old. The fieldwork was performed as personal interviews. Of the net sample, 74.9% participated in the survey. The drop out was not equally distributed over the different age groups, and so weights that compensate for this were applied in the present study. 2.1 Variables The subjects of this study comprised those respondents aged between 55 and 64 who were still in work. The dependent variable was the age at which the respondents wished to retire. In the PSAE study, the question was formulated as follows. Today you have certain possibilities to make your own decision as to when to retire. This might be before as well as after 65 years of age. Taking into account your current health, your financial situation, your work situation, and the current pension rules, at what age do you want to retire? Depending on the answer to this question, a number of follow-up questions were asked about the reasons for the preferred pension age. In the analyses described below, a number of different indicators of work conditions were related to pension preferences. Physical as well as psychosocial work environment indicators were used, and in several cases these were transformed to indexes. The physical work environment index was based on questions about repeated movements, awkward work positions, vibration, and heavy lifting. The psychosocial work environment was defined according to the demand-control model (Karasek, 1990). Work demands were operationalised as both quantitative (amount of work) and qualitative demand (work task complexity). The control index was based on questions concerning the degree of influence the worker had over how, when, and where to do the work. Indicators of instrumental and emotional support from work colleagues and managers were also analysed. Finally, in addition to indicators of work conditions, background variables such as age, education, and social class were also included in the analyses. The panel data in the analyses were applied by comparing indicators of work conditions at baseline (T1), and relating these to how the respondent reported the same indicator at T2. The follow-up period was eight years, according to the panel wave intervals in the ULF survey. The analysis enables us to scrutinise the extent to which pension preferences are affected by long-term exposure to certain work environments, and improvements or deterioration in working conditions.

3 RESULTS We will first describe how the respondents answered the question about their preferred pension age (Figure 1). A majority of the respondents preferred an age below the age of 65; almost 50% preferred an age between 60 and 64, while a small proportion preferred an age between 50 and 59. Four out of ten respondents preferred a retirement age of 65, while only around 7% preferred an age over 65. There were some differences between men and women, but they were very small. Figure 2 shows the reasons for the preferred pension age given by those who reported a preferred pension age below the age of 65. It is important to note here that all respondents had the opportunity to reply yes or no on all alternatives. One point of interest is that more than 80% of the respondents preferring an early exit did so because they wanted to have more spare time. Around a third reported 146

60.0

50.0

40.0 Men Women All

30.0

20.0

10.0

0.0 50–59

60–64

65 Age

66–69

70

Preferred pension age among members of the workforce aged 55–64 years. N = 1,360.

Figure 1. 90 80 70

Percent

60 50

Men Women All

40 30 20 10

r

ill w s sk ta

k

at

ea sa di

tw ns io W or

ic lif ua Q

pp

te an

re no

m s ct pe

ex er

Em

pl oy

d

e tir

ng e

m de o to

e m co be

s ha k

W or

to

an

w

di

or

k

se

nt rre

ith w d fie

at is D

is s

an W

cu

m so

tt o

do

lt na tio

el ng hi

et

k or w o

or tm ra lly ic a om

til

ar sp e

er W an N

ot

ec

on

un

e

re is

re rtn

65

e

tir

tim

ed

e tiv la

ea H

of e ar C

Pa

N or

m

al

pe

ns

io

n

ag

e

lth

0

Figure 2.

Reasons for preferring to retire before the age of 65, among members of the workforce aged 55–64 years. N = 703.

health reasons, and stated that work had become too demanding. One in five wanted to retire early because their partner had already retired; and here there was a considerable difference between men and women. Only a small proportion of respondents reported any of the other reasons. Thus, there were three dominant reasons among those who preferred an early exit. The most common reason – wanting to have more spare time – was fairly equally common across various groups (see Table 1), while larger differences were seen concerning the reason of health. Figure 3 presents the reasons for preferring a late exit (>65). Here it is obvious that aspects related to work were of great importance. Almost nine out of ten stated that work gave their life meaning, and almost as many said that they liked their work. The social importance of work colleagues was 147

Table 1.

Percentages of respondents giving the three most common reasons for a preference for early exit <65. Sign. p = 0.01***, p = 0.05**, p = 0.1*

Gender Male Female Age 55–59 60–64 Socioeconomic position Blue collar I Blue collar II White collar lower White collar middle White collar higher Self employed Education Compulsory schooling Secondary school University

Health

Spare time

35.2 33.0

83.6 83.4 *** 86.3 76.5

33.9 35.0 *** 46.6 45.9 31.5 19.7 20.6 44.6 *** 39.2 26.0 25.6

80.2 82.4 87.0 86.6 88.8 81.8 * 80.5 90.2 85.5

Work too demanding *** 28.1 38.4 33.0 33.0 35.1 34.9 25.9 36.5 29.9 20.8 33.1 28.2 36.8

100 90 80 70 Percent

60 Men Women All

50 40 30 20 10 0 Need the income

Figure 3.

To increase future pension benefit

To retire Like my work Daily contact Work gives Employer Work same time with work life meaning appreciates colleagues as partner colleagues is my want me to valuable competence

Reasons for preferring to retire post-65, among members of the workforce aged 55–64 years. N = 96.

another common reason, as was the possibility of working longer in order to increase future pension income. There were some differences between men and women; however, due to low numbers, no further comparisons between groups are made here. 3.1 Factors affecting pension preferences The next step of the analysis was concerned with associations between pension preferences and a number of independent variables. The independent variables covered background and individual 148

Table 2. Associations between pension preferences and age, health, education, and socioeconomic position. Unstandardised regression coefficients

Age Subjective healthb Number of diagnoses Psychosocial well beingc Education Socioeconomic position (self employed and farmers excluded)

Unadjusted effects

Adjusted for age and healtha

Men

Women

Men

Women

0.257*** −0.626*** −0.193** −0.113 0.673*** 0.277***

0.264*** −0.377*** −0.218*** −0.200*** 0.141 0.126*

−0.628*** −0.210*** −0.02 0.608*** 0.213***

−0.341*** −0.231** −0.166*** 0.302*** 0.154**

Sign. p = 0.01 ***, p = 0.05 **, p = 0.1 * a) For the health variables, only age is adjusted for. b) High values = poor health c) Summation index based on questions about whether or not the respondents were troubled with tiredness, sleeping problems, anxiety, unhappiness, loneliness, or feeling inferior. Low values on the index indicate a lower degree of problems.

characteristics, along with various aspects of working conditions. The dependent variable was the actual preferred pension age reported by the respondent. The presentation includes unadjusted OLS regression estimates and estimates adjusted for age and health. Table 2 shows that age, education, and social position were all positively associated with preferred pension age. The age effect might to a certain degree be explained as a selection effect, since many individuals with health problem would have already left the labour market. There is also a bias involved here, as people may have considered the preferred pension age in relation to their actual age. It comes as no big surprise that health had a strong effect on the preferred pension age. Irrespective of whether the indicator measures health subjectively or more objectively, as is the case with the variable for number of diagnoses, poor health decreases preferred pension age. This holds for both men and women, although there was no significant effect of psychosocial wellbeing among men. Both education and socioeconomic position were positively related to preferred pension age, indicating that work-related circumstances were of importance in this context. The next table presents a closer analysis of the associations between pension preferences and a number of indicators of work conditions. Both physical and psychosocial indicators were tested. The univariate estimates show that most of the indicators were associated with pension preferences in the expected direction; that is, those exposed to poor work environment conditions reported a lower preferred pension age. Some differences can be seen between men and women. Physical strain, control, and monotonous work were significant indicators for men, even after controlling for age and health. For women, indicators of social support also appeared to be of importance, while monotonous work and control were not significant indicators. Additionally, the indicator of subjective physical work was positively associated with preferred pension age among women. Somewhat surprisingly, neither quantitative nor qualitative work demand seemed to have any effect on preferred pension age. A possible explanation for this is that the group studied here, workers aged 55-64, represent a selected group that have learned to cope with the demands of their work. This interpretation is supported by the fact that in this study this specific age group, on average, report lower levels of work demand than do younger employees. Another explanation is that a high level of work demand might be either positive or negative in terms of the work environment (Karasek, and Theorell, 1990). The well-known demand-control model states that 149

64

Pension preference

63.5

63

62.5

62

61.5 High Medium

61 Low Medium Work demands

Figure 4.

Control

Low High

Preferred pension age in relation to different combinations of work demands and control.

high levels of work demand have a negative impact only in certain situations, and that the most disadvantageous work situation is created when high levels of quantitative demand are combined with low levels of work control. Such a combination of work characteristics tends to be associated with poor health and low levels of work satisfaction. On the other hand, a high level of demand in combination with a high degree of control tends to generate a mostly positive work situation. Figure 4 relates the influence of this demand/control interaction to preferred pension age. The results suggest that pension preferences were generally structured in accordance with the model. High demands and low control produced the lowest preferred pension age, while the highest pension age was found in the category with high demands and high control. In addition, in the high control category, increasing work demands were associated with increasing preferred pension age, while the opposite was true for the categories with low and medium levels of control. However, it should be noted that the differences between the categories in figure 4 were not significant. The model was not tested separately for men and women, due to low absolute numbers. So far, we can conclude that the work environment indicators showed the expected associations with pension preferences. In a more complete multivariate analysis, however, it was age, health, and socioeconomic position (only for men) that showed the strongest associations, while work environment indicators showed rather weak or non-significant effects. The variable with the most robust effect was health. However, this does not constitute a rejection of the influence of work environment factors, since such factors are often strongly correlated with both socioeconomic position and education; for instance physical strain at work is associated with social class in an almost linear manner, and work environment strain is a well-known predictor of health. The results may therefore be strongly affected by such interaction patterns and selection effects. Hence, in order to improve the analyses some longitudinal data were applied in analysing the influence of work conditions and health on pension preferences. Longitudinal data is often subject to ‘data poverty’, and to some extent this was the case here, since only a limited number of indicators were comparable over time. By comparing the answers on available indicators at baseline (T1) with the same indicators at T2, we constructed a number of variables showing whether or not respondents 150

Table 3. Associations between preferred pension age and different indicators of work conditions. Unstandardised regression coefficients. Sign. p = 0.01***, p = 0.05**, p = 0.1* Unadjusted effects

Adjusted for age and healtha

Indicators of work conditions

Men

Women

Men

Women

Physical strain (index) Quantitative demand (index) Qualitative demand (index) Monotonous work Hectic work Control (index) Instrumental support from colleagues Emotional support from colleagues Instrumental support from manager Emotional support from manager Getting along with superiors Work/family conflict Lonely work Physical work ability Psychological work ability

−0.109** −0.01 −0.05 −0.963*** −0.377*** 0.148*** −0.234 0.236 0.06 0.408** 0.319** 0.05 −0.300 0.08 0.04

−0.156*** 0.03 0.02 −0.604*** −0.348 0.02 0.321 0.456*** 0.338** 0.476 0.557*** −0.03 −0.157 0.207*** 0.113**

−0.110**

−0.154***

−0.742**

−0.342

0.119***

0.254 0.238

0.258 0.269* 0.270* 0.387***

0.210*** 0.06

had experienced any changes in the variable in question. Such data might tell us somewhat more about the influence of work conditions, since they allow us to control for (a) occasional exposure which in turn may affect how the question about pension preferences are answered, and (b) long term exposure, and changes over time in exposure to different work conditions. The results of this analysis are displayed in Table 4. Again, we can see a clear effect of health. At the univariate level there was an obvious and negative effect of physical strain at work. It is also interesting to note that changes in the work situation that imply increasing physical work demands significantly decreases the preffered pension age also when controlling for age and health. Those who reported monotonous work at both T1 and T2 were associated with a substantial decrease in preferred pension age. One important aspect when discussing older workers relates to the question of human capital, and the extent to which older employees have access to vocational training and skill development. The PSAE data contained one question related to this area; it asked whether the respondents had the opportunity to learn new things at work. It is interesting to note that we found a significant effect here. Work situations which lacked learning opportunities significantly lowered the preferred pension age, especially among those who had experienced the removal of such opportunities. Table 4 also shows that satisfaction with working hours was of substantial importance. Respondents reporting dissatisfaction with working hours also reported a lower preferred pension age; this was particularly the case among those who preferred shorter working hours both at T1 and T2. An important feature of several of these variables is that they interact with health and socioeconomic position in a multivariate setting. This concerns first and foremost deterioration of work quality among the respondents that reported bad subjective health, and among the lower social classes. For example, a combination of poor subjective health and increased physical workload resulted in a strongly significant decrease in preferred pension age. Experiences of increased monotonous work over time among working class employees also showed strong negative effects. These results suggest that working conditions per se, as well as how the work situation changes over time, are essential factors in elderly employees’ thoughts about the timing of retirement. 151

Table 4. Associations between changes/stability in health and work conditions and pension preferences. Unstandardised regression coefficients. Sign. p = 0.01***, p = 0.05**, p = 0.1* Variables Good subjective health T1→T2 No/No Yes/No No/Yes Yes/Yes Good psycho social wellbeing T1→T2 Yes/No No/Yes No/No Yes/Yes Physical workload T1→T2 High/High High/Low Low/High Low/Low Monotonous work Yes/Yes Yes/No No/Yes No/No Psychologically tiring work T1→T2 Yes/Yes Yes/No No/Yes No/No Learn new things at work T1→T2 Yes/No No/Yes No/No Yes/Yes Satisfied with working hours T1→T2 Satisfied/Fewer hours Fewer hours/Satisfied Fewer hours/Fewer hours More hours/Satisfied Satisfied/Satisfied

Crude estimates

Adjusted for age and health

−1.059*** −0.849*** −0.005 Ref. grp

−1.05*** −0.759***

−0.401 −0.153 −0.802*** Ref. grp −0.579** −0.475 −0.664* Ref. grp −1.335*** −0.964*** −0.647* Ref.grp

−0.801***

−0.469* −1.052***

−0.105 −0.032 −0.470* Ref. grp −0.639*** −0.324 −0.400* Ref. grp

−0.579**

−0.919*** −0.455 −1.328*** 0.660 Ref. grp

−0.812***

−0.412*

−1.153*** 0.672*

4 CONCLUDING REMARKS We may now ask what conclusions can be drawn from the analyses presented above. First, it can be concluded that in Sweden there is no wish for a working life lasting beyond the age of 65. On the contrary, the dominant preference is early retirement. This result might be interpreted in terms of an existing early exit culture. This impression is emphasised by the fact that the most common reason given for an early retirement preference was ‘more spare time’ and by the generally stable picture across different social groups. In other words, there was a widespread preference for an early exit based on priorities other than work values. Bearing this in mind, social policy reforms that aim to stimulate a longer working life seem to be adequate measures for influencing workers by means of economic incentives to stay in work longer. 152

At the same time, the analyses showed that a substantial proportion of respondents reported that their reasons for preferring early retirement included factors connected to working conditions and the work environment. In addition, work-related factors proved to be common reasons for a late exit preference. Together with the analysis of the associations between preferred pension age and various indicators of working conditions, this shows that the quality of work factors must not be ruled out as important predictors of people’s views of how long their working lives should be. In order to achieve the social policy objective – postponement of the average retirement age – it is insufficient to merely reconfigure, and restrict access to, economic incentive structures in early retirement programmes. Although to some extent this can be expected to affect both attitudes and overall employment activity rates of older workers, an equally important measure is to improve the quality of work, and to create a working life better in tune with the capacities and qualities of an ageing work force.

REFERENCES de Vroom, B., and Guillemard A-M., (2002). From Externalisation to Integration of Ageing Workers: Institutional changes at the end of the work life, in Andersen J.G., Jensen, P.H (ed) Changing labour markets, welfare policies and citizenship. The Policy Press Ebbinghaus, and Bernhard, (2000). Any Way Out of ‘Exit from work’? Reversing the Entrenched Pathways of Early Retirement in Scharpf Fritz W., Schmidt, Vivien A., ed. Welfare and work in the open economy. Diverse responses to common challenges, Vol II. Oxford, Oxford University Press Esser, and Ingrid, (2006). Why Work? Comparative Studies on Welfare Regimes and Individuals’ Work Orientations. Doktoravhandling. Institutet för social forskning. Gruber, Jonathan, and Wise David A., (1999). Social security and Retirement around the World. Narional Bureau of Economic Research. The university of Chicagor Press. Ilmarinen, J., (1999). Ageing workers in the European Union – Status and promotion of work ability and employment. Finnish Institute of Occupational Health, Ministry of Social Affairs and Health, Ministry of Labour. Helsinki. Karasek Robert, and Theorell Thöres, (1990). Healthy Work, Stress, productivity and the reconstruction of working Life, Basic Books, Inc., Publishers, New York Marklund Staffan, Bjurwald Mats, Hogstedt Christer, Palmer Edward, and Theorell Töres (red), (2005). Den höga sjukfrånvaron. Problem och lösningar. Arbetslivsinstitutet, Institutet för psykosocial medicin, Försäkringskassan. Statens folkhälsoinstitut. OECD, (2000). Reforms for an Ageing society. Social issues. OECD, (2003a). Ageing and employment policies. Sweden. OECD, (2003b). Policies for an ageing society: Recent measures and areas for further reform. Economic department working paper No. 369 OECD, (2003c). Transforming disability into ability. Policies to promote work andincome security for disabled people. OECD, (2006). Live Longer, Work Longer. Ageing and Employment Policies. OECD Publications RFV, (2001). Arbetsgivare attityder till äldre yrkesverksamma. Riksförsäkringsverket analyserar 2001:9 RFV, (2006). Genomsnittlig pensionsålder i de nordiska länderna. Försäkringskassan analyserar 2006:11 Soidre Tiuu, (2005). Vill vi jobba till – eller kanske längre? Arbetsförhållandens betydelse för äldre medelålders kvinnors och mäns preferenser. Arbetsmarknad och Arbetsliv. Årg 11. Nr 2: pp. 107–121 SOU, 2002:5, Handlingsplan för ökad hälsa i arbetslivet. Statens offentliga utredningar. SOU, 2002:29, Riv ålderstrappan! Livslopp i förändring. Diskussionsbetänkande av den parlamentariska äldreberedningen SENIOR 2005. SOU, 2003:91, Äldrepolitik för framtiden. 100 steg till trygghet och utveckling med en åldrande befolkning. Stattin Mikael, (1998). rke, yrkesförändring och utslagning från arbetsmarknaden. En studie av relationen mellan förtidspension och arbetsmarknadsförändring. Akademisk avhandling. Socioligiska Institutionen. Umeå universitet Torgén, Margareta, Stenlund, Carin, Ahlberg, Gunnel, Marklund, and Staffan, (2001). Ett hållbart arbetsliv för alla åldrar. Arbetslivsinstitutet. Wilensky, and Harlold, (2002). Rich Domocracies. Political Economy, Public Policy and Performance. University of California Press. Berkeley. Los Angeles

153

ANNEXE Questions included in the physical strain index Do you, in your daily work, experience: Repetitive movements, Awkward work positions, Sweaty work, Vibration, Heavy lifting? 0 1 2 3 4 5

No on all questions One “yes” answer to the included questions Two “yes” answers to the included questions Three “yes” answers to the included questions Four “yes” answers to the included questions Five “yes” answers to the included questions

Questions included in the quantitative demand index How often do you have enough time to get your work tasks done? How often do you have to work very hard? How often does your work require far too great an effort? Questions included in the qualitative demand index How often do you have keep a great deal of information in your head when you are performing your work tasks? How often do you feel that your work can never be performed well enough? How often do you find it hard to let go of thoughts about your work during your spare time? Choices of answer for the previous six questions were: 1 2 3 4 5

Almost every day At least once a weak A couple of times per month A couple of times per year Seldom or never

Questions included in the control index What possibilities do you have to decide how your work is done? What possibilities do you have to decide what you do in your work? What possibilities do you have to decide when to do your work? Choices of answer for the previous three questions were: 1 2 3 4

Large possibilities Fairly large possibilities Small possibilities No possibilities

154

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Younger managers and considerably older subordinates Ingrid Johansson Department of Social Sciences, Mälardalen University, Eskilstuna, Sweden

ABSTRACT: The purpose of this paper is to present some results of an interview study of younger managers and older subordinates in Sweden. An aim of the study was to increase the understanding of relations between younger managers and older subordinates. Data have been collected through semi-structured interviews with 23 younger managers who had experience of managing subordinates who were 20 years older or more. Furthermore, 19 older subordinates who had experience of being managed by managers who were at least 20 years younger were interviewed. Male and female interviewees have been selected from different kinds of organisations, lines of business and management levels. The results showed that many younger managers and older subordinates perceived that age generally did not matter but they recommended age-mixed teams because of different experience and knowledge. However, most of the interviewees lived an age-segregated private life and socialised mainly with people of their own age. A conclusion was that some of the younger managers’ insecurity in managing older people and a possible reluctance to employ them may be a result of the fact that some younger managers do not know older people well, that ‘like attracts like’ and that there is a general age segregation in society. Keywords:

Intergenerational relations, younger managers, older subordinates

1 INTRODUCTION A consequence of the demographic development and the growing number of older people in Sweden and other countries may be a shortage of personnel in some areas. One of several possible ways of solving this situation is to increase the number of older people in working life. For employers to retain and employ more older workers there is a need for an enhanced understanding and knowledge of different age groups, how they think, what they need and what motivates them (Goldberg, 2000). As Zemke, Raines and Filipczak (2000) point out, never before in our history have so many such different generations been expected to work side by side. Understanding generational differences is critical to creating harmony, mutual respect and joint effort. Still, according to Williams and Nussbaum (2001), research on different age groups in working life, for example the relation between younger managers and older subordinates, seems to be greatly neglected. It is an aim of the study presented here to fill some of this gap. A source of inspiration for studying younger managers was the American writer Betty Friedan, who died in 2006 on her 85th birthday. She wrote the following in her book The Fountain of Age (1994): But those ‘human resource managers’ in their forties simply do not see men and women over sixty as people like themselves in a new stage of development (Friedan, 1994, pp. 174). If younger managers see older people as very unlike themselves and fail to earn their trust, there is a risk that the younger managers may reinforce the trend towards older people’s early exit from the 155

labour market. Kanter (2004), who stresses the importance of trust and confidence in successful management, points out that leadership is not about the leader but about how he or she builds confidence in everybody else. Thus, lack of confidence and bad or inadequate management may either make older people redundant or make them want to leave the labour market on their own as soon as they can. Younger managers may feel uneasiness in their relations to older workers and a reason may be that older workers have for many years built up a certain position in the organisation and therefore are not easily impressed by younger managers (Schabracq, 1994). Older employees may also be critical of managers in general (Glover and Branine, 2001a). On the other hand, Glover and Branine (2001b) point out, older people can feel encouraged and very comfortable working as subordinates of younger managers whom they have reason to respect and appreciate. In short, I believe that it is fruitful to study younger managers as they may play a crucial role for older people’s willingness and motivation to stay longer in working life. 1.1 Aims of the study and research questions One aim of the study was to increase an understanding and knowledge of relations between younger managers and older subordinates from the perspective of individuals. Another aim was to describe and compare younger managers’ and older subordinates’ perceptions of age-related matters in the workplace. The main research questions of the study were: • What is the significance of a considerable age difference between younger managers and their subordinates? • What perceptions and experiences do younger managers and older subordinates have of agerelated matters in the workplace? 2 METHODS The purpose of the study was to obtain, not a set of fixed answers, but as many different and varied answers as possible. Therefore, a qualitative approach was chosen. The choice of qualitative methods was also due to the explorative character of the study. Furthermore, the choice of methods is related to the aim of collecting and analysing the subjective viewpoints of the studied phenomena (Flick, 1998). The research project belongs to the interpretive/constructivist paradigm, which means that the data have been interpreted from a hermeneutical perspective. 2.1 Data collection In 2002 and 2003 semi-structured interviews were carried out with 23 younger managers, aged 22–34, who had experience of managing subordinates who were 20 years older or more. 10 of the interviewees were men and 13 women. The interviewees had been managers for 0.2–9 years with an average of 3.7 years. The number of subordinates varied from four to 150. Only two of the interviewees worked in the same organisation. Fourteen worked in the private sector and nine in the public sector. Fourteen managers had graduated, four had started but not finished their graduate studies and five had an upper secondary school education. Furthermore, in 2006 and at the beginning of 2007, 19 interviews with older subordinates with experience of being managed by at least 20-year younger managers were carried out. 12 of the older interviewees were women and seven men. The interviewees were 50–63 years old with an average age of 60. Their younger managers were 30–40 years old. Eight of the interviewees worked in the private and 11 in the public sector. Six interviewees had graduated, 10 had an upper secondary school education and three had an elementary school education. Male and female interviewees have been selected from different kinds of organisations, lines of business and management levels. The age interval of 20 years or more was chosen in order 156

to get an age difference that was big enough to be associated with a parent-child relationship. The sample strategy was non-random and purposive and the goal was to find informationrich interviewees. The interviews were tape-recorded, transcribed word for word, analysed and interpreted. 2.2 Data analysis and interpretation During the analysing process, data were scrutinised for patterns, themes, and regularities as well as contrasts and irregularities (Coffey and Atkinson, 1996). However, the analysing process has just begun and there is still much more analysis and interpretation to be done. To interpret with a hermeneutical approach means to hold the view that a phenomenon can be looked upon from different aspects (Ödman, 1979). The hermeneutical approach to the interview texts meant that the intention was to explore possible interpretations and meanings rather than providing a definite meaning. 2.3 Ethical issues Ethical criteria compiled in The ethical recommendations of The Swedish Council for Research in the Humanities and Social Sciences (HSFR, 1990) have been followed. Thus, the interviewees were informed of the aims of the study, that participation was voluntary and could at any time be interrupted, that the material would be treated in a confidential way and that it would not be used for other purposes than research.

3 RESULTS Only some parts of the results of the study will be presented here, organised in two sections. One section is about the significance of a big age difference and the other one is about age-mixed teams at work and an age-segregated society. 3.1 The significance of a big age difference between managers and subordinates Generally, the older interviewees seemed to have a very positive experience of younger managers and many maintained that managers’ age and a big age difference were of little importance at work. Other things mattered more, for example the managers’ personality and characteristics such as humility and courage to ask for help when needed, readiness to listen and mutual respect. Many older interviewees also confirmed that their younger managers did listen to them. The younger managers were of the same opinion concerning the importance of listening. A male manager, aged 32, said that listening was the key to successful management. However, there were also examples of when and why age could matter. A female manager, aged 30, talked about younger and older people having different values: Age means that we have different values. Younger and older people have grown up in different kinds of societies so age can not be left out of account. This woman seemed to be referring to the so-called cohort effects, i.e. effects on people’s lives that have to do with the historical period during which they were born and grew up. A male manager, aged 28, expressed the view that it was easier to manage people who were his age. He said this: It may be easier for me to manage a younger person. Older people may have other views on how to handle things. A younger person is easier to tell to do anything, to use, maybe. He or she wants to do a good job in order to progress. 157

A 32-year-old manager explained that employing people of his own age could be nice but totally wrong if he did not consider what knowledge and experiences were needed: If I recruit people of my age we can make good friends and understand each other but it could completely ruin the whole purpose of my work. Instead I try to find people who fit into the picture in order to get a complete picture. Then I may need some older people with experience and some younger ones who are motivated and fast. When this man talked about the temptation to employ someone like himself, he seemed to present an example of the psychological phenomenon ‘like attracts like’. One of the disadvantages of older people perceived by the younger managers was their tiredness of changes. However, a male manager, aged 32, refused to generalise or accept stereotypes about older people and their supposed aversion to changes: Most people are afraid of new things or nervous about them but younger people are more reluctant to admit it and might handle it in a different way. Younger managers, on the other hand, were perceived to enjoy quick changes and sometimes they were simply too quick. Several younger managers talked about how older people could help them to slow down. Here is a quotation from a male manager, aged 35: Older people don’t get carried away and they may have a slowing-down effect on younger people and managers. Their questioning may be of help because if I cannot answer I have to think the matter over and consider whether it is correct or not. They think about things that younger managers don’t think of. The older interviewees were asked what age would be an ideal age for a manager. Half of them answered that there was no ideal age. Any age could be ideal, except for very low ages such as 20 or maybe 25. Most of those older people who did express an ideal age of a manager mentioned 40, 45 or 50. A couple of people argued that they preferred a manager who did not have very young children. A man, aged 63, said the following about his manager: I cannot say that he stays home with sick children very often but he has three children, so it happens that they infect each other. Then he calls and tells us that he is working at home. He is available then but planned meetings have to be cancelled. Generally, having sick children was not perceived as a very big problem, because today much work can be carried out at home. Nevertheless, it could be an inconvenience and it was the same for both male and female managers.

3.2 Age-mixed teams at work and an age-segregated society Both younger and older interviewees recommended age-mixed teams. A man, aged 62, said that you have to be careful with the mix and an older woman, aged 62, said: You take and you give and you have different knowledge and experience. That’s why it is important to have mixed ages. Older people’s experience was often mentioned as an advantage and a reason for the need for mixed age groups. However, when the older interviewees were asked about younger managers’ lack of experience they tended to tone down its significance. They seemed to believe 158

that lack of experience could easily be made up for in various ways if only the younger managers realised that they did not always know best and were open to ask for help when they needed. When the younger managers were asked if they had any private friends that were much older than themselves, most of them answered no. They did not spend time with older people except for relatives such as parents, parents-in-law and grandparents. Many stated that they socialised with people who were not more than 5–10 years older or a few years younger than themselves. The reason was that they had similar values and were at the same stage in life. So, here is another example that can be referred to cohort effects and the ‘like attracts like’ phenomenon. When the older interviewees were asked if they had younger private friends only five answered yes and 14 no. So, the situation was the same as with the younger managers. A young female manager, aged 33, concluded that private life is more age-segregated than working life. Another young female manager, aged 34, said that if you do not know any older people well, they may be like strangers to you. But there were exceptions. An older woman, aged 59, who socialised with younger people in her private life, said that she enjoyed communicating with them. A female manager, aged 34, said that talking to her older aunts or her parents’ friends meant that she got a new perspective on many things. Others had the experience that by talking to older people they could learn to value different things and that discussions with older people were more profound. A woman, aged 62, expressed the view that for younger people it is more difficult to mix with older people than for older people to mix with younger people. A 63-year-old man had a similar view. He declared that young people may find it more difficult to make contacts with older people than the opposite because to young people a certain age difference may seem bigger than the same age difference may seem to older people.

4 DISCUSSION The results show that generally people of different ages seem to work well together when there is mutual respect and when younger and older people listen to each other. However, in this short chapter it was only possible to present a few of many individual variations. A reason for younger and older people to get along well appeared to be the confidence that older interviewees seemed to have in their younger managers. According to Kanter (2004), confidence is based on reasonable expectations. The younger managers seemed to come up to older people’s expectations. A couple of contradictions were found in the material. For example, older interviewees’ low concern for younger managers’ lack of experience contrasts with the great value they ascribed to older people’s experience. Another contradiction was found when many interviewees denied or depreciated the significance of a big age difference and at the same time presented reasons why they recommended age-mixed groups and declared that they preferred to socialise with people of their own age.

5 CONCLUSIONS A conclusion was that having a much younger manager did not seem to have any big effect on older workers’ willingness to stay longer in working life. Another conclusion was that some of the younger managers’ insecurity in managing older people and a possible reluctance to employ them may be a result of the fact that some younger managers do not know older people well, that ‘like attracts like’ and that there is a general age segregation in society. 159

REFERENCES Coffey, A. and Atkinson, P., (1996). Making Sense of Qualitative Data, (London: Sage Publications). Flick, U., (1998). An Introduction to Qualitative Research, (London: Sage Publications). Friedan, B., (1994). The Fountain of Age, (London: Vintage). Glover, I. and Branine, M., (2001a). Introduction: the challenges of longer and healthier lives. In I. Glover and M. Branine (eds) Ageism in Work and Employment, (Aldershot: Ashgate), pp. 3–21. Glover, I. and Branine, M., (2001b). Do not go gentle into that good night: Some thoughts on paternalism, ageism, management and society. In I. Glover and M. Branine (eds) Ageism in Work and Employment, (Aldershot: Ashgate), pp. 47–64. Goldberg, B., (2000). Age Works: What Corporate America Must Do to Survive the Graying of the Workforce, (New York: The Free Press). HSFR (Humanistisk-samhällsvetenskapliga forskningsrådet/The Swedish Council for Research in the Humanities and Social Sciences), (1990). Forskningsetiska principer (Ethical recommendations), Stockholm. Kanter, R.M., (2004). Confidence: How winning streaks and losing streaks begin and end, (London: Random House). Ödman, P.J., (1979). Tolkning, förståelse, vetande: Hermeneutik i teori och praktik. (Interpretation: Understanding, Knowledge. Hermeneutics in Theory and Practice), (Stockholm: Almqvist & Wiksell). Schabracq, M.J., (1994). Motivational and cultural factors underlying dysfunctioning of older employees. In J. Snel and R. Cremer (eds) Work and Aging, (London: Taylor & Francis), pp. 235–249. Williams, A. and Nussbaum, J.F., (2001). Intergenerational Communication Across the Life Span, (Mahwah, NJ: Lawrence Erlbaum Associates). Zemke, R., Raines, C., and Filipczak, B, (2000). Generations at Work: Managing the Clash of Veterans, Boomers, Xers, and Nexters in Your Workplace, (New York: Amacom).

160

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Training older workers and long-term development: Needs and obstacles Jean-Claude Marquié & Liliana Rico Duarte University of Toulouse, CLLE-LTC (Cognition, Langues, Langage, Ergonomie – Laboratoire Travail & Cognition), UTM, EPHE, CNRS, Toulouse, France

ABSTRACT: This paper addresses the cognitive, meta-cognitive and emotional dimensions of some measures likely to promote: (i) greater access to training for older workers, (ii) training methods which are better suited to the adult’s characteristics, and (iii) the integration of continuous learning and long-term development opportunities in the job itself. The paper is illustrated by the authors’ recent and earlier research findings in the field.

1 INTRODUCTION The concept of sustainable development has met with great success in the fields of ecology and the alternative economy, but it has rarely been applied, so far, to human capital in the company. Doing so, however, would invite us to draw attention to an old – but still topical – fact. This is that, through their work, people not only transform their environment but also their own abilities, in a favourable or unfavourable way. If we want to improve and extend employability in the latter part of occupational life, then we must revisit the equally old utopia according to which how we produce something is as important as the product itself. This involves paying more attention than before to the way working conditions, including learning opportunities, can generate new resources, in several domains, including the cognitive domain. A company may thus generate profitable investments for itself and its workers by making them better qualified to take up future occupational challenges. Poor working conditions and inappropriate age-related training practices may have, on the other hand, an adverse effect on occupational behaviour and skills. Examining the literature in the field of psychology, neurosciences and ergonomics, as well as in other relevant fields such as management or education, allows us to identify major obstacles in terms of learning opportunities throughout occupational life and training methods, especially for older workers. We can thus specify research needs and measures likely to overcome some of these difficulties. This paper addresses the cognitive, meta-cognitive and emotional dimensions of some measures likely to promote: (i) greater access to training for older workers, (ii) training methods which are better suited to the adult’s characteristics, and (iii) the integration of continuous learning and long-term development opportunities in the job itself. The paper will be illustrated by some of the authors’ recent and earlier research findings in the field. 2 PROMOTING GREATER ACCESS TO OCCUPATIONAL TRAINING FOR OLDER WORKERS In France, as in other countries, access to occupational training becomes more difficult with advancing age. This fact is old but still relevant (Fournier, 2004). Trying to change it has several implications. It first requires changing representations and throwing off the yoke of negative 161

stereotypes, where they exist, both in employers and in employees themselves (Taylor and Walker, 1994). Surveys have repeatedly shown that occupational training willingness is one of the most frequently-cited occupational qualities among those assumed to decline with age. There is now a broader social and political consensus than before about the need to increase the frequency and regularity of occupational training opportunities throughout the occupational lifetime (national and European programmes for “life-long learning”). Many psychological and social benefits are expected. First, this is a condition for maintaining the worker’s learning motivation because, as shown by extensive surveys on continuing training carried out in France (Fournier, 2004), the less training the worker has received, the less he/she is willing to be trained, and vice-versa. Second, increasing the frequency of training is a means of maintaining learning skills. Learning involves specific skills, which can be lost if they are not used. In a field study on age differences in learning railway maintenance skills, we found a significant negative relationship between the time elapsed since the worker’s last training experience and the learning performance: the more distant the previous training experience, the poorer the performance (Delgoulet and Marquié, 2002). This was even truer for the most abstract part of the tests that assessed learning performance. This suggests that skills involved in generalisation and abstraction are those most affected by disuse. A related issue concerns the relationship between training and overspecialisation at work. Even in highly-qualified working populations, overspecialisation involves several risks with regard to personal and occupational development. For instance, it has been found that compared to younger ones, older pilots from 25 French and foreign airline companies being trained to fly the Airbus A320 on a flight simulator in Toulouse (Amalberti, Pèlegrin, and Racca, 1991) exhibited slower progression rates (as assessed by the number of additional training sessions requested), and greater failure rates in the final examination. One interesting result of the study was that the A320 training seems to have been easier for pilots who had moved around in the previous few years. Indeed a positive correlation was found between failure rates in the final examination and time spent on the most recently piloted aircraft model. Note that more frequent moving around between different aircraft models also implies more frequent training opportunities. A third expected benefit of the increased frequency of the workers’ training is that it would allow them to compensate for any initial inequalities in education. Indeed, people with lower academic credentials are also those with the lowest rates of access to training, in both the public and private sectors (Perez, 2002). Occupational training policies thus maintain initial disparities and may even increase them because of some cumulative processes at work. Fourth, regular training was reported in earlier work such as that of Davies & Sparrow (1985), as counteracting seniority-related obsolescence and preserving performance levels at a relatively greater extent in the older workers.

3 PROMOTING TRAINING METHODS BETTER SUITED TO THE OLDER WORKER’S CHARACTERISTICS Once obstacles preventing senior workers from accessing formal training courses have been overcome, older learners may still encounter specific difficulties in these situations. From studies conducted in real-world training situations and experimental studies carried out in the laboratory, we can extract some recommendations or research directions whenever our knowledge is too scarce. One important measure for older learners consists in making errors less dramatic for them during the learning process, by convincing them that making errors is necessary to progress. Adopting such a view is less easy for them, as we observed in a real training session on learning how to use a word processor (Baracat, and Marquié, 1994). We applied techniques derived from the Signal Detection Theory to assess the decision criteria used by young and older trainees, when judging and deciding whether procedures that were submitted to them, and which they were expected to execute in a subsequent phase, were right or wrong. The older trainees (age range of participants: 25–55) exhibited stricter decision criteria, especially at the beginning of the training week. This means that they made a higher proportion of omission errors, which revealed a greater reluctance to commit 162

to an action when they were not entirely sure that the proposed procedure was correct. This attitude was interpreted as reflecting a higher level of anxiety in this learning situation compared to the younger trainees, an interpretation whose plausibility was confirmed in subsequent studies. This was observed, for instance, in a four-week training course on accounting and office automation (Delgoulet, Marquié, and Escribe, 1997) where the older trainees showed significantly higher anxiety throughout the first three weeks, with learning anxiety being assessed by the Spielberger state anxiety test. The same was observed in a one-week training course for railway operators, who were trained to use magnetoscopic techniques for maintaining bogie trucks (Delgoulet, and Marquié, 2002). In this situation, learning anxiety, measured by the same test at the beginning of the course, was also found to be significantly higher in older participants (r = .45, p ≤ .01). There are domains where age is associated with negative stereotypes. In these cases, trainers are therefore required to use compensatory measures to increase or consolidate the older trainee’s selfefficacy beliefs. New technologies may be an instance of such a domain, as shown by the results we obtained in experiments using feeling-of-knowing (FOK) tasks. The FOK is a prospective judgment of one’s ability to retrieve a particular piece of information that cannot be brought to consciousness at that moment. Participants are submitted first to a knowledge test including a great number of questions. For questions not immediately answered, they are asked to assess, by rating their FOK on a 10-point scale, whether they would be able to retrieve the right answer if they were given some hints and more time. In a subsequent experimental phase, they are submitted to a multiple-choice memory recognition task. This makes it possible to assess the accuracy of their prospective judgement, and to establish whether participants underestimate or overestimate their actual knowledge. For two semantic knowledge domains, general and computer-related, FOK ratings were recorded in young and older adults, after test difficulty was equated across age groups. The older adults were found to be less confident than the young ones when rating their FOK, but only for computer-related items, not for general knowledge items. Thus, although their actual knowledge in the computer domain (as in the general domain) was exactly the same as the young participants, the older participants significantly underestimated their knowledge only in the domain where negative age-related stereotypes are present (Marquié, and Huet, 2000; Marquié, JourdanBoddaert, and Huet, 2001). We believe that this type of judgement bias may have deleterious consequences in learning situations, because people do not make an appropriate effort to mobilise the relevant cognitive resources and do not persevere as much if they do not believe that they possess these cognitive resources. Learning anxiety may also come from the above-mentioned lack of practice in training settings, from previous unsuccessful training experiences, or from a generational complex, with older people comparing themselves to younger ones who are supposedly or actually better prepared. Training methods little suited to the older trainee’s specific characteristics in terms of time pressure, mental overload, especially at the beginning of a training programme, lack of connection with the worker’s own experience, etc, are also a cause of lack of motivation for these populations.

4 PROMOTING THE INTEGRATION OF CONTINUOUS LEARNING AND SUSTAINABLE DEVELOPMENT OPPORTUNITIES IN THE JOB ITSELF Working and learning are too often opposed to each other. Narrowing the links between the two, in the many aspects of the relationship, is a major way of combating qualification obsolescence, and of maintaining and even increasing the worker’s cognitive and motivational potential. This implies, first, a work organisation that actually makes room for formally programmed training sessions, such as those devoted to learning new tasks in companies where job rotation is highly encouraged. In organisations under pressure, these in-service training sessions are often the first to be abandoned when time constraints are too great, as reported in field studies by ergonomists (e.g., Gaudard, 2000). This also implies that the work organisation provides workers with opportunities to experience more or less structured thinking and social exchange about their job practices and working conditions. We have learnt from cognitive psychology that competence does not progress 163

through piling up knowledge, but rather through the close integration of new knowledge within existing knowledge. This is an active, deliberate and effortful process. This also requires a metacognitive effort that makes procedural knowledge, which is encapsulated and clearly unconscious, become declarative, namely accessible to consciousness and thus modifiable. Such opportunities are insufficiently provided in the workplace. Moreover, some current work organisation trends, such as task individualisation, increasing time pressure, and tracking down and chasing idle time lead to the elimination of several of the informal but quite necessary occasions when the worker can perform this “unproceduralisation”. Doing such a meta-cognitive job increases awareness of all the accumulated skills. It allows workers to re-appropriate their skills, to improve or consolidate their self-confidence and thus their ability to take up new occupational challenges. Moreover, current literature on the psychology of ageing and work provides more and more evidence that formal and informal learning occasions in and out of work may also contribute to protecting the worker from cognitive ageing, to some extent (Schooler, Mulatu, and Oates, 1999). In a recent study we examined this hypothesis by using data from the VISAT (ageing, health and work) longitudinal study (Marquié, Jansou, and Baracat, et al., 2002). Data concerned 2,288 workers who were seen twice in a five-year interval and were between 32 and 62 years old at baseline. Cognitive stimulation received at work was assessed on the first measurement occasion through a score computed from seven items, including (i) occupational training opportunities, (ii) qualifying aspects of the job, with items referring to the cognitive richness of the work content and to whether the job allowed workers to increase their abilities (e.g., “My job enables me to learn new things”), and (iii) cognitive effort, with items reflecting a more intensive aspect of the work (e.g., “Having to hold a lot of information in memory at the same time”). Cognitive efficiency was assessed at baseline and follow-up through episodic verbal memory and processing speed tests. Results showed that greater mental stimulation received at work was associated with higher levels of cognitive functioning both at baseline and at follow-up (five years later), and with a lesser decline between baseline and follow-up in the two speed-based tests. These results were obtained after adjustment for age, education, sex, occupational status, intellectual ability, and a variety of other possible confounders. The study thus supports the hypothesis that exposure to jobs that are mentally demanding but offer learning opportunities increases the level of cognitive functioning and possibly attenuates age-related decline (Marquié, 2006; Marquié, and Ansiau, 2007).

5 CONCLUSION Earlier work in the field has already yielded useful information for improving access to training and more suitable training methods for older workers (e.g., see E. Belbin, and Belbin, 1972; Kubeck, Delp, Haslett, and McDaniel, 1996; Paumes, and Marquié, 1998; Czaja, Sharit, Charness, Fisk, and Rogers, 2001). But on most of the issues outlined in this paper, much research effort must still be accomplished to better understand measures most likely to develop the worker’s technical and non-technical skills throughout his/her entire occupational life. One major way of achieving this goal is to favour the “learning organisation” (Sen, 1999), where working and learning are no longer opposed, and where the continuous development of human capabilities is seen as a profitable long-term investment both for the worker and the company (Bryson, and Merritt, 2007). REFERENCES Amalberti, R., Pèlegrin, C., and Racca, E., (1991). Cosynus: a new data acquisition system for aiding pilot training on modern aircraft. Proceedings of the XIXth Conference of the Western European Association for Aviation Psychology (WEAAP), Nice, France Baracat, B., and Marquié, J. C., (1994). Training middle-aged for new computer technology: a pilot study using SDT in a real-life word-processing learning situation. In J. Snel, & R. Cremer (Eds.), Work and aging: a European prospective, Amsterdam: Taylor & Francis, pp. 197–211 Belbin, E., and Belbin, R. M., (1972). Problems in adult retraining. London: Heinemann

164

Bryson, J., and Merritt, K., (2007). Le travail et le développement des capacities. Formation Emploi, 98: pp. 41–54 Czaja, S. J., Sharit, J., Charness, N., Fisk, A. D., and Rogers, W. A., (2001). The Center for Research and Education on Aging and Technology Enhancement (CREATE): A Program to Enhance Technology for Older Adults. Gerontechnology, 1: pp. 50–59 Davies, D. R., and Sparrow, P. R., (1985). Age and work behaviour. In N. Charness (Ed.). Aging and human performance, Chichester: John Wiley & Sons, Ltd., pp. 293–332 Delgoulet, C., and Marquié, J. C., (2002), Age differences in learning maintenance skills: a field study. Experimental Aging Research, 28: pp. 25–37 Delgoulet, C., Marquié, J. C., and Escribe, C., (1997). Training older workers: relationships between age, other trainee characteristics, and learning anxiety. Arbete & Hälsa, 97: pp. 70–78 Fournier, C., (2004). Continuing training in the private sector put to the test of age. Training & Employment, (CEREQ), 54: pp. 1–4 Gaudart, C., (2000). Conditions for maintaining ageing operators at work – a case study conducted at an automobile manufacturing plant, Applied Ergonomics, 31: pp. 453–462 Kubeck, J. E., Delp, N. D., Haslett, T. K., and McDaniel, M. A., (1996). Does job-related training performance decline with age? Psychology and Aging, 11: pp. 92–107 Marquié, J. C., (2006). Le travail, facteur de développement cognitif ou d’usure prématurée? Retraite & Société, 49, pp. 180–187 Marquié, J. C., and Ansiau, D., (2007). Âge et travail : les conditions du développement durable des compétences. In C. Saint Martin, C. Thébault, & G. De Terssac (Eds.). Travail, Organisation, Santé, Précarité, Toulouse: Éditions Octarès (in press) Marquié, J. C., and Huet, N., (2000). Age differences in feeling-of-knowing and confidence judgments in two semantic knowledge domains. Psychology and Aging, 15(3): pp. 451–461 Marquié, J. C., Jansou, P., Baracat, B., Martinaud, C., Gonon, O., Niezborala, M., Ruidavets, J. B., Fonds, H., and Esquirol, Y., (2002). The VISAT (ageing, health, & work) prospective study: overview and methodology. Le Travail Humain, 65, pp. 243–260 Marquié, J. C., Jourdan-Boddaert, L., and Huet, N., (2002). Do older adults underestimate their actual computer knowledge? Behaviour & Information Technology, 21(4): pp. 273–280 Paumès, D., and Marquié, J. C., (1998). Ageing workers, learning and job training. In J. C. Marquié, D. Paumès, and S. Volkoff (Eds.), Working with age, London: Taylor & Francis, pp. 297–311 Perez, C., (2002). Continuing training in the public service. Training & Employment (CEREQ), 48, pp. 1–4 Schooler, C., Mulatu, M. S., and Oates, G., (1999). The continuing effects of substantively complex work on the intellectual functioning of older workers. Psychology and Aging, 14, pp. 483–506 Sen, A., (1999). Development as freedom, Oxford: Oxford University Press Taylor, P. E., and Walker, A., (1994). The Ageing Workforce: employers’ attitudes towards older people. Work, Employment & Society, 8, pp. 569–591

165

Chapter 4 Promotion of Health and Work Ability

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Promoting health and workability in Vattenfall AB Nordic, Sweden Reidar J. Mykletun & Trude Furunes Norwegian School of Hotel Management, University of Stavanger, Stavanger, Norway

ABSTRACT: Vattenfall AB Nordic experienced dropping average retirement ages. Facing an increase in average ages for the workforce, increased expenses and shortage of competence and workers were foreseen. Retaining senior workforce, promoting workability and transferring competence to facilitate the generation relays became an organisational challenge. The General Manager initiated the Ageing Workforce Management Program by increasing pension age to 65 years, and gradually including other programs, among which seminars raising awareness and motivating for prolonged careers, a special program (80-90-100) reducing the workload of workers aged 58+, and age management leadership training were most central. Based on examinations of a questionnaire study, interviews, statistics on work environment, sick leaves and retirement from the company files, it was possible to conclude that the program was installed but has not yet penetrated the organisation. The 80-90-100 schedule was well received and average pension ages increased from 58 to 62.5 years over six years. Senior workers reported generally better work environment profiles than their younger colleagues, and sick leave rates were lower for those in the 80-90-100 program compared to the company average.

1 INTRODUCTION While age discrimination at the workplace has been documented in several studies (Furunes, and Solem et al., 2007) fewer researchers have documented age management interventions at the workplace aiming at retaining mature workers in the organisation. This latter approach was made the scope of the present study with focus on the ongoing workforce retention and generation relays programs in the Swedish company Vattenfall Nordic AB. Workforce retention means here the active efforts of the company to prevent early retirement of the workforce. Generation relay refers to the handover of competence and values from the experienced generations to the newly recruited and younger workers. The research problems driving this study were fourfold: To what extent did the organisation implement an age management program and how was it received by the participants? To what extent did changes in retirement ages occur and to which human costs for the senior workforce? Vattenfall constituted an adequate case for the study as they multiplied their workforce from the 1960s to 1990s reaching 8,000 today and, being highly esteemed as an employer workforce turnover has been low. Consequently, the average age of employees and managers was high and increasing (median 46 years; one third of the employees being between 50 and 60 years of age), reflecting the history of the company as well as the general demographic changes in the Nordic region today. Only 5.7 percent of staff, however, was older than 60 years, and the average pension ages were dropping and reached its lowest of 58 years by the turn of the millennium. Altogether 2,014 employees were entitled to retire from 1998 to 2014, among whom 1,510 would retire between years 2007 and 2014. Most of these would be difficult to replace as the company found it increasingly harder to recruit young and competent workforce, especially to outdoor maintenance and construction work in remote and thinly populated areas. Facing these threats to company efficiency, retaining senior workforce and promoting their workability along with transferring competence to facilitate the generation relays became central 169

organisational challenges. On April 24th 2001 the General Manager announced the expectation that 65 years of age should be the standard retirement age with rights to pension in the company, with no exceptions but health issues and special agreements previously made for outdoor linemen, thereby initiating what became the Vattenfall Ageing Workforce Management Program. The program was communicated to the company in pamphlets like “Myths and Facts about Older Workers (50+)”, and its core messages contained expressions like (p.2): “Early retirements a waste of resources … Vattenfall’s goal is to offer our employees such attractive working conditions that more can and want to remain longer in the company …To succeed, what is required … is that we all …have good knowledge about the connection between age and working ability … It is important to declare that there isn’t any clash of interests between younger and older in the workplace …The coming generation shift, to the contrary, presumes that we develop a way of working that makes young as well as old experience Vattenfall as an attractive employer …We must to a greater degree see to it that our younger people work together with more experienced employees”. Furthermore, the pamphlet aspired at replacing myths about elderly workers with facts in areas like declining physical strength being met with compensating skills, lack of relationships between age and work performance, flexibility and receptivity to change as a lifelong personal capability, criticality versus self-sufficiency and less need of leadership, development of loyalty and social skills, and retention of learning capacities. Clearly, as the program departed from organisational needs of the company to retain the workforce and critical competence, it was grounded within an economic rationality. Moreover, it became to some extent inspired by the extensive work of Juhani Ilmarinen, and especially the optimistic philosophies, strategies and tactics employed to promote health, work-ability and well-being of ageing workers as well as experiences from the FinnAge Respect for the Ageing Program in Finland (Ilmarnen, and Louhevaara, 1999; Ilmarinen, and Tuomi, et al., 1997). Opposite to the Finish project, Vattenfall did not employ the Workability Index (Ilmarinen, and Rantanen, 1999) and the occupational health service was not a central part in the project. One basic assumption was that prolonging working careers were advantageous for the individual, the organisation and the society as outlined by Ilmarinen (2003) in his “orientation matrix”. For the individual the benefits should be improved working ability, health and well-being, add meaning and structure to life, and providing social relationships and the feeling of making valuable contributions to the company. The company should benefit from retention of critical competence and labour, reduced costs on training new staff and early retirement pensions, and fewer needs for supervision. For the society advantages were improved health and well-being of citizens, and reduced costs on pension, health and welfare. Likewise, using the expression of “Age management” Ilmarinen (2003) pointed to remedies and solutions on all three levels. For the individual participating, coping with changes and developing competence, and improving health and personal resources were the imperative. The company could contribute by tailoring solutions like demands and working hours to individual capacities, establish age-mixed teams and provide adequate training. For the society, the central tasks were attitude change campaigns and developments of policies preventing all kinds of age-related discrimination. The actual materialisation of the Vattenfall Age Management Program that should support the new internal retirement policy was developed from 2003 and onwards through several steps and included: (1) One day seminars for workforce aged 57+ in order to raise awareness of the ageing and competence issues and motivating for prolonged careers. (2) A special program (80-90-100) reducing the workload for selected workers aged 58+. (3) Leadership training for age and health management. (4) A mentoring program for transferring competence from the senior workforce to the younger generations. (5) A “dialogue seminar” to discover participants’ tacit knowledge structure and share them across generations. (6) The “Senior Resource Pool” – a special structure for internal re-employment of redundant senior workforce. (7) Networking for managers and workforce on elderly workers issues with the Finish PWD-Technical Division in Helsinki. (No 4-7 are not discussed in this paper due to their small scale trials so far.) The choice of strategies was influenced by Human Relations traditions with emphasise on training and welfare. Central were also the well-established approach of fitting the workplace to the elderly worker. Finally, the program acknowledge that also senior competence has a market place where 170

the company must compete, and mainly with early retirement (Mykletun, and Furunes et al., 2006) in times of labour shortages (Henkens, and Remery et al., 2005). In sum, the program focus on both individual and organisational attributes as recommended by Ilmarinen, Huuhtanen et al. (1999) and thus constitutes a relevant example of a composite approach within the wider field of managing elderly workforce.

2 METHOD The research draws on interviews, questionnaires and company internal statistics, thus applying a multi-method approach (Huberman, and Miles, 1994). The interviews leaned on semi-structured interview guides. For the 57+ seminars 15 participants and five managers were interviewed, and central topics were their reactions to the seminar and how it influenced their retirement plans. For the 80-90-100 schedule, 17 participants and 16 managers were interviewed, and functional aspects along with individual and group outcomes were focused. Notes were taken during the interviews and the analyses followed coding processes as proposed by Huberman and Miles (1994) and Neuman (2003). The first analyses were open coding to identify information relevant to the a priori set themes discussed in the Introduction. However, due to the explorative aspects of the study, analyses of emerging themes were of equal importance. Second, the researchers used axial coding to examine the initial codes followed by selective coding of data and codes to compare and contrast findings and underpin the concepts and theoretical structures that finally led to the conclusions. An electronic questionnaire study was mailed to a total of 720 randomly selected managers in the company and reached 49 percent response rates. Only questions probing for managers’ decision latitudes regarding senior policies (Table 2, see Furunes, and Mykletun et al., 2006), their beliefs about senior workforce, and their observations of discriminatory practices were displayed here (Tables 3 and 4), using frequency analyses. Finally, selected questions from the company’s annual working environment study “My Opinion” were applied to assess a possible wear and tear on the senior workers (Table 5). Statistics for sick leave (Table 1) and retirement ages (Figure 1) were also added.

Table 1.

Sick leave in percent of working hours for participants in the 80-90-100 schedule compared to Vattenfall total sick leaves for years 2004–2006.

Groups

2004

2005

2006

Workforce on 80-90-100 schedule Vattenfall Sweden total

4.3 4

2.7 4.2

3.1 3.5

Table 2.

Managers’ perceptions of own decision latitude for senior policy decision making.

Dimensions of managers decision latitudes

Strongly agree

Agree

Disagree

Strongly disagree

Support from subordinates Possibilities to organise Human resources Budget resources Coin-sides with increasing demands

11 13 8 10 2

54 47 37 36 18

32 33 40 38 44

3 7 16 17 37

Percent (n = 357)

171

3 RESULTS The 57+ seminars – The Human Resource (HR) invited all workers aged 57+ to participate in one-day seminars by using the management hierarchy. Our interviews showed, however, that the invitation was filtered out as “uninteresting” by some managers, and 740 of the potential group of 1,200 participated at the seminars. The participants were generally satisfied with the seminars, which installed a feeling of being perceived as needed by the company due to their competence and labour capacities: “They need us and want us”. Contributing to smoothening of the generation relays through knowledge transfer became of significant concern to the participants and acted as a driving force. Several participants decided to extend their careers to 65, whereas previously planning to retire at 62 or earlier. For some participants the 80/90/100 schedule (see below) became a prerequisite for prolonging their careers, as it was believed to support good health and working ability, which was needed to continue at work. Access to learning and development of wages in senior years were seen as positive attributes, and work content and satisfaction were important for their determination to go on working to 65. Moreover, the employees thought that managers should have taken part in the seminars as they claimed that managers’ attitudes made the difference. After their instalment and first wave the 57+ seminars were handed over to the Vattenfall Professional Training, which is the company’s course and training group. They charged end-users for participation fees and the demand for the seminars evaporated. Adapting workload to senior workforce capacities: the 80-90-100 schedule – Retaining workforce by adapting work to the individual worker’s capacity was the rationale behind the 80-90-100 schedule, which allowed the individual to work 80 percent of his/her position while receiving 90 percent of salary and earning 100 percent of pension points. The unit should cover up for the loss of workforce if needed, or work had to be slowed down, as no compensations were given from the central organisation to replace the losses of working hours at the team level. The schedule was offered to workers aged 58+ and could be initiated on their own request, independently of their health status. The line managers were empowered to accept or reject such applications, and the decisive argument was the actual demands for labour at the workplace. The agreement was made between the individual and his/her manager, and should be evaluated and renegotiated after six months to be prolonged for another half-year period. The interviews informed us also that managers appreciated the schedule as it led to higher motivated and engaged workforce with more ideas for problem solving and more capacity to pursue their undertakings. 260 workers from a potential group of 900 have been into this schedule once, for several periods, or “on and off”, as the schedule depends on contextual factors and allows for flexible use both for the worker and the manager. All participating workforces were satisfied with the 80/90/100 working schedules and gave several reasons for this. It also allowed preparing oneself for retirement practically, socially and emotionally. All interviewed felt more motivated and energised for, and alert at work, and would continue working to 65 provided a continuation of the schedule. The schedule was perceived and valued as recognition for seniority and unique competence at the workplace. Some felt uncertain about the continuation as the schedule should be re-negotiated every 6th month. The schedule gave one day less per week on the road for long distance commuters and facilitated continuation of work with chronic health problems. The sick leave rates for the group on 80-90-100 schedule were lower than the average sick leave rates for Vattenfall total (Table 1), implying that workforce on this schedule fare better than the average workers of Vattenfall with respect to sick leaves even though they are among the senior workers. The schedule was tested in two business units in 2003 and later extended to the entire organisation. However, it did not materialise in all units as their managers claimed it to be incompatible with the work demands and available resources. If adapted by the managers it became practised with different criteria by different managers as some “sub-interpreted” the rules. Some managers felt uneasy about making these decisions, and some felt it added to their administrative burden. There were also reports on challenges related to managers’ decision latitudes, thus managers felt they were not fully empowered to put the schedule into operation. These findings were supported by results from the questionnaire study (Table 2). Limitations to managers’ decision latitudes were 172

Table 3.

Managers’ beliefs about senior workforce capacities.

Managers beliefs

Strongly agree

Agree

Disagree

Strongly disagree

Passive in change Negative to change Less flexible Uninterested in training

1 3 3 4

25 25 34 38

51 57 49 44

22 16 14 14

(Percent) (n = 357) Table 4.

Managers’ perceptions of discriminatory practices. (Percent) (n = 357).

Discriminatory practices against senior workforce

Strongly agree

Agree

Disagree

Strongly disagree

No promotion Less training No new technology Less wage increase

8 4 5 20

38 18 47 51

45 53 39 24

10 26 9 6

most pronounced with regards to combining the 80-90-100 schedule with increasing demands on unit or team efficiency, and most managers also saw limitations in relation to available budget and human resources, but more compatible with the attitudes of the workforce and that organising work according to the schedule was not a great difficulty. However it should be noticed that one third of the managers or more found difficulties in complying with the age management requirements. Leadership training for age and health management – A special leadership training approach was designed and run by external consultants to support the Ageing Workforce Management Program, and they also documented their approach (Skoglund, and Skoglund, 2007). Issues focused included empowering the workforce and their managers, improving the workplace dialogue and use of research based insights in optimising working conditions for all, including elderly workers. The program was only offered to 100 managers, but was well received. However, some managers in Vattenfall still held negative beliefs about the capacities of senior workforce as measured by the questionnaire (Table 3). The managerial scepticism applied mainly to low interest in training and age-related loss of flexibility, while one of four managers also kept the beliefs that the senior workforces were passive or even negative to changes in the work context. Moreover, some discriminatory practices seemed to prevail; as judged by the managers’ responses, this applied foremost to lower increase in wages for the senior workforce and to their exclusion when new technologies were to be introduced. No further promotion and less access to training for the senior segment were also reported (Table 3). Unexpectedly, the Ageing Workforce Management Program got external media support as it was an innovation in Sweden, and Vattenfall became the Employer of the Year in 2006. This fed back and boosted the organisational culture and self-concept. Changes in retirement behaviour – to which costs for the workforce? – From 2000 to 2006 the average retirement ages increased by more than four years from 58 to between 62 and 63, with total retirement ages being slightly lower than retirement ages excluding retirement on disability/longterm sick leaves (Figure 1). The increases in the former measure, which include the sick and disabled, is the most conservative evaluation of the actual retirement behaviour. It should be noticed tough that in spite of the findings on negative beliefs about the capabilities of elderly workers, the majority of the workforce aged 51+ rated their relationships to their managers as positive, and even more so than their younger colleagues (Table 5). This applied also to their relationship to colleagues, and to issues like finding meaning and satisfaction in their work, access to 173

Average ages at retirement

65 64 63 62 Total

61

Excl disability pensions

60 59 58 1998 1999 2000 2001 2002 2003 2004 2005 2006

Figure 1.

Table 5.

Changes in average retirement ages for employees and managers from 1998 to 2006.

Selected work environment dimensions from “My Opinion” presented by Yes (positive) versus No (negative) responses for age groups >30; 31–50; and 51+ for years 2005 and 2007. (N = 7,072 for 2005, and 7,801 for 2007 populations). (The “uncertain” responders are not included in the table). Yes

No

Selected work environment dimensions from “My Opinion”

Year

>30

31–50

51+

>30

31–50

51+

My manager is actively interested in my personal development

2005 2007

66 65

68 71

68 68

26 27

24 22

22 20

Manager’s feedback helps me improving my achievements

2005 2007

50 50

57 63

69 64

39 35

34 27

31 24

I am often feeling work pressure

2005 2007

53 52

40 38

40 40

45 44

58 58

55 55

Work interferes with my private life

2005 2007

70 67

55 57

55 56

27 29

42 38

40 37

I have access to relevant training in my work

2005 2007

73 73

74 76

77 78

23 20

22 19

16 15

My colleagues go out of their way to help each others

2005 2007

88 90

88 89

88 89

11 7

10 8

8 7

My work gives me personal satisfaction

2005 2007

77 78

83 83

86 84

21 35

15 27

12 30

I have no intention to leave Vattenfall

2005 2007

20 23

13 19

8 13

59 57

69 61

79 73

relevant training, and experiencing pressure from work and imbalance between work and family life. Also they reported finding satisfaction in their work to a higher extent than their younger colleagues.

4 CONCLUSIONS Vattenfall implemented their Ageing Workforce Management Program in April 2001 and intensified it with new activities from 2003. Increases in average retirement ages have been observed accordingly, and sick leave rates were lower for the 80-90-100 schedule group than for Vattenfall total. Workers aged 51+ held as positive views on their work, the work environment and their interactions with managers as their younger colleagues. The program was well received by the 174

workforce and was generally accepted by managers. Positive external media support enhanced its direct effects, and the company was declared “Employer of the year” in 2006. However, the program had not fully penetrated the organisation in 2007. Some managers and units found it hard to implement, and a significant portion of managers reported lack of decision latitude for implementing age management strategies. A substantial minority of managers still held negative views on mature workers’ capabilities, and indication of age discrimination practices were reported. The competence transfer has not been accelerated and time may run out as the increase in number of retirements is sharp. 100 managers only have been in the special leadership training for Ageing Workforce Management. Thus the case study inform us that implementing an Age Management program trough a management structure is time consuming and requires several integrated approaches (Ilmarinen, J., and Huuhtanen, P. et al., 1999). After all, leadership is a complex process materialised through several organisational and behavioural channels like: delegation, written procedures, organisational cultures and face-to-face interaction (Yukl, 2006), and a significant force of worker’s self-management (Sims, and Lorenzi, 1992) should be accounted for in retirement decisions. The moderate penetration of the program may be traced back to weaknesses in or inconsistencies between these systems, where one may counteract another. Persistency in program continuation may lead to a better organisational penetration in near future, and the total leadership systems may profit from examinations with regards to the implementation of the Ageing Workforce management Program. At present the program may become an important tool to prolong workforce careers while also contributing to a strong external image for Vattenfall. REFERENCES Furunes, T., Mykletun, R. J., and Solem, P. E., (2006). Age Management in the Public Sector: Managers’ Decision Latitude. (Subm). Furunes, T., Solem, P. E. and Mykletun, R. J., (2007). Age Discrimination as a Barrier to Employment of Older Workers. (Subm). Henkens, K., Remery, C. and Schippers, J., (2005). Recruiting Personnel in a Tight Labour Market: an Analysis of Employers’ Behaviour. International Journal of Manpower, 26(5): pp. 421–433 Ilmarinen, J., Huuhtanen, P. and Louhevaara, V., (1999). Developing and Testing Models and Concepts to Promote Work Ability During Ageing. In: FinnAge - Respect for the Aging: Action Program to Promote Health, Work Ability and Well-being of Aging Workers in 1990-99 edited by Ilmarinen, J. and Louhevaara, V. (Helsinki: Finnish Institute of Occupational Health. People and Work. Research Reports 26), pp. 263–267 Ilmarinen, J. and Louhevaara, V., (1999). FinnAge - Respect for the Aging: Action Program to Promote Health, Work Ability and Well-being of Aging Workers in 1990–96. (Helsinki: Finnish Institute of Occupational Health. People and Work. Research Reports 26). Ilmarinen, J., (2003). Promotion of Work Ability during Aging. In: Aging and Work, edited by Kumashiro, M. (London: Tylor and Francis) pp. 23–35 Ilmarinen, J., (2004). Finnish National Program on Ageing Workers (FINPAW) 1998–2002. Paper presented at the Third International Course on Age Management in the Information Society. Ilmarinen, J., Tuomi, K. and Klockars, M., (1997). Changes in the Work Ability of Active Employees over an 11-year Period. Scandinavian Journal of Work Environment and Health, 23: pp. 49–57 Ilmarinen, J. and Rantanen, J., (1999). Promotion of WorkAbility DuringAging. American Journal of Industrial Medicine, Supplement 1: pp. 21–23 Miles, M. B. and Huberman, M., (1994). Qualitative Data Analysis. (Thousand Oaks: Sages Publications). Mykletun, R.J., Furunes, T., and Solem, P.E., (2006). Extending Senior Workforce Careers: Public Sector Managers’ Beliefs about Adequate Measures. (Subm). Neuman, W. L., (2003). Social Research Methods: Qualitative and Quantitative Approaches. (Boston: Allyn and Bacon). Skoglund, B and Skoglund, C., (2005). Åldersmedvetet ledarskap. Att inte skjuta problemen framfør seg. (Age Management. Face the Problems Now.) (Boden, Sverige: Age Management i Sverige AB (AMSAB)). Sims, H. P. Jr. and Lorenzi, P., (1992). The New Leadership Paradigm. Social Learning and Cognition in Organizations. (Newburry Park, CA: Sage). Yukl, G., (2006). Leadership in Organization. (Upper Saddle River: Pearson Prentice Hall).

175

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Managers’ decision making latitudes in relation to managing ageing workers Trude Furunes & Reidar J. Mykletun Norwegian School of Hotel Management, University of Stavanger, Stavanger, Norway

ABSTRACT: In order to put age management into practice, it is important that managers at all levels are involved. Drawing on three different samples, this study illustrates how manager decision making latitude in regard to age management practise can be defined and measured across samples. The core dimensions identified were access to human and budget resources, possibilities to organise work to meet the demands of ageing workers, and possibilities to unite age management with increasing demands on effectiveness. The decision latitude measure was found to differ with managers’ beliefs about and attitudes towards ageing and older workers, and may also differ in size across organisations. Managers’ age and work ability did not affect their perception of decision latitude.

1 INTRODUCTION Although there has been an increasing focus on age management and work ability in the Nordic countries, results in terms of extended work careers and better possibilities for older workers at the labour market have been slow in coming. For that reason, it is suggested that managers may have difficulties in handling their age management duties. One of the intentions of introducing the age management concept as stated by Ilmarinen (Ilmarinen, 2004), was to empower both white and blue collar workers to reach both personal and corporate goals regardless of their age. This should be done through considering age-related factors in daily management in relation to management styles, task design, work environment, and organization of work. Hence implementing age management may, by and large, be an encompassing managerial task. To better understand managerial challenges with age management, this paper describes and explains which issues managers perceive as limitations to their decision making latitude on age management. Decision making latitude implies the amount of freedom managers have to perform and to change existing practices or modify current structures .(Furunes, Mykletun and Solem, 2007) in order to implement age management within their unit. Similar to this, Meijman and Mulder (1998, pp. 22) describe decision latitude as “the possibility of control which the situation allows”, and their reasoning is seemingly appropriate for age management. For a manager to utilise his or her decision latitude, he or she has to perceive it. The core components identified as defining a manager’s job as well as strongly influencing managerial behaviour are demands, constraints, and choices (i.e. Stewart 1982, in Yukl, 2006). Yukl adds on the three situational determinants pattern of relationships, work pattern, and degree of exposure to decision making and associated consequences. Based on these components and from information contained in relevant management literature, seven different aspects that potentially influence managers’ decision making latitude on age management emerged and were included in the survey. According to Bowditch and Buono (2001) it is important for managers to have the power to influence work procedures including practical possibilities to organise work for older workers (i.e. choice). Further potential prerequisites for contributing to organisational performance through 177

optimal handling of older workers could be availability of adequate budget resources, as well as the availability of human resources (i.e. constraints) (Bowditch and Buono, 2001). Implementing age management might be conceived of as breaking the informal conceptions of organisational justice (Colquitt, Conlon, and Wesson et al., 2001). Along the same lines, Yukl stated that managers often have less freedom of action, here defined as decision latitude, as one goes down through the authority hierarchy. This also implies that lower-level managers have less time, and to a lesser degree are able to allocate resources, as well as make their own decisions. Hence, the fourth potential aspect could be managers’ possibilities to fit age management duties into a work schedule with increasing demands for efficiency within the organisation (i.e. demands). Three additional aspects that according to Yukl could influence managers’ decision latitude are patterns of relationships and communication, here defined as managers’ communication with older workers; support from co-workers for age management issues, as well as employees’ attitudes towards age management. Implementing age management might be conceived of as breaking the informal conceptions of organisational justice (Colquitt et al., 2001). In this study a scale for managers’ perceived decision making latitude on age management was tested and established. In this process it was interesting to explore if managers’ perceptions were equal across samples, or if perceptions would vary from sample to sample. The second aim of this study was to see how managers’ decision latitude differs with manager’s age, work ability, their own beliefs about the role of the employer in relation to preventing early exits and work design for the mature workers, his/her knowledge about relevant work re-design for elderly, and his/her own attitudes to the timing of their own retirement.

2 METHOD The study utilizes three samples of managers, respectively in public sector organizations (n = 672), a public hospital (n = 114) and an energy company (n = 352). Data were collected by use of survey questionnaires, and the response rates were respectively 42.3, 60.3 and 49.1 percent. All participants answered a set of seven items concerning their possibilities to manage their age management program. The seven items were developed from management literature as discussed in the introduction and covered the following issues: (1) possibilities to organise work; (2) available budget resources; (3) available human resources; (4) managers’ possibilities to fit age management duties into a work schedule with increasing demands for efficiency within the organisation, (5) communication with elderly workers, (6) support from co-workers, and (7) employees’ attitudes towards age management. In a factor analysis (Principal Component Factor Analysis, Varimax rotation), four items (listed 1-4 above) loaded on the first factor for all three samples. Internal consistency tested by Cronbach’s Alpha was .69, .78 and .76, correspondingly. For each sample, a sum score variable for ‘decision latitude’ was created, based on these four items. These sum score variables were used as dependent variables in three separate multiple regression analyses. In the regression analyses, the predictors were manager’s beliefs about and attitudes towards older workers and age management, managers’ work ability and manager’s age.

3 RESULTS This study developed a measure for evaluating managers’ decision making latitude on age management duties. Across these three samples, the data showed that four variables load on factor 1 in a Principal Component Factor Analysis, with varimax rotation. The first factor explained 30–34 percent of the variance across these three samples. There were at least two different ways of applying and interpreting these factors. First it was interesting to see how managers’ perception of decision making latitude varied across different samples. Although the same four variables loaded on factor 1 in all three samples, this does not imply that the internal distribution of responses was equal across the samples, as can be seen in the 178

Table 1.

Factor loadings for component 1 for all three samples.

Possibilities to organise work Availability of human resources Availability of budget resources Possibilities to unite age management with increasing demands on efficiency

Table 2.

Energy company

.674 .786 .772 .611

.732 .767 .821 .744

.727 .723 .766 .732

Strongly disagree

Disagree

Agree

Strongly agree

6.7 30.1 34.8 36.4

45.6 49.6 50.8 43.3

38.2 18.3 12.7 18.2

9.5 2.0 1.7 2.1

Frequency levels for public hospital managers (n = 114).

Possibilities to organise work Availability of human resources Availability of budget resources Possibilities to unite age management with increasing demands on efficiency

Table 4.

Public hospital

Frequency levels for public sector managers (n = 672).

Possibilities to organise work Availability of human resources Availability of budget resources Possibilities to unite age management with increasing demands on efficiency

Table 3.

Public sector

Strongly disagree

Disagree

Agree

Strongly agree

10.5 41.4 37.8 39.4

52.8 39.6 46.6 43.1

31.6 15.8 14.2 14.7

4.4 2.6 0.9 2.8

Strongly disagree

Disagree

Agree

Strongly agree

7.8 16.9 17.8 27.0

32.8 39.6 37.6 37.4

45.7 35.8 34.9 27.5

13.1 7.4 9.4 8.1

Frequency levels for energy company managers (n = 352).

Possibilities to organise work Availability of human resources Availability of budget resources Possibilities to unite age management with increasing demands on efficiency

factor loadings and frequency distributions (Tables 1–4). In itself this was an interesting finding, as it illustrated that managers’ decision latitude on age management is perceived differently in different organisations. As displayed in Table 2, public sector managers report that they to some degree have possibilities to organise work for older workers (ca. 48% agree). However, lack of human and budget resources are perceived to constrain managers’ decision latitude. About 20 percent perceive to be in control of demands; the remaining 80 percent disagree in having possibilities to unite age management intentions with increasing demands. 179

Table 5.

Regression analysis for ‘decision latitude’-variable, beta correlation coefficients and significance levels.

Predictors To what extent can the employer prevent early exits I have a good dialogue with elderly workers and know their needs I think it is important to design work to fit older workers I think there is too much focus on age management issues I miss information about how to redesign work for elderly workers Age Work ability To what extent would you prefer early retirement

Public sector

Public hospital

Energy company

.203∗∗∗

.046

.257∗∗∗

.128∗∗

.015

.006

.000

.033

.116∗

.098∗

.322∗∗∗

.161∗∗

−.179∗∗∗

−.187∗

−.261∗∗∗

−.014 −.024 −.093∗

−.043 .111 .066

.020 N/A N/A

* p < .05, ** p < .01, ***p < .001

Among public hospital managers about 36 percent agree that they to some degree have possibilities to organise work for older workers (Table 3). Also here, lack of human and budget resources is perceived to constrain managers’ decision latitude. In this sample only 17 percent perceive to be in control of demands; the remaining 83 percent disagree in having possibilities to unite age management intentions with increasing demands. Although the same variables loaded on the decision making latitude factor for all samples, the frequency pattern was somewhat different for energy company managers than observed in the previous two groups (Table 4). Almost 59 percent report that they have possibilities to organise work for older workers. Respectively, 43 and 45 percent agree that they have adequate budget and human resources, whereas the remaining perceives resources to constrain age management efforts. About 35 percent perceive to be in control of demands; the remaining 65 percent disagree in having possibilities to unite age management tasks with increasing demands. As the second aim of this study was to see how managers’ decision making latitude differed with manager’s age, work ability, and their beliefs and attitudes, multiple regression analyses were applied (Table 5). According to these analyses, managers’ age and work ability did not affect their perception of decision making latitude, as low Beta values and no significant values were observed for these predictors. In the public sector and energy company samples managers’ beliefs in employer’s influence on early exits were significant predictors. Managers who believe that the employer can prevent early exits (i.e. believe in age management intentions) reported to have more decision making latitude. Similar results were found for managers who think there is too much focus on age management, this effect was significant across all samples, but strongest for public hospital managers. Another predictor that was significant illustrated that managers who miss information on how they can redesign work for older workers perceive to have less decision making latitude. In the public sector sample two additional predictors were significant, indicating that managers having a good dialogue with older workers perceive to have more decision making latitude, and managers who would prefer early retirement themselves perceive to have less decision making latitude. In the energy company a significant effect was found for managers who think redesigning for older workers is important, as these managers also perceived to have more decision making latitude. In sum, the analyses showed that managers’ own work ability and age did not predict managers’ perceived decision latitude, however managers’ attitudes towards age management duties to a varying degree did. 180

4 DISCUSSION Managers’ decision making latitude may be of salience to develop work organisations towards more age inclusive practices. The findings of this study support the statements of Stewart (1982, in Yukl, 2006) on leadership being influenced by demands, constraints, and choices. Managers feel that they have limited freedom to act when it comes to adequate adaptation of the workplace to mature workers’ needs. At least seven measurable dimensions restrict managers’decision making latitude on age management, and four of these were consistently integrated across three studies, supporting the proposition of an age management decision making latitude construct. Furthermore, accepting an organisational need and obligation to develop age-inclusive workplaces, owners and boards of work organisations should be concerned about granting managers more choices to organise work and access human and budget resources, and less conflicting demands and fewer constraints in combining unit effectiveness with considerations on individual human capacities (Furunes et al., 2007). Managers’ perception of decision making latitude may vary from one industry to another as a consequence of organisation specific differences in traditions, cultures, type of work, and budgets. Generally the study showed that the energy company selling their products and services on the private market reported larger decision making latitudes on age management, while the public sector and the hospital being publicly owned “budget organisations” have narrower frames for managers in these issues. However, these findings are based on one study with limited sampling and need to be substantiated by future research. Managers’ perception of decision making latitude appears to vary according to their knowledge, attitudes, beliefs and the situational determinants like relationships to elderly workers. This may imply that organisations have to put more effort into raising awareness and educating their managers on ageing and the age management concept as well as influencing their attitudes. Requiring managers to report to their supervisors on age management results would most likely be useful for an adequate use of increased decision making latitude, as managers tend to do what pays off or gives access to rewards. Finally, contextual factors may influence managers’ attitudes and beliefs. Managers’ attitudes towards employing elderly workers have changed from 49 percent being positive in 2004 to 63 percent being in favour of elderly workers in 2006. Lack of labour has been proposed as the main explanatory factor behind this radical change (Dalen, 2007), forcing managers to also employ elderly workers. ACKNOWLEDGEMENT The authors are indebted to Professor Linda Stromei for skilful help with the manuscript. REFERENCES Bowditch, J. L., and Buono, A. F., (2001). A primer on organizational behaviour 5th Edition. (New York: John Wiley & Son, Inc). Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., and Ng, K.Y., (2001). Justice at the millenium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology. 86 (3): pp. 425–445 Dalen, E., (2007). Norsk seniorpolitisk barometer: Ledere i arbeidslivet. (Oslo: Synovate/MMI. Report). Furunes, T., Mykletun, R. J., and Solem, P. E., (2007). Age management in the public sector: Managers’ decision making latitude. Submitted for publication. Ilmarinen, J., (2004). Finnish National Program on Ageing Workers (FINPAW) 1998–2002. Paper presented at the Third International Course on Age Management in the Information Society. Meijman, T. F., and Mulder, G., (1998). Psychological Aspects of Worklaod. In Work Psychology. 2nd ed. edited by Drenth, P. J. D., Thierry, H., and Wolff, C. J. D. (East Sussex: Psychology Press). Yukl, G., (2006). Leadership in organizations 6th Edition. (Upper Saddle River: Pearson Prentice Hall).

181

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

The DRUVAN-project: A major increase in Occupational Health Service based on the Metal Age method in a Finnish municipality improved the work ability and gave significant financial return Ove Näsman Dalmed OyAb, Dalsbruk, Finland

Guy Ahonen Department of Management and Organisation, Hanken Business School, Helsinki, Finland

ABSTRACT: Following increases in sick-leaves and early retirements among employees in the small municipality of Dragsfjärd in Finland, it was decided to initiate a health promotion and prevention program. The annual investment in occupational health and safety (OHS) increased from 20 to over 400 Euro/employee. Method: The multidimensional OHS model is a process which continues throughout the whole working life from recruitment to retirement, including health checkups, rehabilitation courses, medical treatment, ergonomics, work place analysis, physical activities and intoxicant programs. The expert advice given by the OHS personnel to the management is an important part of the Druvan project. The development of the work content and work community is based on a systematic use of the Metal Age method for participative planning for increased wellbeing, work ability and total productivity. Results: At the end of the project, there was a significant decrease in sick-leave and fewer early retirements. The Kiva personnel questionnaire showed that, compared with other employees, the employees in the municipality of Dragsfjärd were more satisfied with their situation at work. The work ability index (WAI) developed favourably. In a cohort (N = 67) there was during four years only a 0.13 point decline in the WAI score as opposed to an expected decline of 2. The monetary investment in the project gave a 46% return on investment (ROI). Discussion: The results are very encouraging. The participative approach using the Metal Age method played a crucial role in this. A 20-fold increase in investment in multidimensional OHS is only possible if the organization has a low expenditure in OHS before the increase. Conclusion: The results indicate that a work place health promotion program based on participation and collaboration can significantly increase the wellbeing and work ability among the employees. A major increase in the OHS expenditure can be very profitable.

1 INTRODUCTION Due to an increase in sick leaves and early retirements 1998–2001 among the 190 employees, the council of the small municipality of Dragsfjärd in Finland, decided unanimously on starting a threeyear project (2002–2005), named the DRUVAN project. The annual investment in Occupational Health Service (OHS) increased from 20 to over 400 Euro/employee. The multidimensional OHS model was modified from the local steel industry. 2 METHOD The Druvan model (Druvan = Grape in Swedish) is a multidimensional OHS process (Figure 1), which continues throughout the whole working life from recruitment to retirement including health 183

Municipality of Dragsfjärd 2002–2005 Promotion of Well-Being, Work Ability and Total Productivity

Retirement

Basic tasks of Occupational Health Service

Activities based on age or years of service

• Medical treatment 57–58 years Sundia (10 empl./year)

• Ergonomics • Physical exercise -pause gymnastics, ball games, Nordic walking etc.

↑ Health check-up every 3rd year 50years ↓ Health check-up every 5th year

• Intoxicant programs - Shall we talk? - No-smoking activities - Drugs • Work place investigations

 15–30 years Redirecting resources (Wire-Impuls)

• Statutory health controls

Expert advice for the line organisation • Development of the work community - Metal-Age, 3 groups/year - Special topics - Solution based development - Personnel questionnaire x2/year

 3–15 years

Wireys-course (3 empl./year)

• Development of work contents • Development of vocational skills • Continuous development of leadership • Birthday discussions

Employment health check-up

Employment Figure 1. The Druvan Program.

checkups, rehabilitation courses, medical treatment, ergonomics, work place analysis, physical activities and intoxicant programs. The expert advice given by the OHS personnel to the management is an important part of the Druvan project. The development of the work content and work community is based on a systematic use of the Metal Age method for participatory planning for increased well-being, work ability and total productivity. The Kiva-questionnaire (Kiva = Nice or Fun in Finnish) was used during the project to assess the working climate. The Birthday discussion is a mini development discussion. The supervisor gives a small present to the employee at his/her birthday and they have a cup of coffee together while discussing the employee’s well-being at work. 2.1 Metal age The Metal Age (MA) method for participatory workplace health promotion was developed by Näsman and Ilmarinen (1999). The model is a solution based method which identifies problems at work and turns the problems into opportunities and finally generates positive manageable solutions to them. 184

In the MA process the personnel of the company or organization is divided into groups of 5–20 people, often by departments or people normally working together. Each group has development sessions that take about eight hours in total, followed by a shorter follow-up session usually 3–6 months later. The MA process consists of the following phases: Orientation Identification of development (problem) areas Prioritizing Consensus decision-making about concrete, manageable actions Follow-up In the orientation phase, the workplace is seen as a matrix of problems/opportunities, solutions and results on an individual, work unit and enterprise level. In the identification of development areas, the session participants are encouraged to list all relevant development areas they can think of in order to increase their well-being at work. The prioritizing of initiatives is made using a priority matrix. All participants are asked to rate the importance, prevalence and possibility to influence initiatives on a scale of 1 to 10. A total score is achieved by multiplying the ratings. The development area with the highest score is then taken for a more detailed analysis in order to agree on concrete actions, on the person in charge of them, and on dead lines for their implementation. During the first follow-up session, the achievements concerning the concrete actions of the first development area are discussed one by one. Thereafter, the session participants continue by planning concrete actions for the second development area. During the Druvan-project there were nine Metal Age groups. The first Metal Age planning group consisted of employees at the Central administration working in the same building. This was considered a good way of introducing Metal Age in the municipality as many managers worked in this building. It is always important that the manager knows about methods used in his/her organization or problems may appear when implementing the concrete actions agreed on in the Metal Age planning group. The Nursing home’s second development area “Information, meetings, communication” is presented in Table 1. 2.2 Kiva-questionnaire The Kiva-questionnaire is a short, user-friendly questionnaire with seven standard questions, (Table 2), to evaluate the work climate. The work climate was measured twice a year using the standard formula during the Druvan project. The VOOP program enables an interactive presentation of the Kiva-questionnaire results. When presenting the Kiva-questionnaire results online, it is possible to cross-tabulate all Kiva-questions according to the needs and wishes of the respondents. The Metal Age planning sessions were often started with a presentation of the Kiva-questionnaire. This procedure created engagement to develop, for example, comradeship when the respondents could see how, in their own work community, bad comradeship (Question 4) influenced the answer in Question 1 (Enjoyed coming to work). The same changes appear in Question 1 when the respondents have low score in Question 5 (Leadership). A high score in Question 7 (Influence) usually gives high scores in Question 1.

3 RESULTS All measured indicators showed favorable figures and the general opinion among the employees in the municipality was that the Druvan project was a big success. A questionnaire at the end of the project showed that aging employees in particular had benefited from Druvan. 185

Table 1.

Metal Age, Concrete actions for the Nursing home. Development area 2 “Information, meetings, communication”.

Development area 2: Information, meetings, communication Concrete Actions

Responsible person

Schedule

Modul meetings x 3/year, 1–3 pm – spring, autumn, around New Year – participation “compulsory”

Bettina Petra Lena

W23/2005

Department hour – every week, different day, 2–3 pm – short documentation in a booklet – documentation from the unofficial morning coffee gathering in the same booklet but from the other direction

Dea

W21/2005

Personnel meeting 1.30–2.30 pm – January, March, May, September, November – all the personnel in the nursing home – participation “compulsory”

Eva

Sept. 2005

Development of the reporting technique – during module meetings and department hours

Dea

W21/2005

Hannahemmet, Nursing home 19/5/2005

Table 2. The Kiva-questionnaire. Have you enjoyed coming to work in the last weeks? 1 2 3 4 5 Not at all

6

7

8

9 10 Yes, very much

I regard my job as 1 2 3 Not meaningful at all

4

5

6

7

8

9 10 Very meaningful

I feel in control of my work 1 2 3 Not at all

4

5

6

7

8

9 10 Very much so

I get on with my fellow-workers 1 2 3 4 Not at all

5

6

7

8

9 10 Very well

My immediate superior performs as superior 1 2 3 4 5 Very poorly

6

7

8

9 10 Very well

How certain are you that you will keep your job with this employer? 1 2 3 4 5 6 7 Not certain at all

8

9 10 Very certain

How much can you influence factors concerning your job? 1 2 3 4 5 6 Not at all

8

9 10 Very much

7

3.1 Work Ability Index (WAI) The WAI (Ilmarinen et al., 1997, Ilmarinen and Tuomi, 2004) was completed by 67 persons before (2001) and after (2005) the Druvan project. The expected decline in WAI during four years among 186

8.1 8 7.9 7.8 7.7 7.6 7.5 2002

2003

2004

2005

2006

Ref.4000

Figure 2. The development of the average score of the seven Kiva-questions. The reference score is from 4,000 respondents from different work places.

municipality workers with mean age over 40 years is of about 2.4. The decline during the Druvan project was of only 0.13 in the above cohort. The WAI-cohort study will be completed in 2008. Correlations between different Druvan activities and the WAI-development are being studied. These activities are physical activities, workload, sick leaves, development discussions, gymnastics pauses, ergonomics, weight control, tobacco, visits to the OHS unit, rehabilitation courses, health check-ups etc.

3.2 Kiva-questionnaire The work climate measured by the average of the seven Kiva-questions was already initially at a good level compared with a reference material that compiled results from 4,000 respondents from different work places (Figure 2). The total average scores showed a slight u-formed curve for the measured years. The first measurement was done half a year after the start of the Druvan project, so no baseline measurement exists. The high first measurement may be due to a Hawthorne effect (Roethlisberger and Dickson, 1939).

3.3 Sickness absence There was a dramatic decrease in sickness absence. Dragsfjärd had a much higher sick leave rate than the average Finnish municipality before the Druvan project. After the project, the situation was the opposite (Figure 3). The comparison between Dragsfjärd and all Finnish municipalities shows that, in reality, the development is even better than that seen in Figure 2. The numbers from Dragsfjärd are real numbers whereas numbers from other Finnish municipalities are taken from annual interviews. Weekends and days off are also excluded from the results for the numbers for all Finnish municipalities. If this were also calculated using Dragsfjärd’s sickness absence, the numbers would decrease by more than 20%.

3.4 Retirements The retirements are few because there were only 190 employees. However, the trend since the start of the Druvan project has moved from disability pensions to old age pensions (Table 3). 187

18,0 16,0 14,0 12,0 10,0

Dragsfjärd Finland

8,0 6,0 4,0 2,0 0,0 2001

Figure 3.

Table 3.

2002

2003

2004

2005

2006

Sick leave days/employee in Dragsfjärd and in all Finnish municipalities.

Old age and disability pensions in Dragsfjärd 2001–2006.

Old age pension Disability pension Partial disabilty pension

2001

2002

2003

2004

2005

2006

2 2 0

1 1 1

1 2 2

4 1 0

3 1 0

0 1 1

3.5 Economic analysis A Cost-Benefit Analysis (Ahonen et al., 2002) was made using the Potential Model (Miljodata, 2002). The analysis shows that there was a 46% annual return on investment (ROI) as a result of the project. The labor cost of efficient working hours before the intervention was calculated with this model. This unit value and direct monetary savings were used for calculating the total economic benefit of the intervention, which was compared with the direct monetary investment to calculate the financial return on investment (ROI) of the project.

4 DISCUSSION The results were very encouraging. The multidimensional Druvan model made it possible for almost every employee to participate in something he or she benefited from personally which is crucial in creating engagement. This was the main reason for Druvan becoming “Our Project” among the municipality workers. The participatory approach using the Metal Age method and the Kiva-questionnaire were very important in this aspect. It is difficult to engage employees in a project focusing on decreased sickness absence and longer working life. It is much wiser to use the positive marketing approach of the project as being a project aimed at increasing wellbeing. None can oppose such a project and the outcome still remains decreased sick leaves and longer working life. The collaboration and participation of many external experts from different fields gave important advice to the local actors and lent authority to the project. The economic analysis could not have been made without external expertise. 188

A 20-fold increase in investment in multidimensional OHS is possible only if the organization has a low expenditure in OHS before the increase. The Druvan project ended in 2005 but the municipality has decided, due to the good results, to continue with multidimensional OHS in the same manner as during the Druvan project. The project activities have since become normal procedure. In conclusion, the results indicate that a work place health promotion program based on participation and collaboration can be very successful as measured with wellbeing, work ability and monetary units and may lead to a new, lasting way of maintaining and developing the wellbeing of the employees and the whole organization. REFERENCES Ahonen, G., Bjurström, L-M., and Hussi, T., (2002). Economic Effectiveness of Maintenance an Promotion of Work Ability in Peltomäki et al. (Eds.): Maintenance of Work Ability Research and Assessment: Summaries. Ministry of Social Affairs and Health, FIOH, Social Insurance Institution, Report No 7: pp. 33–44, Helsinki Ahonen, G., and Näsman, O. The Economic Consequences of Multidimensional Development of Human Capital by Using the Metal Age Method in a Finnish Municipality. Melbourne: Monash University, Melbourne 2007. Paper presented at The Eighth International Research Conference on Quality, Innovation and Knowledge Management, New Delhi, India 12–14 February 2007 Ilmarinen, J., and Tuomi, K., (2004). Past, present and future of work ability. People and Work Research reports, 65: pp. 1–25 Ilmarinen, J., Tuomi, K., and Klockars, M., (1997). Changes in the work ability of active employees over an 11-year period. Scandinavian Journal of Work Environmental Health, Vol. 23 suppl 1: pp. 49–57 Miljodata, (2002): The Productivity model. A computer based model to calculate the business economic effects of work environmental investments. www.miljodata.se Näsman, O., and Ahonen, G. The DRUVAN-project: Participatory development of wellbeing, work and productivity by using the Metal Age method in a Finnish municipality. 39th Nordic Ergonomics Society Congress. CD. October 2007. Lysekil, Sweden Näsman, O., and Ilmarinen, J. Metal-Age: A process for improving well-being and total productivity. Experimental Aging Research Vol 25, Number 4, USA Roethlisberger, F., William, J., and Dickson, W.J., (1999). Management and the Worker. In: The Early Sociology of Management and Organization, Ed. Thompson, K., Routledge (UK) 1939.

189

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

A study on effects of support systems for KAIZEN in casting production of non-ferrous alloy by aged foundry workers Yuki Mizuno & Nobuyuki Motegi Institute for Science of Labour, Kanagawa, Japan Musashino University, Tokyo, Japan

Miyuki Sugiura Juntendo University Graduate School, Chiba, Japan

Fumiko Matsuda Institute for Science of Labour, Kanagawa, Japan Musashino University, Tokyo, Japan

Toru Yoshikawa & Kazuhiro Sakai The Institute for Science of Labour, Kanagawa, Japan

Tetsuo Misawa Chiba Institute of Technology, Chiba, Japan

ABSTRACT: This study evaluated the effects of work environment improvements on aged workers by applying the KAIZEN method. It was based on a field study conducted at a foundry in Japan. Using the viewpoint of Ergonomics we investigated the work loads involved in the operation of smelting machines, which provide a key part of the casting production process. Based on our findings, we developed support systems to reduce the work loads. We found that workers complained of high fatigue of their eyes and feet. Because there were fumes created by the burning of heavy fuel oil, workers complained of pains in their eyes and throats. Following the principles of the KAIZEN method, support systems were developed that reduced heavy life work and created work environments in which it was easy to perform work. Because changes were made in the fuel, changing from heavy fuel oil to gas, the amount of exhaust gas decreased. In addition, the pain in workers’ eyes and throats was reduced. Moreover, a new gas furnace that managed the ignition and fire power adjustment collectively by using a control panel was introduced. Operations such as ignition and temperature control have needed expert skills so far, but these aims can now be achieved and all workers can easily carry out their work by using the control panel of the new gas furnace.

1 INTRODUCTION This study aimed at evaluating effects of work environment improvements for aged workers by applying the KAIZEN method. It was based on a field study in a foundry in Japan. It is known that the physiological stress is high in the work environment of the casting production line in a foundry because of the high temperatures and heavy lifting work. In particular, foundry workers easily develop chronic fatigue. Aged foundry workers over fifty-five years old constitute 80% of the total, and many aged workers choose early retirements. There is now a problem that young workers can’t smoothly acquire the skills of the aged workers. 191

Working processes include “preparation of materials”, “checking of fuel”, “igniting a furnace”, “controlling the temperature of a furnace”, “throwing materials into a furnace” and “taking out smelted materials to a sand form”. The work of controlling the temperature requires high skill and often relies on aged workers who have become skilled. Because the fireplace is on the 2nd floor, workers must carry heavy materials by hand from the 1st floor. Therefore, throwing materials into the fireplace is a job which places a heavy responsibility on aged workers. In this study, using the viewpoint of Ergonomics, we investigated the work loads in smelting work using machines which provide a key part of the casting production process. Then, we developed the support systems to reduce work loads.

2 METHOD 2.1 Work involved The working process includes “preparation of materials”, “checking of fuel”, “igniting a furnace”, “controlling the temperature of a furnace”, “throwing materials into a furnace” and “taking out smelted materials to a sand form”. 2.2 Investigation items 1) Time Study The subject was a worker (1 aged worker). We observed the working process of casting products on a working day. Work operation and work place were recorded each second. 2) Investigation by questionnaires Subjects were four workers (1 young, 3 aged workers). They were also asked to fill in a survey sheet of Subjective Fatigue Symptoms and one of Tired Body Parts four times in a working day. The statistical testing examined the significance level by using t-test.

3 RESULTS AND DISCUSSION 3.1 Results of existing condition investigation 1) Time Study The works of controlling the temperature and the ignition required high skills, so aged workers who had become skilled performed this work. In addition, the workers had to take a halfcrouching position during their work many times, so there were high work loads (Fig. 3). Because the furnace was on the 2nd floor, aged workers carried weighty materials by hand from the 1st floor. They performed a lot of movements. 2) Investigation by questionnaires Workers complained of blurred vision, flickering and strain on their eyes caused by fumes resulting from the burning of crude petroleum. They also complained of lassitude, such as backache and fatigue of their feet, because of the wide range of movements required (Fig. 1). 3.2 Design and development of support system 1) Indicators of System Design As part of the KAIZEN method, support systems were developed that reduced heavy life work and created work environments in which it was easy to perform work. The conditions for system design are shown below. #1. Making the work environment easier: Because there were fumes created by the burning of heavy fuel oil, workers complained of pains in their eyes and throats. 192

100% 75% 50% 25% 0% 8:00 12:00 12:50 17:00

8:00 12:00 12:50 17:00

Aged subject A

Drowsiness

Aged subject B

Unbalance

Indisposition

8:00 12:00 12:50 17:00

Aged subject C

Lassitude of body

Figure 1.

Subjective fatigue symptoms; existing works.

Figure 2.

Controlling the temperature (left: existing work, right: new systems).

8:00 12:00 12:50 17:00

Young subject D

Blurred vision

#2. Reduction of the worker’s physiological load: Workers assumed a half- crouching position when controlling the temperature and the work posture involved bending the head forward. #3. Batch processing in igniting the furnace and fire power adjustment: There were a lot of movements because in the ignition and fire power adjustment air was controlled in the 1st floor, and fuel on the 2nd floor. 2) Development of Support Systems Trial systems were developed with workers, managers and researchers. On the basis of the above described requirements, new support systems were developed as follows. #1. Change to gas fuel: The amount of exhaust gas was decreased in burning because the fuel was changed from t heavy fuel oil to gas. #2. New control panel (Fig. 2): A new gas furnace managed the ignition and fire power adjustment collectively by means of a control panel. The working position was changed from a half-crouching to standing position. The color of the display was changed so as to avoid improper operation, and an operating manual was placed on the panel. #3. Use of new bucket: It became possible to carry materials weighing up to 500 kg at once with a new bucket, and carrying by hand was decreased.

3.3 Reduction of work loads by support systems By using new systems, work which had required high physiological loads and difficult movements decreased and the complaints of subjective symptoms decreased sharply (Fig. 3, 4). Especially, the pains in worker’s eyes and throats were reduced (p < 0.01) and they were able to work in a healthy and safe environment. 193

Working hours Number of working times 0

10

20

30

0 500 1000 1500 2000 2500 40

Carrying and placing materials

Carrying and placing materials

Ignition a furnace

Ignition a furnace

Controlling temperature

Controlling temperature

Remove used stairs

Figure 3.

3000 sec

Existing works

New systems

Number of working times and working hours.

100% 75% 50% 25% 0% 8:00 12:00 12:50 17:00 Aged subject A

Drowsiness

Figure 4.

8:00 12:00 12:50 17:00

8:00 12:00 12:50 17:00

Aged subject B

Unbalance

Aged subject C

Indisposition

Lassitude of body

8:00 12:00 12:50 17:00 Young subject D

Blurred vision

Subjective fatigue symptoms; new system.

4 CONCLUSION In this study from the viewpoint of Ergonomics we investigated the work loads involved in smelting work using machines which perform a key part in the casting production process. Based on our findings we developed a support system to reduce work loads. As part of the KAIZEN method, support systems were developed that reduced heavy life work and created work environments in which it was easy to perform work. Because the fuel was changed from heavy fuel oil to gas, the amount of exhaust gas decreased. In addition, the pains in worker’s eyes and throats were reduced. Moreover, a new gas furnace that managed the ignition and fire power adjustment collectively by means of a control panel was introduced. Works such as ignition and temperature control which formerly needed expert skills can now be easily performed by all workers by using the control panel of the new gas furnace. Additionally, they can look forward to an expansion of job categories among workers in the middle and advanced age groups. REFERENCES Mizuno Y, Matsuda F, Yamada Y, & Misawa T. (2004), The Case Study of the Participative KAIZEN by Employees and Researchers. The Japanese Journal of Ergonomics, Vol. 40, Supplement, pp. 210–211. Mizuno Y, Toriizuka T, Misawa T, & Horie Y. (2006), Case Study on the Development of High Usability Equipment and the Work Improvement, Proceedings of the XVIth Triennial Congress of the International Ergonomics Association, CD-ROM. Motegi N, Mizuno Y, Sugiura M, & Misawa T. (2007), Investigative Study of Workload in the Non-Ferrous Metal Casting Manufacture. The Japanese Journal of Ergonomics, Vol. 43, Supplement, pp. 148–149.

194

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Discovering the treasure: The use of the Work Ability Concept and the WAI in a bus company within a workplace health promotion process Jürgen Tempel & Jutta Schramm Occupational Health Practice, Verkehrsbetriebe Hamburg Holstein AG – Pinneberger Verkehrsgesellschaft mbH Corporation (VHH PVG Corporation), Hamburg, Germany

ABSTRACT: Bus drivers generally like their work and try to avoid premature disability. With increasing age they are involved in a decreasing number of traffic accidents, whether they are responsible or not. There is a win-win situation for both the staff and the company in promoting and maintaining work ability. Since 2002, the company doctors of the VHH PVG Corporation (a regional public transportation company in Northern Germany) have established the use of the Work Ability Concept (WAC) and the Work Ability Index (WAI) within the existing work place health promotion process. According to the WAI-results, a “treasure” was discovered. There exists a successful balance of “good work quality and productivity” and “good quality of life and wellbeing”. But a more detailed and complex view of the results based on the WAC components and all data and experiences available reveals a certain instability for the future: The ageing staff will not be able to cope with growing work demands. To maintain the existing state of work ability and to ensure the future success of the enterprise there are three major tasks ahead: Continuation of the appreciative exchange of experience and integration of the new style of leadership into the company climate. Analysing the services and successfully implementing a structure of driving time and breaks that is supportive for the drivers. And reducing the overwork to decrease the risk of premature invalidity with increasing age. Both the WAC and the use of the WAI have proven to be helpful to rise to future challenges.

1 SALUTOGENIC ASPECTS OF WORK Bus drivers generally like their work and are particularly afraid of premature disability. A lot of them are “communicators” and they enjoy the contact with passengers, especially when they get older and have gained supportive professional experience. Many are proud of their good driving style: they easily manoeuvre a big bus in narrow streets, arrive without delay, and are not involved in accidents or minor damages. Others appreciate their autonomy and decision latitude (Karasek and Theorell, 1990): “when I’m on the bus, I’m the boss!” (Geißler-Gruber and Geißler, 2000). Recent research within the VHH PVG Corporation (a regional public transportation company in Northern Germany) and others has focussed on learning from workers who manage to maintain a good state of health and work ability or from those who have successfully recovered from diseases or other restrictions of their work potential (Geißler-Gruber and Geißler, 2000). With increasing age, bus drivers have a decreasing number of traffic accidents, whether they are responsible for them or not (Ell, 1995). This may help explain why 74% of the participants in our study wanted to continue working until retirement age, while only 14% were uncertain and 12% unwilling (N = 132). A win-win situation can be created for both the staff and the company if a mutual interest exists in work place health promotion and a low rate of premature disability. Nevertheless, from 1987 to 2005, 127 employees of the “Verkehrsbetriebe Hamburg-Holstein AG” had to leave work because 195

of invalidity. Their average age was 56 years. The company therefore required an analysis of the actual potential of its staff, so as to define the optimal balance between – growing – demands of the job and the quality of life and well-being of its employees. 1.1 Workplace health promotion and work ability concept In 2001 the VHH PVG corporation started a process based on the above-mentioned studies of Geißler-Gruber and Geißler. The crucial change in paradigm was to avoid a discussion about absenteeism in the company, but rather to learn systematically from those who successfully stayed at work. The defined aims of the process are as follows: • • • • • • • •

Raising responsibility for health among staff and management. Increasing health potential at the workplace so as to decrease the cost of illness. Considering employee health as an advantage in competition with other companies. Achieving a perception of support for their health by the company among at least 80% of the employees. Achieving measurable effects and evaluating results of installed measures. Organising the demands of the job so as to improve the individual’s situation while accounting for the company’s competitive needs. Requiring employees to participate in the process. Reducing the number of sick days to a certain amount.

Within this context, the company doctors began in 2002 to use the scientifically and practically proven model of the Work Ability Concept (Ilmarinen, 1999) to organise and evaluate the change process, to gain valid data using the Work Ability Index (WAI) for cross-sectional as well as longitudinal assessment (Hasselhorn and Freude, 2007), and to develop the so-called “WAI dialogue”. The seven items of the WAI are helpful in creating an employee-centred discussion about what a person is going to do, what the company could do and how the company doctor can support this process (Tempel, 2004). The basic concept is illustrated by the following scheme: 1.1.1 Conditions The corporation had started its workplace health promotion program after a period of restructuring, dismissals, premature retirements and reduction of staff – a development which eventually led to the company’s economic stabilisation. There was also a growing gap between the intentions of senior management and the somewhat risk-centred interest of the previous company doctors. So we found the doors wide open for our program when we started. Management as well as the workers’ committee accepted the use of the WAI and eventually began to show interested in the Work Ability Concept as an explanatory model and practical tool. Since 2002 we have gathered data from 663 WAI-questionnaires, more than half of which stem from a WAI-dialogue.1 Every five years bus drivers are required to pass a medical assessment, which presented a natural fix point for our work. Other employees come to see us because of personal problems or other occupational assessments. Participation in the program is strictly voluntary, and until 2006 we did not link the data to the person assessed. We were just trying to develop a sufficiently valid cross-sectional WAI – epidemiology as the first step, and this decision proved to be very helpful: Bus drivers are forced by law to consult us and to pass the examination. Therefore it took two to three years until they started to trust in our balancing of the obligatory check-up and the support we were offering within the health promotion process. Another important point was to win over the company’s leadership (supervisors, station managers) for the use of the questionnaire, because this would essentially benchmark them against others, which requires their consent. Both

1 Occupational Health Service of the VHH PVG Corporation – Dr. med. Jutta Schramm, Dr. med. Jürgen Tempel (VHH), Dr. med. Peter Schmidt-Wiederkehr (PVG)

196

parts needed and took the time to observe how we carried out the interviews, how we handled the data we generated, and how we organized feed back within the enterprise. Today we follow a rather strict separation: first we administer the driving license examination, fill in the papers, and hand them over to the person being assessed. Then we ask them to participate in the health promotion-interview. Since 2006, with the consent of the employee, we have been able to add WAI-index, category, and results of the different items to personal health records. We ask the interviewees to create a password, which helps us with the follow ups (generating longitudinal data). Since 2002, less then ten persons have declined to join the interviews and there where no complaints about the use of the questionnaire. We need 20 to 30 additional minutes to do this work in a friendly and open minded manner. The company agrees to these proceedings and pays us for the extra time we require. 1.1.2 Work demands and facts Riding the bus is one of the safest means of transportation one can choose (Vorndran, 2006, pp. 17). In the VHH PVG Corporation you will find 1750 employees, about 1300 of whom are bus drivers and 12% are women. They perform irregular day-time work amounting to more than 40 hours a week, depending on public demand and possibilities of organising shift schedules that are based on scientific standards are reduced. 90.7 million passengers are transported annually in 600 busses. Work demands involve mostly local and commuter traffic with 159 bus routes, but also long distance driving and shuttle and substitute service on special occasions. There is a growing need for drivers, which effectively leads to overtime work among existing drivers and difficulties in compensating these hours by granting spare time as is customary. The employees’ average age is 46 years, 42 years among women and 46.6 among men. Split by occupations, bus drivers are on average 46 years old, office staff 42 years, repair workers 45 years, cleaning personnel 47 years, and we find about 150 temporary workers with an average age of 53 years. Most of these are retired bus drivers who wish to continue working, either to support their company or for financial reasons. These figures also show that there is no tendency towards “age discrimination” within the company. On the contrary, maintaining their ageing workforce is a decisive part of the VHH PVG Corporation’s strategy to cope with the lack of employees and to ensure regular service hours. Within the next 15 years 37% of the drivers will reach their regular pension age and about 30 drivers per year will retire. So every driver who must leave work prematurely because of invalidity is one too many, a fact that is well known among the company’s leadership. The local public administration exerts increasing economic pressure: They demand to receive the same service at lower cost or less service at lower cost, transferring the resulting problems to VHH PVG passengers, drivers and the company as a whole. Since July 2007, a bus driver starts working for 1.830 a gross monthly income, advancing towards the limit of 2.030 a over the course of 13 years. From the very beginning, overtime is required, which may be only a minor problem for younger drivers, but can lead into a vicious circle for the aging driver: recovery of health instead of additional pay, and modest but satisfactory living standards are both important components of what we call “quality of life” and “well-being”. Finally, there is a shortage of bus drivers in neighbouring European countries such as Denmark. It is reported that there are better working conditions and higher wages to be had there, and the first VHH PVG drivers have already set off to raise their fortunes. At the same time, former employers try to lure back into their industries drivers who were once mechanics, construction or automotive workers. 1.1.3 Tools Following the stress-strain concept (Wakula, 2007), the work ability index (WAI) (Tuomi, and Ilmarinen et al., 1998) generates data on strain, defined as the individual or collective reaction or coping strategy devised in the face of a certain work demand (stress). This concept has been developed by Rohmert, Rutenfranz, Ulich and others since the 1960s. The surrounding debate 197

Work and Work Conditions (ergonomics, occupational hygiene, occupational safety) • Organization of work • Work spaces and tools • Work postures and movements • Physical load

Employee (resources, health) • Functional capacity • Physical activities and other lifestyle factors • Self-inititative

Work Ability Maintenance

Work Community (management, interaction) • Work organization • Age management • Work arrangements • Schedules

Professional Skills (competence) • Learning • Versatile skills • New technology

Good work ability and health

Good work quality and productivity

Good quality of life and well-being Active and meaningful retirement ‘‘third age’’

Figure 1. Basic model for work ability and its influence in enterprises and work organisation (Ilmarinen, 2006).

continued between Rutenfranz and Ilmarinen and finally had an important influence on the origins of the work ability concept and the related questionnaire. The WAI’s index points or categories measure and describe the balance between these demands and the coping potential of the individual, but higher or lower point scores cannot explain what is going well or wrong. To do this we must gather further information about “work and work conditions”, “professional skills” and the “work community” (Figure 1, chapter 1.2), so as to learn more about the stress, or rather the resources the enterprise creates by means of work-organisation. The WAI has been combined with a number of questions used in the PIZA- project (Frevel, and Geißler-Gruber et al., 2006) and the ABI-NRW-project (Tempel, and Giesert et al., 2005), together with basic facts about the work of the drivers. Apart from this WAI-assessment in 2002, 2005 and 2007, the entire VHH PVG staff was asked what they think about the ongoing work place health promotion process. This data provided an additional view of the company climate, the opportunities for participation open to employees, and their relationship with the company’s management. The results will be presented in the following chapters. In 2006, after four years of experience with the VHH, we occupational physicians asked the staff about their satisfaction with the medical assessment. 198

Mean WAI-index (points)

50

Excellent Good

40

Moderate

30

20

10

Poor R-Quadrat linear
0.015 20

30

40 50 Age (years)

60

70

Figure 2. Work ability and age, VHH PVG Corporation. 2002 to 2007, cross-sectional data, scatter plot.

2 RESULTS: HOW TO DISCOVER A TREASURE After five years we have gained data that represent the staff well according to age (VHH PVG average age 46 years / WAI-sample 46 years), but rather than to gender (percentage of women in VHH PVG 15% / WAI-sample 10%), because more women work in the office and often do not have a bus driving license, hence do not come to us for their annual check-up. The mean WAI among participants of our study was 43 points, and there was significant negative correlation with age (Pearson −.121∗∗ ), though much smaller than we had expected. A scatterplot (Figure 2) gives a first impression of the relationship between age and WAI-points. In all age groups we find a predominance of excellent (total 55%) and good (total 35%) balance of work demands with individual or collective potential of the workers. There is little variation in relation to age and a negligible linear regression: Age does not have the expected negative influence on the work ability of the staff. 7.5% of the surveyed employees have a moderate work ability and 2% have a poor one. The latter categories are more or less evenly distributed across all age groups. Among the occupations, the lowest mean WAI-index is found in the cleaning department where we find three work shifts, higher physical demands and a higher average age of workers. These findings seem quite similar to those of other studies across Germany. The German WAInetwork describes a WAI-index for bus and tram drivers between 43 and 40 points, with an increase at the age of 40 plus to 50 plus years (Hasselhorn and Freude, 2007, 18). However, they are remarkably different from what we know from Finland when looking at the WAI –index points. In terms of the WAI – classification, the most important difference is found among the repair units.2 The VHH PVG sample is completely inconsistent with the findings of Karazman (1999). He found 11.7% of his sample population with a poor and 36.2% with moderate work ability, with a plainly reduced work potential among 55% in the age group of 50 to 59 year-olds (Ilmarinen, 1999). As the work demands of the different driver samples can be considered consistent, more details are required to define and understand the “treasure” we found in our study. The relatively low share of 10% of staff members whose future work ability is uncertain according to WAI points and

2 Sini

Palo, 2000, permission of Ilmarinen and Tuominen. We would like to thank Eva Tuominen, FIOH Helsinki, for the discussions we head in 2006.

199

categories means that about 160 members of staff are threatened by premature disability and may not reach retirement age (Ilmarinen, 1999).

3 DISCUSSION: WHAT CAN WE FIND WITHIN THE TREASURE CHEST? Basically, our positive findings can be considered a result of VHH PVG’s work place health promotion process and occupational safety standards. A more detailed view shows four components that should be analyzed: Staff developments, the new leadership style, occupational health management, and the limits of cross-sectional WAI data within the process of promotion and maintenance of work ability. 3.1 Reduction of staff From 1995 to 1997 the number of VHH bus drivers was reduced by 200, as mentioned above. Within this context it is interesting to note that many of the dismissed staff (may have) had a history of chronic diseases, decreasing work ability, and increasing incidents of sick leave or absenteeism. But staff-reduction measures, taken by the leadership within a specific historical and economic situation, do not explain why 8 to 10 years later such high potential was found among the staff of the VHH. Though at times maybe necessary, such measures are neither suitable to solve the problems of an aging workforce, nor can they promote or maintain work ability. Hence, staff-reduction cannot explain our positive findings. 3.2 The appreciative exchange of experience – a new leadership style Once restructuring was completed – the number of bus drivers has been constant since 1997 – the leadership initiated a new style of communication and support for the employees, which led to the creation and implementation of what is called an “appreciative exchange of experience” (Geißler, Bökenheide et al., 2003). Every single employee or worker is seen and appreciated as an experienced carrier of ideas and knowledge. Within this context managers can influence the work ability by “reflecting on and changing their opinions of the workforce, adopting a credible participative-cooperative leadership style, and considering, maintaining and even increasing the existing resources at work that healthy and recovered staff members regard as indispensable” (Geissler-Gruber, Geissler et al., 2004, 215;, Geißler, Bökenheide et al., 2003). The implementation of this appreciative exchange of experience throughout the corporation is not yet completed, but first results show its contribution to the current high level of work ability, and demonstrate the advantage of the concept for both supervisors and staff (see chapter 3.2.4). 3.3 Work place health management since 2001 From the very beginning of the program there was a mixture of measures applied to raise individual health (see chapter 3.2.1) as well as to improve work demands in general (see chapter 3.2.2). In 2007, more than 50% of the staff rated the project as “very good” (7%) or “good”, 32% were “indifferent”, and 4.6% were not aware of the project. In 2007, 58% agreed with the statement that the company is supporting their health, while 29% were indifferent. As to the work demands, more than 60% of the participants are generally content, while 32% are indifferent. All ratings have been improving since 2002. In 2005 a further question was added: “How content are you with extra driving shifts during your free period?” 30% of the respondents are rather content, 30% are indifferent, and another third are not content. This rating is decreasing, showing growing dissatisfaction with overtime work. Asking about contentment with the duration and timing of breaks within the service provides similar results, also with a decrease in rating during the past two years. Two time-study groups have 200

been set up to gain deeper understanding of the details involved, but had to be cancelled due to the threat (and fear) of additional costs. All in all, we find growing acceptance among VHH PVG employees of the health promotion work that is being done, though recent opinion polls contain early hints at excessively tight work schedules (see chapter 3.2.2). Even though there was no WAI-epidemiology before the health promotion process started, we can say that its activities contributed to the positive results. 3.4 Limits of cross-sectional WAI-data Up until now we have gathered cross sectional data based on a mean age of 46 years. Finnish longitudinal research shows that the most significant decline in work ability appears at 45 to 60 years of age (Ilmarinen, 1999), where the decline follows an exponential pattern rather than a linear one. In the VHH PVG sample there seems to be no decline of work ability with growing age, but in fact we can observe an enormous change in work demands: longer services, overtime work, increasing traffic problems, social difficulties with passengers, and economic measures to cut costs by up to 20%. Work within the bus company entails many salutogenically complex tasks and requires learning new things (“you never know what will come next, that’s why I like my job”). But it also involves tight schedules changing for the worse, decreasing possibilities to manage one’s work and free time, and few chances to take a break according to individual needs. Aging employees or workers are especially sensitive to these changes, and only when they “are able to make their own choices and adjust to their work according to their own resources, aptitudes and routines, they will cope well in their work” (Ilmarinen, 2006). Actually, bus drivers are gradually loosing the possibility of choice. The company has sufficient time to plan for further steps, but the direction of the process is still somewhat open. The results we found can help us look back at the past years and explain our WAI-findings. But they have small predictive value for the future. The next steps to promote and maintain the prevailing work ability must be defined by the steering committee.

4 THE USE OF THE WORK ABILITY CONCEPT “The most important asset of employees in work life is their work ability. Because enterprise profits are made possible by the work ability of its personnel, the enterprise has a certain role in supporting and promoting the work ability of its employees. Concepts of work ability have changed and developed during the last decade in a more holistic and versatile direction” (Ilmarinen, 2006). Accordingly, it is the task of the steering committee to integrate all existing knowledge, data and scientific understanding that is available and necessary to support the company in taking its next steps in the right direction. In other words: State-of-the-art technology and most comfortable buses cannot replace the driver. His or her work ability is based on four components that should be kept in mind and view. 4.1 Employee Over 80% of the staff rate their current work ability between 8 and 10 points with “lifetime best” at 10 points (WAI-item Nr. 1). The overwhelming majority can use their potential to cope with the daily work demands, physically and mentally (WAI-item Nr.2). About 75% have no or only one current disease, and 38% have mostly one illness or health problem (“own opinion”) with which they cope successfully (WAI-item Nr.3). 85% have no estimated impairment due to diseases, and 10% experience some symptoms, but are able to do their work (WAI-item Nr.4). 47% have taken no sick-leave during the previous 12 months, and 26% have had nine or fewer days off for health reasons (WAI-item Nr.5). We found this self-reported data to be consistent with data we collected from VHH personnel in 2006. The 2006 data was gathered anonymously and categorized in the 201

same way as in the current study. In 2006, 95% of the respondents were relatively certain that, from the standpoint of their health, they would still be able to do their current job two years from then (WAI-item Nr.6). About 5% had rather reduced mental resources, with only 2 points or less as compared to the average score of 4 or 3 points (WAI-item Nr.7). Apart from the medical tests described above, the bus drivers must pass a neuropsychological test to acquire the driving licence and repeat it every 5 years once they reach the age of 50. The parameters tested for here (resilience, orientation, concentration, awareness, and speed of reaction) do not really inhibit one’s driving skills (and license). These are limited by diseases such as musculoskeletal deficiency and disorders, psychosomatic illnesses and sufferings, overweight, diabetes and hypertension (“metabolic syndrome”), coronary heart disease, infarction of the heart, and stroke. Incidence and prevalence of these diseases are not higher among VHH PVG bus drivers than in the German population as a whole. The crucial question is how to get the employees to play their own part within the work place health promotion program. In the WAI-dialogue, we used the items of the index to structure the assessment and to develop employee-centred communication. The aims of the dialogue are not only to give information about health risks, but to raise individual initiative, self-responsibility and enjoyment in getting involved in the health promotion process. In 2006 we assessed the quality of communication between bus drivers and the physicians during the health check-up meetings by asking: “When you have an appointment with the company doctor, do you have the feeling that you can talk about your problems?” “Do you have the feeling that the company doctors are listening to what you say?” “Are the decisions and explanations of the company doctors comprehensible to you?” 75% of the answer were positive (“very well” and “well”), 15% had not had an appointment within the relevant time frame, and 10% rated their experiences as negative (“indifferent”, “little” and “bad”). The following figure compiles our experiences and ideas concerning the WAI-dialogue. Table based on discussion with Ilmarinen, 2006; Geißler-Gruber, Geißler and Bökenheide, 2007. In the steering committee, we are altogether deeply convinced that fear is an inacceptable healthinfluence which should be substituted by providing options, free choice and participation, individual and collective support, sustainable contracts, and a defined right for both the company and the employee to end the health promotion activities without discrimination. The corporation has prepared different offers to accommodate its employees: Some bus stations now have fitness rooms, others offer physiotherapy and massages (company pays the therapist, employee participates in spare-time), and there are walking-groups and soccer teams. Poor physical fitness can be considered the highest risk factor for impaired health, followed by smoking, hypertension, cholesterol and obesity (Blair et al., 1996, stated in Ilmarinen, 1999). Therefore we launched the two year-program “active with all one’s heart” in 2006 in order to encourage employees to exercise. In whichever way they like: (belly) dancing, inline skating, swimming, or walking are all fine, as long as the activity is carried out twice to three times a week in their spare-time. The company contributes up to 500 a per year towards expenses like gym memberships, swimming pool entry, or a bicycle. The participant signs a contract and has a little notebook to register all his activities. There are regular meetings for information and to share experiences, comments and questions. To spread the idea, all bus stations have been visited and possibilities and wishes were recorded. 76 people have joined the program so far. Their average age is 49 years, and about two thirds are “real beginners” in sports after a lengthy period of inactivity. Sometimes it is necessary to restore health and work ability in a very profound and individual manner. In 2007 a partnership contract was signed with a rehabilitation centre specialised on the specific professions of the company and with the pension insurance which bears the costs of disability retirement. With the employee’s consent, the company doctor can now prescribe a stationary treatment at the centre, shorten the employee’s waiting time, ensure the quality of treatment, and support the employee during treatment. All these projects are carried by the idea of maintaining a win-win-situation for the employee and the company. 202

Table 1.

FBC VPG

WAI

Table 2.

Finnish Bus Company (FBC)/VHH PVG Corporation (VPG): Mean WAI/SD/number/age if available/by occupational groups and WAI classification. Office

Supervisors

Repair

Cleaning

Drivers

41.7/4.2/6 43.2/5.3/82/42.4 ys office and supervisors classification

37.2/6.7/9

35.5/7.3/13 43.5/5.1/ 34/44.7 ys

./. 33.2/8.0 7/47 ys

39.2/7 7.9/77 43/5.6/529 46.5 ys

good/same

moderate/good

good/same

Good reasons for company doctors and employees to start a WAI-dialogue.

Mutual benefits using the WAI – items for a dialogue: For the company doctor (CD) 1. CD gets an employee centred view. 2. What does it mean for this person?! 3. Getting to know the minor social back- ground (family and friends) 4. Getting a more holistic view of the potential of the person. 5. Better and deeper understanding: How can I support the person best? 6. More joy at work.

For the employee 1. Own standing gets more clear. 2. What can I do for my work ability? 3. I am not alone. Where do I get personal support? 4. Evaluation of the actual situation (personal strength and deficits). 5. Ranking and acting: What can I do, what could the company do? 6. Comfort: CD listens, understands and explains well.

4.2 Work and Work conditions Since 2001 the VHH company was awarded prizes twice by the Hamburg occupational health and safety authorities for its health and safety standards. In the repair and cleaning departments, high ergonomic standards support the aging workers in staying at work successfully, and all age groups are enabled to prevent musculoskeletal diseases. The physical work load has been reduced by the use of carriers (a young woman can easily change a bus tire or brake lining), and mechanics and workers have appropriate possibilities to manage their work breaks, organise the order of tasks, and a satisfactory choice of methods, work pace and amount of work (Ilmarinen, 2006). The number of occupational accidents is below the German average. Workers in the office as well as in the repair and cleaning department enjoy satisfactory working hours (opinion poll 2007), but the technical department (repair and cleaning) has the highest number of workers with moderate or poor work ability. To date, the number of participants in the study is low and hence the results can not be considered representative (Table 1, chapter 2.2, 41 out of 170). Therefore there will be a specific assessment of this issue in 2008. It was mentioned already (section 2.2.3) that tight schedules, workweeks of more than 40 hours, irregular day-time work and two-shift or night shift work have become a problem for the bus drivers during the last years. 10% of the bus drivers regard their work demands as primarily physical, 51% as mental and 44.5% as both mental and physical. With increasing age there is a definite (Pearson .211∗∗ , mean age ANOVA ,000) change in this view towards a mixed work demand. Long services3 , overtime and tight schedules are experienced as more physically and mentally demanding. 3 Service

means a mixture of driving time with minor and major breaks. Every service has a certain structure that is created by occupational or traffic law and the organizer, whose possibilities are limited by public demands and costs.

203

46

WAI-index of VHH PVG bus drivers compared with night shift workers with different work loads

Mean WAI-index (points)

44 42 40

Night shift workers (N
368)

38

Night shift workers without firefighters (N
271) VHH PVG CG bus drivers (N
522)

36 34 32 30 Total

Figure 3.

Up to 39 40 to 44 45 to 49 50 to 54 years years years years Age groups

55 plus years

Comparison of WAI scores of VHH PVG bus drivers (2002–2007) with night shift workers (1996–1997) with different work loads (Tempel 2002), data by permission of GP Forschungsgruppe, Munich, 2002.

34% of the drivers want spare time as compensation for overtime work, 32% want money and 31% want a mix of these two according to their needs. It is reported that drivers have begun to accept extra services only when they are guaranteed additional time off. Figure 3 (above) may give an impression of this. Comparing the work ability of bus drivers with “real” night shift workers with different workloads such as fire fighters, the chemical or paper industry points to important and helpful differences. Salutogenic factors have been described and the mean age of the sample is between 38 and 46 years. The samples of night shift workers were not representative and the participation rate was low. All this put together shows that more important differences remain: The physical workload for bus drivers is low, the risk is rather one of monotonous underchallenge, but autonomy at work is rather high compared to other shift workers (Karasek and Theorell, 1990), and finally there is a successful workplace health promotion process which contributes to the actual findings. To date (N = 101), we have found a positive correlation between work ability and contentment with choice alternatives and decision latitude, but there is weak significance and more data is needed to support this. In 2007 we started discussing that even the change of leadership style will not be able to compensate the problems the drivers have (and will have in the future) with increasing age. Bus driving is considered moderate to heavy work (physical, mental and social) by the leadership as well as the drivers, and with increasing age there is a growing need for recuperation: “It is essential for the efficiency of recovery that it can start immediately after a peak load or a heavy burden” (Ilmarinen, 1999, 1997). But in practice there is a tendency to keep up the service until it ends, to go on until the next day off, to extend the working week to the next bank holiday or scheduled vacation. All these coping strategies are helpful and necessary to navigate the actual changes in work demand with which the company is confronted. But they are not a sustainable strategy to cope with the demographic changes. On the contrary, growing fatigue is a severe accident and health risk for both the company and the drivers. So finally a project to study this issue was started in two steps: A pilot study (Amelsberg, and Tempel et al., 2007) was to define scientific standards for creating a service concept that a bus driver can follow until pension age. That means that the services have to be organized within a shift schedule that satisfies modern occupational knowledge (Härmä and Ilmarinen, 1999; Knauth and 204

Hornberger, 1997). Finally, an “alarm system” was created (red: service not acceptable according to occupational science; yellow: problematic result, has to be developed; green: you can go on up to pension age as far as we know, preserve!) and the prevailing services were analysed. After this successful pilot study, the further timetable will be analysed in 2008 as the actual work demand set by the company and the results are discussed with drivers at the different stations. We must learn how they perceive the different services and routes, e.g. with respect to traffic, passengers, or working time. The drivers may hold completely different views about the quality and content of a break. Minor breaks of one to five minutes are highly valuable according to work physiology, but in traffic – there is a common consent on this – they easily “get lost”. And the efficiency of pauses is not the same with regard to physical or mental work demands. The drivers need much more time to cope with mental problems than with physical ones. We will compare the views and ideas of the drivers concerning the 2008 timetable with driving reality: The company uses CAD/AVL (Computer Aided Dispatch/Automatic Vehicle Location) to survey, organise and maintain public transport. The same system can be adapted to check actual driving time by gathering mean scores. The board and works committees have agreed to end the endless discussion about what it is actually possible to drive. So the work demands set by the company will be compared with actual driving time and conditions, and finally be categorized. 4.3 Professional Skills Learning and vocational training have a long tradition in the corporation. There is an own driving school and the participants have to drive various buses with different types of technology. Every bus driver has to spend five hours per year on special subjects such as how to cope with difficult passengers, driving techniques, first aid on the bus, or how to deal with accidents. From 2008 on these procedures are enforced by European regulations. Corresponding rules exist for the whole staff. Among the leadership, learning and vocational training is considered to be a decisive part of maintaining the potential of the employees and workers as well as a contribution to “good quality of life and well-being”. 4.4 Work Community Since 2006 we have been assessing the correlation between work ability and the quality of communication with the supervisor, as well as the possibility of discussing and solving problems with the supervisor and with colleagues. Leadership has the strongest positive or negative influence on a sustainable balance between work demands and individual potential of an employee or worker (Ilmarinen, 1999). Within the context of the work ability concept, our findings seem comparable with the Finnish results that are based on more robust statistical methods (Ilmarinen, 2006). There are different scores for the bus drivers: The quality of communication has no significant correlation with work ability, and the correlation with the possibility to discuss and solve problems is less significant (Pearson −.363∗∗ , ANOVA .003, R2 .132). This can be taken as an early indicator that the climate for the drivers might be changing for worse.

5 COMPANY CLIMATE AND PUBLIC POLICY Since 2002 we have been trying to gather more details about the existing climate within the whole VHH PVG Corporation. Important components of the climate are “the general impression the company makes on the staff, the colleagues, the leadership, information and participation, the works committee and company’s payment, fringe benefits and appreciation” (Rosenstil and Bögel, 1992). To assess the climate, we use four questions about satisfaction with pay, the climate, social welfare 205

50

Mean WAI-index and communication with supervisor or possibility to discuss and solve problems (N
131) Quality of communication with supervisor (a)

Mean WAI-index

45

Discuss and solve problems with collegues and supervisor (b)

40 35 30

Po or

In di ffe re nt M od er at e

G oo d

go od Ve ry

To ta l

25

Categories

Figure 4. WAI of employees and leadership: Quality of communication and possibility to discuss and solve problems are highly significantly correlated. (a: Pearson −,278∗∗ . ANOVA .010; b: Pearson −.418∗∗ , ANOVA .000, R2 .175).

70

General satisfaction - VHH PVG Company Group

60

Percentage

50 Very satisfied Rather satisfied Rather dissatisfied Very dissatisfied

40 30 (n=63) 20 10 0 Payment

Figure 5.

Company climate Social welfare benefits Categories

General leadership style

General satisfaction of the staff with pay, company climate, social welfare benefits and general leadership style.

benefits, and the general leadership style. Figure 5 shows first results that should be followed up during the next years – nevertheless they are valuable already as early indicators. About 65% of the employees are “rather satisfied” with their pay and 35% are not. Especially younger drivers and newcomers receive lower wages. 79% are “very and rather satisfied” with the company climate, 83% with the social welfare benefits and 79% with the general leadership style. The perception of company climate is weakly correlated with work ability (Pearson −.284∗∗ , ANOVA .0014, R2 .081). 206

Bus companies experience high public pressure and are often exposed to discussions about costs, and many drivers report that they no longer feel appreciated by the public and the passengers. Since 2006, we have been using the question “I experience the public situation as supportive for my work ability” (very much so, rather much, indifferent, rather not, not at all). 42% answer “not at all” (7%) or “rather not”. If we look at perceived public support for work ability, there is a U-shaped distribution which might be interesting: Those bus drivers who are “indifferent” have the lowest mean WAI-index (41.4 points), and both the group which feels supported and the one which does not display much higher mean scores. In our WAI-dialogues we find two coping strategies for this issue: “At work I don’t care about politics and don’t want to think about it” or “I mind my own business”, and “hard times need strong efforts to get along”. 6 CONCLUSIONS AND OUTLOOK The Work Ability Concept has proved helpful for structuring the work place health promotion process of a company. Based on a scientific and practical understanding, the steering committee can analyse the past and present situations and thus support the company (management and works committee) in structuring the future. The use of the WAI together with the concept creates a deeper understanding of the current state of balance between the “Work Quality and Productivity” the enterprise needs and the “Quality of Life and Well-Being” of the staff. The different measures taken to promote and maintain work ability can be evaluated according to their influence on the WAI-index or the different WAI-items. As for the VHH PVG Corporation, the actual balance seems to be somewhat instable. The positive results of the work place health promotion process that was started after a period of change and reduction of staff can explain the present situation. To ensure this for the future, three major tasks must be taken on: Continuing the appreciative exchange of experience and integrating the new leadership style into the company climate. Analysing the services and successfully implementing a structure of driving time and breaks that is supportive for the drivers. Reducing the overtime work in order to reduce the risk of premature invalidity with growing age. ACKNOWLEDGEMENT The presentation in Hanoi and the article were funded by VHH PVG Corporation. Jens Peter Becker, Miriam Krieger, Jan Krohn and Heike Tempel gave me valuable support to write the article in English. REFERENCES Amelsberg, S., and Tempel, J. et al., (2007). Alters- und alternsgerechte Schichtplan- und Dienstgestaltung unter schwierigen wirtschaftlichen Bedingungen zur mittel- und langfristigen Zukunftssicherung der VHH PVG Unternehmensgruppe – Abschlussbericht. Hamburg und Bremen, VHH PVG UG und Universität Bremen – ZeS. Ell, W., (1995). Arbeitszeitverkürzung zur Belastungsreduzierung älterer Arbeitnehmer im öffentlichen Personennahverkehr – 10 Jahre Erfahrung aus den Interventionsmaßnahmen in den Verkehrsbetrieben in Nürnberg. Alt, erfahren und gesund. Betriebliche Gesundheits-förderung für älterwerdende Arbeitnehmer. H. G. l. Rudolf Karazman, Irene Kloimüller, Norbert Winker. Gamburg, Verlag für Gesundheitsförderung G. Conrad. 1: pp. 160–170. Frevel, A., and Geißler-Gruber, B. et al., (2006). Dialoge verändern, Partizipative Arbeitsgestaltung – Voraussetzungen, Methoden und Erfahrungen für eine zukunftsfähige Arbeitsforschung. Köln, Kölner Wissenschaftsverlag.

207

Geißler-Gruber, B., and Geißler, H., (2000). Von den Gesund (et) en lernen. Der Nahverkehr – Personen- und Güterverkehr in Stadt und Region 10: pp. 1–6. Geissler-Gruber, B., and Geissler, H. et al., (2004). Appreciative exchange of experiences: an “empowerment” tool for and by managers. Applied Participation & Empowerment at Work. C. R. Johansson, A. Frevel, B. Geißler-Gruber and G. Strina. Lund, Studentlitteratur: pp. 211–222. Geißler, H., and Bökenheide, T. et al., (2003). Der Anerkennende Erfahrungsaustausch – Das neue Instrument für die Führung. Frankfurt / New York, Campus. Härmä, M. and Ilmarinen, J., (1999). Towards the 24-hour society – new approaches for aging shift workers? Scand J Work Environ Health 25(6, special issue): pp. 610–615. Hasselhorn, H. M., and Freude, G., (2007). Der Work Ability Index – ein Leitfaden. Bremerhaven, Wirtschaftsverlag NW. Ilmarinen, J., (1999). Ageing Workers in the European Union – Status and promotion of work ability, employability and employment. Helsinki, Finnish Institute of Occupational Health, Ministry of Social Affairs and Health, Ministry of Labour. Ilmarinen, J., (2006). Towards a longer worklife! Ageing and the quality of worklife in the European Union. Jyväskylä, Gummerus Kirjapaino Oy. Karasek, R. and Theorell, T., Eds. (1990). Healthy Work. Stress, Productivity, and the Reproduction of Working Life. USA, BasicBooks. Knauth, P., and Hornberger, S., (1997). Schichtarbeit und Nachtarbeit. München. Rosenstil, L. v. and Bögel, R., (1992). Betriebsklima geht jeden an! München, Bayerisches Staatsministerium für Arbeit und Sozialordnung, Familie, Frauen und Gesundheit. Tempel, J., (2002). Der Einfluß der Nachtarbeit auf den Gesundheitszustand. Ernährungszustand von Nachtschichtarbeitern. D. Korczak, S. Klotzhuber, J. Tempel, C. Eggerdinger and G. Schallenmüller. Bremerhaven, Wirtschaftsverlag NW. S 68. Tempel, J., (2004). The Work Ability Index (WAI) is an useful instrument to structure the client/patient-doctorrelationship in occupational medicine. Proceedings of the 1st International Symposium on Work Ability – Past, Present and Future of Work Ability. J. Ilmarinen and S. Lehtinen. Helsinki, FIOH. 65. Tempel, J., and Giesert, M. et al., (2005). Arbeitsfähigkeit 2010: Von 16 bis 65 in einem Unternehmen! Abschlussbericht zum ABI-NRW-Projekt. Düsseldorf, IQ-Consult gGmbH. Tuomi, K., and Ilmarinen, J. et al., (1998). Work Ability Index. Helsinki, K-Print Oy Vantaa, Finland. Vorndran, I., (2006). Unfallgeschehen im Straßenverkehr 2005. R. V. Statistisches Bundesamt, Referat V C “Verkehr”, Pressestelle, Wiesbaden. Wakula, J., (2007). Belastungs-Beanspruchungskonzept. Lexikon Arbeitsgestaltung. K. Landau. Stuttgart, Gentner Verlag – ergonomia Verlag: pp. 305–306.

208

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Prolonging working life in intellectual work Ülo Kristjuhan Department of Working Environment and Safety, Tallinn University of Technology, Tallinn, Estonia

ABSTRACT: A later retirement age is typical of intellectual workers, e.g. in the case of university teachers and researchers. The ability to do part of the work at home, including telework, is probably part of the reason. During 2005–2006 questionnaires were designed and a research of telework was carried out in Tallinn University of Technology. Among 260 academic respondents, telework usage did not vary significantly between the age of 40 and 70. Employees younger than 30 years were using it less than the older. Computer and Internet experience might be a better predictor than age. Data shows that doing some work at home, 10–20 hours per week, is healthier and is associated with less stress and hypertension. There is also a tendency to report less visual fatigue. The option to work one to three days per week at home is the best.

1 INTRODUCTION At present many developed countries including Estonia are entering a workforce situation never experienced before. There is a shortage of young workers as birth rates are low. The average age of manual and professional workers is rapidly increasing. Compared to previous times, older people are working or seeking employment more often (see: EU job situation improves . . . , 2006). There are several causes: • Good pensions are becoming more difficult to receive. Pension replacement rate is decreasing. • In previous times the life expectancy increased mainly because of the decreasing mortality of young people. At present the situation is different. Life expectancy is increasing as a result of years of good health. The ageing workforce will pose new challenges to enterprise and organisational management. These challenges are different in the cases of blue-collar and white-collar workers. This article considers intellectual workers. Older white-collar workers are physically weaker compared to younger ones and many of their physiological and psychological characteristics are worse. However these workers mostly compensate the deterioration of their functional capacities with knowledge and competence. Older workers have more accumulated (crystallized) knowledge. Therefore scientific research has shown that their productivity is practically the same as the productivity of younger persons, and sometimes higher. There is a need to transfer both knowledge and experience to younger generations in society and to create appropriate conditions for this process. There are also older intellectual workers who finished their studies many years ago and need training courses. However this training costs less than preparing new young specialists. Older intellectuals want to continue teaching and research in the university. They want to have more privacy and autonomy within their organizations. They sometimes need extra time for recreation. They don’t want to have a long way from home to the workplace. 209

2 TELEWORKING AS A SOLUTION Teleworking (working from home, satellite office or a telework centre near home), full-time or part-time, is one possible solution, especially in the case of university teachers and researchers. Teleworking contributes to using modern information technology, and improves the use of accumulated knowledge and experience. Older intellectual workers who are partly working from home have an extra 1–2 hours per day of free time (decreasing travel times), affording time to be with their families. These workers can thereby avoid the impact of negative social changes on their personal well-being in later life and therefore better maintain health. Working from home has many different positive effects for society. While cars, buses, trams and trains need to be used, it is safer and healthier for people to change their thinking, changing the transport idea by the idea of telecommuting. Telecommuting provides many benefits: avoiding road accidents, protecting the environment, and conserving energy. Sometimes working from home is criticized because of the possibility of overworking. Factually, this possibility always exists. Even without telework most intellectuals do not limit their working by eight hours per day. Sometimes these workers are not thinking about their work during traditional working hours but during traditional leisure hours. There are a lot of studies about telework. However, in the case of older intellectual workers telework is often a new phenomenon. It is studied superficially, especially when it regards health issues. We have recently carried out research in telework at Tallinn University of Technology. 3 THE SITUATION IN ESTONIA Here are some factors that push older people to work and use computers in Estonia: • The unemployment rate is low (less than 5%). The employment rate for olderworkers has rapidly increased during the last few years. • Wages and salaries are relatively low as in most former Soviet Union countries but they are also rapidly increasing. In the second quarter of 2007, the average monthly gross wages and salaries were 11,549 kroons (738.1 EURO). The average monthly old-age pension was 3,515 kroons (224.6 EURO), about 30% of average wages and salaries (see: Main Social and Economic. . . , 2007). • Estonia is very intensively introducing and using modern information technology everywhere. Personal identity cards are used instead of tickets in public transport. The Internet is widespread as in the most developed Western countries and teleworking has become a standard element in the Estonian workforce. Computers with data projectors appeared in the laboratories and cabinets of Tallinn University of Technology in the 1980s and in auditoriums in the 1990s. There are many computer training and education courses and seminars (virtual learning environments WebCT, Moodle, etc) for teachers at Tallinn University of Technology and virtual learning environments are widely used in practice at present. The average age of academics at Tallinn University of Technology is similar to most western universities, but the University has a long tradition of using an older workforce and offering opportunities to work from home. The University encourages age diversity. Even during the Soviet period academics were permitted to work part-time at home according to negotiations between employers and employees. At present there is no mandatory retirement in the University. Currently the oldest researcher is 85 and the oldest associate professor is 79 (born 1922 and 1928, respectively). Most academics retire in their late 60s and 70s. Studies in Tallinn have shown that productivity is highest among teachers and researchers 56–65 years old (Kristjuhan and Taidre, 2005). There is no decrease in work efficiency among older academics in their 50s and 60s as they are not waiting for retirement. 210

4 RESEARCH ON TELEWORK AT TALLINN UNIVERSITY OF TECHNOLOGY 4.1 Method We carried out a survey of telework at Tallinn University of Technology during 2005–2006 (Arvola, 2007). A questionnaire was developed that consisted of 18 questions and was available on paper. There were open questions as well as multiple choice questions. Academics were asked to answer the questions about different characteristics of work, health indices and complaints in different parts of the body, and conditions at home. 260 university teachers and researchers completed the questionnaire. There were questions about the number of hours of telework and attitude toward it, about different factors that compared the working environment at home to the University office, including contact with chemicals, and about the negative aspects of telework. The questions were about teachers’ mastery working with a computer and communication tools, about the size of their family, their income and age, about the number and pages of publications and hours spent on scientific work (working with literature, planning and carrying out the research). Data about time spent travelling between the University and home and about income were also included. Questionnaires were sent to academics by e-mail or by post. Participation in the research was voluntary. All data collected were subjected to statistical analysis. 4.2 Results and discussion There were significant differences in responses regarding the use of telework. Telework usage was 44.3% when it was used 1–10 hours per week and 30.2% when used 11–20 hours per week. 8.0% of academics did not use telework at all. Only 13.2% of respondents used telework more than 30 hours per week. Consequently the common fear of overworking as a result of the unlimited possibility of working at home was not confirmed. Telework usage did not vary much by gender. The research showed that personnel younger than 30 years used teleworking less. Telework usage did not vary significantly between the ages of 40 and 70, and did not depend on time of travel between home and the workplace. Computer and Internet experience, not age, probably predicts telework usage. The research showed that the academics’ stress level was significantly lower when working outside of the employer’s workplace (mostly at home) instead of working at the office. A low stress level is very important to older intellectual workers. There was also less hypertension among teleworkers compared to non-teleworkers. Subjects were better able to focus on work at home. There was more privacy. Visual fatigue was an overall problem, but it was more widespread in non-teleworkers. There were minimum complaints of visual fatigue in the case of those using telework 21–30 hours per week. Most data regarding health and productivity characterized part-time teleworking 10–20 hours per week as the best variant. It appears that maintaining health is easier in the conditions of home. General recommendations about older workers workplaces described in detail by researchers (Ilmarinen, 2005) are easier to put into practise at home than in the office. The ability to plan and regulate work is better at home. Telework offers to the worker the opportunity to design an office environment at home that best fits his/her needs. This takes into account the timely recognition of body signals in order to prevent health disorders, flexible, individual, and ergonomic working hours, individual rest breaks and periods for movement, relaxation and eye exercises, dynamic sitting behaviour (regularly changing body positions), and individual workstation organization, including lighting. The ability to postpone the aging processes by means of a beneficial environment is a common understanding among biogerontogists at present. Aging is the gradual change in the structure and function of organisms that occurs for intrinsic processes and extrinsic influences and increases the probability of death. There is modest age postponement in developed countries, the ages of the concrete probabilities of death increasing every year. It is likely that aging will be postponed to some extent for those working in the beneficial conditions of freedom and home (see Kristjuhan, 211

2007, 2006). Also, using computers and the Internet improves information about health issues of the computer user and opens new horizons for creating optimum working environments and computer-assisted instruction (Kristjuhan, 2004). The knowledge and experience of intellectual workers should be better used. Using experienced teachers in scientific work and as supervisors of doctoral students has some advantages for the institution, and also provides more years of good health in academics. Complex solutions are the best. One of them is the ‘flexicurity’ approach. The concept of ‘flexicurity’ attempts to find a balance between flexibility for employers (and employees) and security for employees. Flexicurity combines active labour market policies, flexible contractual arrangements, lifelong learning and modern social protection systems. It can help to confront the challenges of globalisation and demographic aging. In the future life probably will not be divided into three blocks – hard studying, earning and enforced leisure – but will show a mixture of education, work and leisure over the course of life.

5 CONCLUSIONS AND RECOMMENDATIONS Our studies show that: • Older intellectual workers should have the possibility to work without any mandatory retirement age. • Telework should be an option for older academics. • Working partly at home, around 10–20 hours per week, can maintain better health in intellectual workers. • Using older academics in research, instruction and as experts is especially important for using accumulated knowledge and experience.

ACKNOWLEDGEMENT I am grateful to PhD student René Arvola and former bachelor student Mari Arnover for carrying out the survey at Tallinn University of Technology. REFERENCES Arvola, R., (2007). New data of working from home (Research in case of intellectual work). Telework as Solution for Senior Workforce. Ü. Kristjuhan & R. Arvola, eds., (Tallinn University of Technology Press), pp. 13–27, ISBN 978-9985-59-701-9. EU job situation improves, but new reform push needed to hit 2010 target. Europa. Rapid Press Releases. Brussels, 6th November 2006. Ilmarinen, J., (2005). Towards a Longer Worklife. Ageing and the Quality of Worklife in the European Union, (Helsinki, Finnish Institute of Occupational Health). ISBN 951-802-685-8. Kristjuhan, Ü., (2007). Vista of youth maintenance and body sensations. Ü. Kristjuhan & R. Arvola, eds., Telework as Solution for Senior Workforce, (Tallinn, Tallinn University of Technology Press), pp. 5–12 Kristjuhan, Ü., (2006). Soft strategies for postponing aging and prolonging human life. Rejuvenation Res. 9: pp. 329–332. Kristjuhan, Ü., (2004). Computer as a tool in postponing aging. First Central European International Multimedia and Virtual Reality Conference. Veszprém, Hungary, 6–8 May 2004, Veszprém University Press, pp. 95–100, ISBN 978-9985-59-701-9. Kristjuhan, Ü., and Taidre, E., (2005). Workability and health of older academics. Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers. Elsevier, pp. 101–105. Main Social and Economic Indicators of Estonia (e-publication). Monthly bulletin, Statistics Estonia. July, 2007.

212

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Ergonomic and safe design of railway vehicles for elderly and handicapped people Manfred Rentzsch & Denis Seliger Research Department, IAS Institute for Occupational and Social Hygiene Foundation, Berlin, Germany

ABSTRACT: According to COST 335 disabled people represent around 13% of the population of Europe. This is approximately 63 million people. The share of the elderly in the total population of Europe is expected to rise from 21% now to around 31% by 2020, and to around 34% by 2050. The starting point of the investigation was to analyse the current access situation for disabled and elderly persons in passenger trains, considering the relevant regulations, standards, operation instructions as well as the large volume of research work that has already been carried out at the international and national levels. A user-focused analysis based on an interview manual and questionnaires was carried out. The results of the user-focused analysis were used as inputs for the design guidelines. Altogether 67 subjects (33 females and 34 males) with different handicaps took part at the tests. The mean age of the subjects was 53 years. The distribution of age of the test persons is presented in Table 2. The test duration was about three hours. Approximately 63% of the test persons evaluated the accessible design as good or very good. Regarding the influence of age it can be summarized that there is not a big difference between the two age groups. That means handicapped persons as well as elderly persons have assessed the quality of the product design as positive. There is still the need for further improvements to enter the train, to get the necessary information within the train and to operate the emergency and service facilities without any problems and thus to achieve a barrier free design for all. This concerns e.g. to increase the character size and the contrast between the characters and the background on the touch screen, and to increase the contrast within the entrance and compartment area and the toilet as well as environment related elements like handholds, handrails, buttons for soap, water and drier.

1 INTRODUCTION This paper is an output of an European collaborative project financed by the European Commission with experts of railway operators (DB, SNCF and Trenitalia), manufacturers/suppliers (Bombardier, Siemens, Alstom), associations of persons with disabilities and scientific institutions (FAV Berlin, IAS Berlin, VUT Vienna, UPC Barcelona) from six European countries taking part. The project is coordinated by UNIFE Brussels. The objective of the project is to optimise the access, the entrance vestibule, information systems inside and outside the train, emergency facilities, toilet with all conveniences and the additional test arrangements regarding push buttons, steps and emergency equipment as an important segment of a barrier free travel chain. For this reason physically, visually and hearing impaired people, persons of small stature, elderly people and parents with small children were included from the very beginning (Seliger and Rentzsch 2006). According to COST 335 (1999) disabled people represent around 13% of the population of Europe. This is approximately 63 million people. The share of the elderly in the total population of Europe is expected to rise from 21% now to around 31% by 2020, and to around 34% by 2050. The median age in Europe is growing up from 37.7 in the year 2000 to 49.5 by 2050 (Höhn, 2002). 213

VUT, IAS foundation, UPC, FAV

Transfer of knowledge

User focused analyses

Design

Functional

Guidelines

Mock-up

Test criteria

Test programme

Derivation

of standards

Innovative phases of the project partners

Figure 1.

Basic structure of the methodology.

In Germany 6.6 million people are handicapped (3.5 million males and 3.1 million females). 75% are older than 55 years and 52% older than 65 years (Statist. Bundesamt, 2004). Handicapped people favour public transport over private cars. Therefore the vehicles of the local public and long distance traffic have to be designed in such a way that they can be used from all the interested persons in the sense of a barrier free travel chain including elderly and handicapped people.

2 METHODOLOGY 2.1 Basic structure Figure 1 illustrates the basic structure of the methodology used in the project. The starting point of the investigation was to analyse the current access situation for disabled and elderly persons in passenger trains, considering the relevant regulations, standards, operation instructions as well as the large volume of research work that has already been carried out at the international and national levels. A user-focused analysis based on an interview manual and questionnaires was carried out. The main structure of this manual is listed below: • • • •

Classification of people with reduced mobility and their aids. Ergonomic design of the access area. Information. Emergency facilities.

The main components and items of the interview manual are summarized in Table 1.

2.2 Test equipment The results of the user-focused analysis were used as inputs for the design guidelines. A computer version of a mock-up (see Figure 2) and finally a functional mock-up in the scale 1:1 were then produced. Further, two additional elements were included in the tests. The first was a test panel consisting of different kinds of push buttons to open and to close the doors and the other was a special test stand consisting of steps with different depths and heights. 214

Table 1.

Main components and items of the interview manual.

Components

Items

Ergonomic design of the access area

– General aspects – Access doors and its controls – Horizontal and vertical gap

Information

– Acoustic signals – Visual signals – Emergency brake – Alarm signal – Toilet – Changing table

Emergency facilities Service facilities

2600 2500

900

2400

2150

1880

– Boarding aid devices – Access room – Lighting, colour – Materials – Tactile signals

700

1900

1900

728

2913

955 900

4650

Figure 2.

8008

CAB (floorheight 1360)

REG/HST (floorheight 800)

Eupax Mock-up layout.

2.3 Subjects Altogether 67 subjects (33 females and 34 males) with different handicaps took part at the tests. The mean age of the subjects was 53 years. The distribution of age of the test persons is presented in Table 2. The test duration was about three hours.

3 RESULTS The general assessment of the handicapped accessible design depending on age is presented in Figure 3. Approximately 63% of the test persons evaluated the accessible design as good or very 215

Table 2.

Number of subjects and distribution of age.

Number of subjects

Distribution of age

30 15 12 5 4 1

20–49 50–59 60–69 70–79 80–89 >89

Mock-up general

100 90 80 70 60 50 40 30 20 10 0

Bad Fair Good Very good

All (n
63)

 50 (n
28)

 50 (n
35)

Handicapped accessible design p
0.05

Figure 3.

Product design for handicapped and elderly people.

good. Regarding the influence of age it can be summarized that there is not a big difference between the two age groups. That means handicapped persons as well as elderly persons have assessed the quality of the product design as positive. Finally the subjects were asked to distribute 100 points to the four different aspects and thus setting priorities for the barrier free design. The aspects were: Access to the train/entrance area, compartment area, use of control elements (e.g. push buttons) and perception of information (outside and inside the train). The results of prioritization of different aspects depending on age are shown in Figure 4. That means for all test subjects the barrier free access to the train is the most important in handicapped accessibility design. The other three parts are more or less equally ranked at about 20% each. For people older than 50 years the perception of information has the second priority due to visual and hearing problems as these kinds of impairments strongly increase with rising age.

4 CONCLUSIONS There is still the need for further improvements to enter the train, to get the necessary information within the train and to operate the emergency and service facilities without any problems and thus to achieve a barrier free design for all. This concerns e.g. • To increase the character size and the contrast between the characters and the background on the touch screen. • To increase the contrast within the entrance and compartment area and the toilet as well as environment related elements like handholds, handrails, buttons for soap, water and drier. 216

45 40 35 Access to the train/ entrance area

30 25

Compartment area

20

Use of control elements (e.g. push buttons)

15

Perception of information (at and in the train)

10 5 0 All (n
63)

Figure 4.

 50 (n
30)

 50 (n
34)

Prioritization of handicapped accessible design and age.

REFERENCES COST 335, (1999). European Commission, Directorate General Transport; Passengers’Accessibility of Heavy Rail Systems, Final Report of the Action; Office for Official Publications of the European Communities; Luxembourg Höhn, Ch., Alterung der Bevölkerung, (2002). http://www.berlin-institut.org/pages/ buehne/buehne_beventw_ hoehn_alterung.html Seliger, D., and Rentzsch, M., (2006). Barrier free design of the access area and the information system of railway vehicles, 14th International Symposium EURNEX – Zel “Towards the competitive rail systems in Europe”, Proceedings edited by Jiri Zahradnik, Pter Nagy and Michal Mikulas Zilina, Slovenska Republika ISBN 80-8070-552-6 Statistisches Bundesamt, (2004). Wiesbaden, http://www.destatis.de

217

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Female workers’ superior peculiarity and consideration for aging for the activation of “super-advanced age and fewer children” society Koki Mikami Department of Humanity and Social Science, Hokkaido Institute of Technology, Sapporo, Japan

Kenichi Iida Department of Production System, Hokkaido Industrial Research Institute, Sapporo, Japan

Masaharu Kumashiro Department of Ergonomics, IIES, University of Occupational and Environmental Health, Kitakyushu, Japan

ABSTRACT: Taking Japan’s “super-advanced age and fewer children” society into consideration, one measure to promote the economy of Japan is efficient use of women. In order to contribute to the activation of women’s work, this paper describes 1) women’s superior peculiarity compared with men’s, 2) the influence of aging and 3) consideration points for women’s work, using the results of experimental research and a case study conducted to solve these KAIZEN problems. Keywords: KAIZEN

aging, female workers’ superior peculiarity, super-advanced age and fewer children,

1 INTRODUCTION Japan cannot avoid a “super-advanced age and fewer children” society. In 2013, it is estimated that one out of four people will be over 65 years old. And in 2035, one out of three people will be over 65 years old. The Employment Stability Law for the Aged was amended and put into effect on April 1, 2006, and this obliges employers to continue employing their employees until the age of 65, the final starting age of pensions. And then, since April 2007, the government has promoted the expansion of continued employment until the age of 70 for employees wishing to continue working. Taking Japan’s “super-advanced age and fewer children” society into consideration, one measure to strengthen the economy of Japan is efficient use of women. In order to contribute to the activation of women’s work, the authors made a study of women’s superior peculiarity, the influence of aging and consideration points for women’s work, using the results of two kinds of experimental research and a case study.

2 METHODS 2.1 Women’s superior peculiarity compared with men’s (experimental study) Women’s work adaptability in four tasks of visual information processing work was examined compared with men’s from the viewpoints of workload and efficiency in order to appreciate women’s work ability and make the best use of it in the workplace. 219

2.1.1 Experimental conditions 1) Task Classification Task A: An eye-hand coordination task with a microscope Task B: A wave inspection task requiring both hand movement and visual judgment with a VDT Task C: A task of pattern recognition requiring eyesight, thought, and judgment with a VDT Task D: A task of choice of the correct one from the three options requiring knowledge and thought with a VDT 2) Subject: 10 females, average age 20.0 ± 1.0, 10 males, average age 0.9 ± 0.5 3) Work control: Each subject’s own pace 4) Work posture: Each subject adjusted the height of his or her chair. 5) Composition of the work periods, breaks, and measuring points of each task.

M1

1st work Period 70 min. M2

M3

Break 5 min.

2nd Work Period 90 min. M4

M5

3rd Work Period 120 min. M6

Lunch break 45 min.

4th Work Period M7 105 min.

M8

Break 5 min.

6) Measuring items: (1) VRT (Visual Reaction Test), (2) CFF (Critical flicker Fusion frequency), (3) Near-point Accommodation. (4) Subjective Feelings of Fatigue, (5) Sites of Physical Fatigue Symptoms These items were measured eight times, before and after each work period. (6) Subsidiary Behavior, (7) Heart rate, (8) Performance, (9) Error Rate. These items were measured during work. 2.2 The influence of aging (experimental study) The characters of the work efficiency and workload of middle-aged or elderly females who had no job for a year or more were examined and compared with the results for younger females in task A and task B. Experimental Conditions (1) Task Classification: Task A, Task B (2) Subject: 10 middle-aged or elderly females, average age 48.0 ± 3.4 10 younger females, average age 20.0 ± 1.0 (3) The other conditions were the same as the above 2.2. 2.3 Consideration points for women’s work (case study) Target company: The main products were various kinds of noodles (Factory 1) and box lunches (Factory 2). The research was performed based on the Ergoma Approach (M., Kumashiro, 1987, K., Mikami, et al., 1997, and K., Mikami, 2002)

3 RESULTS 3.1 Female workers’ superior peculiarity (experimental study) 3.1.1 Females’ work adaptability to four visual information-processing tasks Females’ fluctuations in the psycho-physiological functions did not show significant lowering compared with those of males. Although in some tasks their complaint rates of subjective feelings of fatigue about the local muscles were higher than those of males, there were no significant 220

Table 1.

Relative evaluation on the basis of males.

Fluctuation in function VRT CFF N.P Heart Rate Subsidiary Behavior Subjective Feeling of Fatigue Sites of Physical Fatigue Symptom Ocular Discomfort Performance Error Rate Finished Parts

Task A

Task B

Task C

Task D

N S I N S N N

N S N S S N I

N S N N S N I

N N N N S N N

N S N S

N N S S

N N S S

N N S S

No Superiority 3 1 4

difference 4 1 3 3 4 2

1 3 4

Inferiority

1

2

4 3 1

N: no difference between genders; S: females are superior to males; I: females are interior to males

differences observed in subjective feelings of fatigue and ocular discomfort, and females had less behavior of escape from work. In respect to work efficiency, there was no significant difference in performance between females and males, but the numbers of females’ inferior areas was generally smaller. As a result, in all the tasks the numbers of finished areas females made a day was larger than those of males, which clarified females’ higher adaptability to this kind of work. 3.1.2 Females’ predominant characters in the visual information-processing work Females’ lowering of the cerebral cortical activity level was less and their durability of the cerebral cortical activity level was higher, compared with those of males. In all the tasks, the females’ occurrence rates of subsidiary behavior were lower, and their behavior of escape from work was less. In thought work, such as the task requiring thought and judgment and the task requiring knowledge and thought, females showed lower work efficiency at early stages of work, but the combined effect of their marked learning effect in the working process, higher durability of the cerebral cortical activity level, and less behavior of escape from work led to an increase in performance] and decrease in inferior areas, which made economical, effective production possible. 3.2 Effects of aging (experimental study) In Task A and Task B, many of the psycho-physiological functions of the middle-aged or elderly females showed lower values than those of the younger females, but no significant decrease was observed over the course of time. Subsidiary behavior and complaints of feeling of fatigue increased in the middle-aged or elderly females, which showed their difficulty in maintaining concentration on the work. The above results show women have excellent ability compared with men, but that they cannot avoid the effects of aging. Therefore, it is important to conduct KAIZEN to best utilize the ability of female workers. 3.3 Consideration points for women’s work (case study) A study on KAIZEN for female workers at a food processing factory for the activation of “superadvanced age and fewer children” society 3.3.1 Aspect of the target company The main products were various kinds of noodles (Factory 1) and box lunches (Factory 2). 221

Figure 1. The main products of this company.

Figure 2. An electronic work manual with animated cartoons.

Many of their products were daily foods, and a timely and flexible production system was required. Part-time workers with long experience played the most important role in this company. However, 60.5% of the 1st factory workers answered in the questionnairing conducted there, “I don’t want to be a full-time worker.” The main reasons were “That will reduce my free time” (42.3%) and “I cannot work full time for family reasons such as housework and childcare” (36.0%). At the 2nd factory 86.8% answered in the negative. For the continued employment of female part-time workers who had the key to the existence of the food company, we made a study of the creation of a healthy workplace, which would be easy to work in and accommodate the needs of the changes in the female lifecycle including housework. 3.3.2 Practice of work management 1 We introduced a new part-time employment system: The personnel department in the head office controls part-time employment directly and decides whom to employ after giving applicants a sufficient understanding of the company’s policies and work substance. Workers wishing to continue working can work until the age of 65. To utilize part-time workers efficiently, quick introduction of clear and effective education/training was indispensable. So, we made an electronic work manual with animated cartoons and a work standards book which would give a sufficient understanding of the substance of their work at the time of employment or orientation. 3.3.3 Practice of work management 2 Internal enlightenment lectures were held for a strong company composition that is the base of continued employment. The following five kinds of lectures were offered. 1) Necessity of higher productivity 2) Women’s superior peculiarity 222

3) IE basics 4) Production management and JIT production method 5) Standard work composition drills

3.3.4 Practice of work management 3 We conducted 11 kinds of support apparatus KAIZEN. 1. KAIZEN for keeping and cleaning cutters

Before

Figure 3.

After

Making of a work table for keeping and cleaning cutters.

2. KAIZEN for vacuum refrigeration

Before

Figure 4.

After

Introduction of an automatic stop device, a buzzer, and pilot lamps for vacuum refrigeration to avoid careless mistakes.

3. KAIZEN for putting noodle dough in the refrigeration machine

Before

Figure 5.

After

Making of an aluminum push car for vacuum refrigeration and guiding lines on the floor.

223

4. Workload reduction of standing inspection work

Before

Figure 6.

After

Introduction of a rubber mat to reduce the workload.

5. KAIZEN of the cutting machine for egg noodles

Before

Figure 7.

After

Lowering of the cutter position and making of a new wider, two-roller cutter.

6. KAIZEN for checking the residual quantity in the feeder

Before

Figure 8.

After

Installing a mirror for checking from the 2nd floor.

224

7. KAIZEN for checking the tanks

Before

Figure 9.

After

Making of an acrylic-resin tank with an automatic valve.

8. Safety measure for the noodle roller safety bar

Before

Figure 10.

After

Improvement of the bar so that it could move in two directions (one direction before this).

9. KAIZEN in producing deluxe egg noodles

Before

Figure 11.

After

Simplification of the machine structure for cleaning and prevention of contamination by foreign matter.

10. Environmental KAIZEN for standing inspection work

Before

After

Figure 12. A shift of ceiling lamps and setting of a dimming cover on the nearby lamp.

225

11. KAIZEN for feeder cleaning

Before

Figure 13.

After

Making of a feeder work stool.

4 CONCLUSION Although women have excellent ability compared with men, they cannot avoid the effects of aging. It is important to conduct KAIZEN to best utilize the abilities of female workers. We are certain that we have been able to make the company an appealing one from the viewpoint of part-time workers’ continued employment, and form a basis for a continuous KAIZEN structure. REFERENCES Kumashiro, M., (1987). Work load , – Postures and Job Redesign – An Ergonomic and Industrial Management (Ergoma) Approach, NEW METHODS IN APPLIED ERGONOMICS, Taylor & Francis. London, pp. 247–252. Mikami, K., (2003). Aging and Work. Taylor & Francis. London, pp. 233–244 Mikami, K., Kumashiro, M., (1997). A scientific Approach to Work Improvement (I) – From the viewpoint of Ergoma Approach and Virtual Simulation-, The 14th International Conference on Production research, Osaka, Japan, pp. 1152–1155

226

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Usability research on the older person’s ability for Web browsing Daiji Kobayashi Department of Human Relations, Nihonbashi Gakkan University, Kashiwa city, Japan

Sakae Yamamoto Department of Management Science, Tokyo University of Science, Tokyo, Japan

1 INTRODUCTION 1.1 Usability and accessibility issues concerning Web browsing for the elderly in Japan The percentage of the Japanese population aged over 65 years is now over 21 percent, which is the highest ever. Although nearly 20 percent of the Japanese people in the over 65 age bracket have Internet access, most of the elderly lack sufficient information about the facilities available to them in order to make their life as comfortable as possible. Much of the information, from both public and private sources, about facilities such as healthcare, social insurance and other services, is available on the Internet. Among people aged over 65 years, the most popular method of obtaining information through the Internet is via web pages; therefore, making their Web browsing experience more comfortable could consequently enhance their skills and expand the number of job categories they are eligible for. However, Japanese elderly have less experience in using the typewriter. This is also the reason why they have an aversion to using the personal computer (PC). In this regard, the universally relevant issue of accessibility should be discussed. 1.2 Japanese response to usability and accessibility issues Usability is defined in ISO 9241-11:1998 as follows: a product can be used by specified users to achieve specified goals with effectiveness, efficiency and satisfaction in a specified context of use. This means that the products having usability are used to the specified users’ satisfaction in a specified context. Thus, the concept of usability specifies the user, the user’s goals and the context of use. On the other hand, these restrictions on the targets are not present in the concept of accessibility. Bergman and Johnson (1995) described accessibility as follows: ‘Providing accessibility means removing barriers that prevent people with disabilities from participating in substantial life activities, including the use of services, products, and information’. Therefore, specifying or identifying the barriers is difficult for persons without disabilities. Moreover, the degree to which a barrier is perceived is subjective and depends on the user’s characteristics. Hence, evaluating whether a product or service is accessible is difficult. However, some relevant standards on usability and accessibility have been issued by the International Standards Organization (ISO)1 , and the United States has enforced the regulations for making products for use by the elderly. Since the Amendment to Section 508 of the Rehabilitation Act came into effect in 1998 in the United States, Japanese electronics companies and computer makers have been paying attention to the concept of accessibility. In 2001, the METI2 published the original of ‘ISO/IEC Guide 71’which contains the guidelines for addressing the needs of the elderly and persons with disabilities. Since then, the guidelines have been providing directions to Japanese 1 For

example, ISO 9241-11 and ISO FDIS 9241-20 are well known. Ministry of Economy, Trade and Industry

2 METI:

227

manufacturers for making products for the elderly and persons with disabilities. Moreover, the Japanese government has actively promoted standardising through ISO, and a few committees and associations have been formed for the purpose of laying down standards, guidelines, and recommendations for providing accessible devices and services. Thus, the Japanese government has been actively involved in addressing the issue of accessibility. 1.3 Does the concept consider Web browsing accessibility issues for the Japanese elderly? Although guidelines, standards, regulations and recommendations exist, these have just abstractly informed us about how to realise accessibility. On the other hand, some researchers have attempted to address the accessibility issues experimentally. For example, Fisk et al. (2004) proposed design guidelines for older adults based on human factors approaches. However, these guidelines did not fit the Japanese elderly, who do not live in the English-speaking world as described in our previous work. Our previous research on Japanese older persons showed that they preferred fewer pages and menu branches, with information condensed using a smaller font size, rather than more pages with a larger font size. This was because they felt uncomfortable with multiple selections and preferred having all the information displayed in one place, rather than having to select individual topics Kobayashi and Yamamoto (2003). This observation indicates that the concept and technique of providing accessibility could depend on the various characteristics of Japanese users. According to the ‘General principles concerning measures for the aged society’ issued by the Japanese government (2004), the individual situation of Japanese elderly people and their living circumstances can vary widely according to gender, state of health, financial situation, family structure, and housing. Thus, it is incorrect to treat them all as similar. Therefore, the conventional knowledge about providing accessibility should be reconsidered from an older person’s perspective. 1.4 Aim of this study In this paper, through some case studies, we discuss the reason why only vague ideas for realising accessibility are indicated in the existing standards and recommendations, and propose some ideas for providing accessibility to the Japanese elderly. The concept of accessibility is applicable to people with disabilities; however, the people with disabilities have such a wide range of characteristics that considering all the issues for all of them would be difficult. Therefore, we narrowed the scope of our study to Japanese older persons. We investigated their characteristics in the context of Web browsing using a Web browser on a notebook PC and focussed on the issue of providing accessibility to the users. Thus, we considered the issue of Web browsing accessibility based on investigation among a range of users. Thus, the present study is basic research focusing on the issues involved in expanding older persons’ ability for participating in the Japanese Internet environment. 2 CASE STUDY OF HANDLING A COMPUTER MOUSE The computer mouse can be assumed to be the device that is commonly used for obtaining information from Web sites, although some types of notebook computers have other pointing devices. In Japan, a computer mouse with two buttons is most common pointing device used by the elderly, and most textbooks on using Web browsers and application software assume the use of a computer mouse. Therefore, we first observed a few older persons handling a computer mouse in order to identify the barriers they faced. 2.1 Participants The participants were 16 older persons, 11 males and 5 females, with ages ranging from 63 to 83 years (mean = 72.9, S.D. = 6.2), and all of them were right-handed. All of the participants were volunteers. Four male participants had prior experience browsing the Web using a computer. None of the participants had any physical or psychomotor disabilities. All participants were residents of the Shinjuku ward in Tokyo and visited a public community center in their neighbourhood. 228

Figure 1.

Home page of the official Web site of the Shinjuku ward.

2.2 Apparatus and materials All participants visited a public Web site (official Web site of the Shinjuku ward) using a notebook computer (IBM ThinkPad R40) with a mouse (Microsoft IntelliMouse® ) having two buttons and a wheel. The notebook computer was running the Microsoft® Windows® XP Professional operating system, Japanese edition, and had a 1024 × 768 pixel (15 inches) display. The Web browser used was Microsoft Internet Explorer Version 6. The public Web site provides a great amount of useful information for residents of the ward (see Figure 1). The Web site provides information regarding public and private services, the national pension, health care, etc. It consists of hierarchical pages with more than four layers requiring multiple selections. However, this study used a part of the Web site that was copied and was stored on the hard disks of the users’ notebook computers because the public community center did not provide Internet access facility to users. Further, a custom software was installed on the participants’ notebook computers that recorded mouse events on the hard disk of the notebook computer. Therefore, the locus of the mouse pointer and click events were obtained from the recorded files. 2.3 Procedure We instructed every participant on how to hold, move, and click the mouse, and allowed them to practice mouse operation. After receiving the instruction, each participant browsed the Web site without restraint, in succession. They stated their impressions of browsing each Web page of the Web site. The participants felt slightly nervous when a video camera pointed at them to record their actions. Their reports were recorded using a tape recorder, and their performance was recorded as mouse events using the custom software. After their Web browsing, their opinions about the experience were obtained. 2.4 Results of the observation The mouse events recorded in the data files indicated that the many erroneous clicking around each target in different font sizes occurred. Further, we observed the following characteristics of their Web browsing performance: • Several gaps were observed on the screen between the mouse pointer and the location of correct hypertext link that needed to be clicked (see Figure 2). 229

Figure 2.

Gaps observed on the home page of the Web site in the mouse events of all 16 participants. (The length of the gaps has been magnified 10 times in the figure).

• Unintentional manipulation such as clicking the middle or right buttons on the mouse caused the browser to scroll the Web page automatically or to display pop up context menu, which caused thorough confusion among the participants. • There were numerous small target areas on the Web pages, making it difficult for some participants to accurately point at them using the mouse. • There were some Japanese sentences in 8 point font, which was difficult for many participants to search and read. 2.5 What are the barriers they face? From the results of the above-mentioned observation, the difficulties faced during Web browsing by the elderly participants can be classified into the following three points: • Moving the mouse pointer to the desired hypertext link. • Keeping and holding the mouse in order to push and release the button. • Using small hypertext links on Web pages. Taking the above results into account, we had to identify the measures to be taken in order to improve the accessibility of Web browsing for the elderly. The results suggest that it is difficult for them to adjust their movements for correct mouse operation. In other words, they would continue to face mouse operation as a barrier. To remove this barrier, therefore, the layout of Web pages should be rearranged according to accessibility guidelines such as ISO/FDIS 9241-151 and W3C. Small hypertext links increase the difficulty of handling a computer mouse, and enlarging the font size of the Web page affects the layout and size of the Web page. Thus, improving the layout of Web pages alone is not sufficient to remove the barrier. Therefore, we should identify alternative methods for improving their Web browsing experience. 2.6 How can the barrier be removed? In order to improve the Web browsing experience for the elderly, we tried to develop an accessibility tool. This accessibility tool aimed to remove the barriers faced by the participants in the observation. Thus, we concluded that the following three functions must be incorporated into the Web pages in order to modify the Web browsing environment to the needs of the elderly: • Moving the mouse pointer to the desired hypertext link on the screen. • Enlarging the click area. • Accepting commands only from the left mouse button (see Figure 3). 230

Figure 3. Three functions for expecting providing accessibility.

Figure 4.

Experimental Web pages: Each page shows nine Japanese hypertext links using 8 point font size (left), 12 point font size (centre) and 16 point font size (right).

The above three functions are provided by two modules of the accessibility tool. The first module processes the mouse events using a system standard Application Program Interface (API), and the second module, which processes Hypertext Markup Language (HTML), embeds JavaScript and Cascade Style Sheet (CSS) in the HTML file through an HTTP proxy3 . 2.7 Verification of the accessibility tool’s effectiveness The effectiveness of the tool should be tested under use by the elderly; therefore, we conducted experiments and compared their performances with and without using the tool. The results suggested that the difficulty of handling a computer mouse could be related to the font size of the hypertext link. Therefore, we investigated the number of erroneous clicks and time taken for completing three different tasks (see Figure 4). The participants of the test were 21 older persons with ages ranging from 60 to 74 years (mean = 65.1, S.D. = 4.4). Eleven participants had prior experience operating a PC. None of the participants had any physical or psychomotor disabilities, and all of them were right handed. They were instructed on how to use the mouse and about the requirement of the task. They were allowed to practice mouse operation, on which no time limit was set, and assistance was provided on demand. The task was locating and clicking the hypertext links on the first Web page with 8 point font size, 3 Takahashi, Y., Kobayashi, D. and Yamamoto, S., 2005, Development of Accessibility Tool for Elderly People.

In Proceedings of HCI International 2005, Las Vegas.

231

5

Normal With the tool

***

Erroneous clicks

4

***: p  .01 **: p  .05 **

3

** 2 1 0

Figure 5.

8

12 Font size (point)

16

Comparison of the number of erroneous clicks.

100

Normal

***

With the tool ***

Time taken (sec)

80

***: p  .01 ***

60 40 20 0

Figure 6.

8

12 Font size (point)

16

Comparison of time taken for a task.

as shown in the image on the left of Figure 4. When all the hypertext links were clicked, the next page, with 12 point font size, was displayed (see the image in the centre of the figure). Similarly, the last page with 16 point font size was displayed after the participant had clicked all hypertext links on the previous page (see the image on the right of the figure). Every participant was presented with three trials without using the functions of the accessibility tool, and three more trials were completed after the functions were enabled. The apparatus used by the participants was the same as that used in the observational study described earlier. Based on the mouse events recorded in the data files, we counted the number of erroneous click events around each click area in every font size. The time elapsed between the first and last clicks on a page. The number of erroneous clicks and the time taken for the task are shown in Figures 5 and 6. Figure 5 presents the comparison between the average numbers of erroneous clicks. This result statistically indicates that the tool significantly reduced erroneous clicks (p < .05 or p < .01). The participants responded ‘Mouse operation became easy’, ‘The link objects were enlarged, so I could 232

see the links clearly’, ‘I could locate the link easily’. Therefore, we conclude that some of the barriers leading to erroneous clicks were reduced by the following: • The function for enlarging click area, which made it easy for the participants to select the link object. • The function for moving the mouse pointer, which helped the participants to move the mouse pointer to the desired hypertext link. • The function for moving the mouse pointer, which reduced erroneous clicks by decreasing the gap between the mouse pointer and the click area. Figure 6 shows that the average time taken was significantly reduced (p < .01) when the tool was used. Moreover, no assistance was required when all the functions of the tool were enabled.

2.8 Discussion Opinions about the tool and the trials were obtained from all participants after the trials. Seventeen participants felt that the operation became easier than normal, using the tool. However, four participants responded that they did not experience any changes in the operation. Two of these participants cited the reason ‘I did not have difficulty in the operation even without the tool’, and the other two participants said that ‘I did not understand the functions of the tool well because I am not interested in operating the PC’. In spite of the preferable performance using the tool, these opinions suggest two things. Firstly, it was difficult to improve accessibility for every participant, and secondly, it would be difficult to satisfy them regardless of whether accessibility is improved or not. The concept of accessibility is different from that of usability, in that the user’s satisfaction is not considered. However, the barriers faced by the elderly could be subjective distress and the feelings in their subconscious. Therefore, the satisfaction of users may represent the accessibility for them, so every user’s opinion is important. In this regard, we conducted further research, which is described later in the paper.

3 VERIFIYING THE EFFECTIVENESS OF THE ACCESSIBILITY TOOL’S FUNCTIONS Through another observational study, we attempted to identify the reason why the accessibility tool we developed could not satisfy every elderly user. Although the concept of a user should not be restricted by the accessibility concept, we should consider the factors affected by experience because it could affect the perception of a barrier for some elderly users. Further, since the accessibility tool contains three functions, we checked the effectiveness of each function. From these viewpoints, we conducted further investigation to ascertain the effectiveness of the accessibility tool.

3.1 Method Participants were 17 older persons with ages ranging from 65 to 83 years (mean = 73.3, S.D. = 6.3). Five participants had experience operating PCs for more than one year. None of the participants had any physical or psychomotor disabilities, and all of them were all right handed. The task was to find out the telephone number of four municipally-owned facilities on the public Web site (see Figure 1). Every participant first attempted to complete the task without the functions of the tool; later, another trial was completed with the functions enabled. In each trial, we randomly selected a facility and asked each participant to find its telephone number. After the trials, the participants’ opinions about the trials were obtained. The apparatus used in this study was the same as that used in the above-mentioned investigations. Thus, the participants’ performance was recorded as mouse events, using the custom software. 233

3.2 Results Based on the experimental data, the effectiveness of the three functions is discussed as follows. Firstly, based on the mouse events, we investigated the effectiveness of the function ‘moving the mouse pointer to the desired position’. Erroneous clicks were observed in six inexperienced participants’ performance when they did not use the tool, although many gaps were observed in the performance of every participant. When the tool was used, none of the participants realised that the gaps were corrected by the tool’s function; however, erroneous clicks were greatly reduced when the function was used. Therefore, it is fair to say that the effectiveness of the function was clarified. Secondly, the effectiveness of the function ‘enlarging click area’ was investigated based on the participants’ opinions. The results revealed that five participants, including an experienced older person, indicated that the click area was eye-friendly; however, the other five participants, including three experienced older persons, complained about the function because the enlarged click area overlapped other contents of the Web page. Therefore, this function was not useful for the elderly in some cases. Lastly, the function ‘accepting commands only from the left mouse button’ was evaluated. Although the participants could complete the task using the left button and a center wheel for scrolling, the 12 inexperienced participants pressed the middle and left buttons with an average frequency of about ten, without using the tool. On the other hand, the experienced participants’ frequency was about two. These results suggest that the function was effective just for the older persons inexperienced in handling a computer mouse. Further, we observed that some participants unintentionally browsed the same page repeatedly. The inexperienced participants said that they were not able to perceive the change in the colour of hypertext link that they clicked. These perception errors confused the inexperienced participants; therefore, during trials, most of these participants asked queries such as ‘What should I do?’ The performance of these participants’ indicates that the perception error causing confusion may be a barrier to Web browsing for the elderly. 3.3 Discussion The results revealed that the functions, except for ‘moving the mouse pointer to the desired position’, have limited effect in providing accessibility. This suggests that the context in which the tool is used affects the level of accessibility provided, in accordance with the participants’ characteristics such as experience and skill. Therefore it is difficult to standardise the recommendations for accessibility. Moreover, the factors causing the participants overlook the clicked hypertext links may raise a barrier for them; therefore, these factors should be addressed in order to provide accessibility for the elderly. There may be some methods for addressing these factors; however, it is important that the methods do not induce nervous strain on the older persons, as training does, given Japanese older persons’ aversion to the PC. Based on this point, another function prevents the older persons from overlooking the change in the colour of the hypertext link was proposed. 3.4 Effectiveness of reversing clicked hypertext links The function we tried was to highlight the clicked hypertext in order to prevent the participants’ perception error. The clicked hypertexts links were indicated by reversed characters using the Stylesheet Language for Web pages. This function was believed to be extremely effective and easy to use. Therefore, the effectiveness of this function was tested through the following experiment. Participants were eight older persons with ages ranging from 65 to 82 (mean = 70.5, S.D. = 6.9), and all of them were right-handed. Two of them had experience operating personal computers for more than one year. The apparatus, the task, and the procedure were the same as those used in the experiments described earlier. The results revealed that every participant subjectively agreed with our idea; therefore, we conclude that the function is effective in providing subjective accessibility, based on the context of use. 234

4 CONCLUSION In order to expand the older person’s work ability, some characteristics of elderly participants have been studied through some experiments and trials. The results revealed the characteristics of Japanese elderly that raised some barriers for them with regard to Web browsing. These barriers fall under two categories. The first category concerns their ability for Web browsing, which is covered under the conventional concept of accessibility. On the other hand, the second category concerns their manner and motivation for browsing the Web. Further, it was found that the barriers in the second category could not be removed by approaching the issue by the conventional accessibility concept alone. The experimental results also revealed that the level of accessibility felt differed from one participant to the other because their respective contexts of use were also different. This finding suggests that it is difficult to evaluate the accessibility level with respect to their physical or psychomotor abilities. Therefore, a measure of users’ satisfaction should be incorporated into the concept of accessibility as well as the usability concept. The definitive method for providing accessibility has not yet been proposed in the several existing accessibility guidelines and standards. In this regard, we should develop more methods for expanding the older person’s ability to utilize information technology, according to their respective contexts of Web use. REFERENCES Bergman, E. and Johnson, E., (1995). Towards Accessible Human-Computer Interaction. In: Advances in Human-Computer Interaction, Vol. 5, 1st, edited by Nielsen, J. (New Jersey: Ablex Publishing Corporation), pp. 87–113. Fisk, A. D., Rogers, W. A., Czaja, S. J., Charness, N. and Sharit, J., (2004). Designing for Older Adults, (London: CRC Press).

235

Chapter 5 Age Affected Functions

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Impact of psychosocial work environment factors measured by the COPSOQ on the need for recovery after work in aging workers. Preliminary results Philippe Kiss & Marc De Meester Securex Occupational Health Service, Ghent, Belgium Department of Public Health, Ghent University, Ghent, Belgium

1 INTRODUCTION In recent research it has been demonstrated that long-term adverse health effects are preceded by short-term effects (Sluiter et al., 1999). These short-term effects are signs of fatigue which are mostly experienced during or immediately after a day’s work. This is thought not to be a problem if enough recovery time is offered between two periods of work. If there is not enough time to recover from this fatigue in between two periods of work, the cumulated effects of this fatigue will lead to long-term adverse health effects (Sluiter et al., 1999). This is concordant with the cognitive activation theory of stress (CATS), where a sustained response may lead to illness and disease (Ursin and Eriksen, 2004). In the CATS stress response is defined as an alarm in a homeostatic system, producing neurophysiological activation. The activation can be reduced by coping mechanisms, triggered by the same alarm. If the coping mechanisms are inadequate to reduce the activation level, a certain aroused activation level remains. If sustained this may lead to adverse health effects (Ursin and Eriksen, 2004). The short term effects, manifested in feelings of temporary overload after work or subjective need for recovery, are recognizable in the immediate off-work situation and might be a useful indication of the personal psycho-physiological homeostatic balance at that moment (Sluiter et al., 2001). The subjective need for recovery can be measured by “The need for recovery scale”. This scale, used to measure early indications of fatigue at work, proved to be a powerful predictor of experienced health problems (Sluiter et al., 1999; Sluiter et al., 2003; van Veldhoven and Broersen, 2003). This was also confirmed in a study performed in 541 employees working in the public sector where the relationship between work stress and need for recovery was explored. The results showed that the presence of psychosomatic health complaints was significantly associated with the need for recovery (De Meester and Kiss, 2002). In earlier research, also the relationship between short-term effects and high psychosocial work load was demonstrated (Sluiter et al., 1999). More recently, it was shown that subjects who reported more job demands also reported more need for recovery from work (Sluiter et al., 2001). In a recent study (Kiss et al., 2007), it was pointed out that ageing workers had a higher need for recovery, even after taking into account several occupational, personal and social and family life factors. However, a detailed exploration on the influence of psychosocial work environment factors on the need for recovery was not performed up to now. The Copenhagen Psychosocial Questionnaire (COPSOQ) is a new tool for the assessment of psychosocial factors at work. The purpose of the COPSOQ concept is to improve and facilitate research as well as practical interventions at the workplaces. The COPSOQ concept turned out to be a valid and reliable tool for workplace surveys, analytic research, interventions and international comparisons. The questionnaire includes most of the relevant dimensions according to a number of important theories on psychosocial factors at work (Kristensen et al., 2005). 239

The aims of this study were to explore the impact of psychosocial work environment factors on the need for recovery using the COPSOQ and subsequently to explore differences between older and younger workers. However, since data collection is still in progress, these results can only be considered as preliminary.

2 METHODS This study has been approved by the Ethics Committee Progecov (Commissie voor Medische Ethiek OG 211, Ghent, Belgium). 2.1 Subjects 998 subjects employed in the public sector were asked to participate in a cross-sectional questionnaire study. All subjects received a standardized self-completed questionnaire at their home addresses by mail. One month later a second data collection was carried out amongst the nonresponders. Eventually 781 subjects returned the questionnaire (78.3% response rate). 15 subjects with missing values on the outcome variable and six subjects with unknown age were excluded from the analyses. This reduced the number of participants to 760 subjects (76.2% participation rate). Mean age was 41.6 years (SD 9.8), ranging from 19 to 63 years. For comparison of older and younger workers the subjects were divided into two age groups: older workers (45 years or older) and younger workers (younger than 45 years), according to the definition of the World Health Organisation (WHO 1993). 317 subjects (41.7%) were 45 years or older and 443 subjects (58.3%) were younger than 45 years. 2.2 Need for recovery The dependent outcome variable was the experienced need for recovery, which was assessed by “The Need for Recovery Scale” questionnaire. The need for recovery scale was computed by summing up the scores of the 11 constituent dichotomous (yes/no) items, resulting in a score ranging from 0 to 11, which was transformed to a 0 to 100 scale. Reliability of the scale in this study reached 0.89 (Cronbach’s α), which was comparable to the reliability found in earlier studies (Sluiter et al., 1999; van Veldhoven and Broersen, 2003; Jansen et al., 2003). A need for recovery score higher than 50 was defined as a high need for recovery, while a score of 50 and lower was defined as a low need for recovery. 2.3 Psychosocial work strain For the assessment of psychosocial work environment factors 17 dimensions of the COPSOQ were used, each transformed to a 0-100 scale: quantitative demands (seven items, Cronbach’s α 0.85), emotional demands (three items, Cronbach’s α 0.83), demands for hiding emotions (two items, Cronbach’s α 0.60), cognitive demands (eight items, Cronbach’s α 0.89), influence at work (10 items, Cronbach’s α 0.79), possibilities for development (seven items, Cronbach’s α 0.84), degrees of freedom (four items, Cronbach’s α 0.66), meaning of work (three items, Cronbach’s α 0.80), commitment to the workplace (four items, Cronbach’s α 0.62), social support (four items, Cronbach’s α 0.77), social relations (three items, Cronbach’s α 0.56), role clarity (four items, Cronbach’s α 0.67), role conflicts (four items, Cronbach’s α 0.82), predictability (two items, Cronbach’s α 0.72), feedback (two items, Cronbach’s α 0.61), sense of community (three items, Cronbach’s α 0.84) and quality of leadership (eight items, Cronbach’s α 0.94). The reliability figures were comparable with those found in the original study, where the detailed composition of the different scales was described (Kristensen et al., 2005). The dimension ‘role conflicts’ was coded reversed: high scoring for this item was concordant with a low degree of role conflicts. 240

2.4 Physical work strain and other occupational factors Physical work strain was assessed using a three item physical work load scale, comprising sustained physical efforts, lifting heavy weights and working in awkward positions (Cronbach’s α 0.83). Scoring for each separate item was analogous to scoring for the job demands items. The physical work load scale was computed by summing up the scores of the three constituent items, resulting in a score ranging from 0 to 9. The highest score was concordant with the highest physical strain. Other relevant occupational factors that could be of influence on the need for recovery were included in the questionnaire: part-time (<38 hours a week) or full-time work, shift work and more than one employment. 2.5 Personal factors Apart from age, following personal factors were asked for in the questionnaire: gender, presence of chronic musculoskeletal disorder and degree of ability to cope with stress. The degree of ability to cope with stress was assessed by the coping questions in the COPSOQ. 2.6 Family and social life factors Social life and family related matters which could be of influence on the need for recovery were considered as well. For this study questions were developed about: number of children at home, taking care of disabled or elderly people, degree of satisfaction (“very satisfied”, “fairly satisfied” and “not very satisfied”) on relationships with close relatives and friends and on social contacts (contacts with people in leisure activities, neighbours, hobby clubs, associations, . . .). 2.7 Statistical analyses All data analyses were performed using SPSS, version 15.0 for Windows (SPSS 2006). The categorical variables (age groups, gender, presence of musculoskeletal disorder, full/part time work, shift work, more than one employment, number of children at home, taking care of ill or old people, contacts with close relatives and friends, social contacts) were described by number and percentage for the total study population and for the younger and older workers separately. The non-categorical variables or scores (need for recovery, considered COPSOQ dimensions and physical work strain) were described by mean and standard deviation for the total study population and for the younger and older workers separately. For testing differences between older and younger workers the Mann-Whitney U test and the chi-square test were used appropriately. To prevent occurrence of multicollinearity, correlations between all exposure variables were checked beforehand and no high collinearity was found (Checkoway et al., 2004). Stepwise forward conditional multiple logistic regression analysis was used to calculate the odds ratio and their 95% confidence intervals for the presence of a high need for recovery (dependent variable). The regression analyses were performed for the total population and for the groups of older and younger workers separately. Following independent variables were entered into the regression model: age group (only in analysis of the total population), the 17 considered dimensions of the COPSOQ, physical work strain, full/part time work, shift work, more than one employment, gender, presence of musculoskeletal disorder, stress coping, number of children at home, taking care of ill or old people, contacts with close relatives and friends, social contacts. For the stepping method criteria, the p-value for including a variable was set at .05 and the p-value for excluding a variable at .10. The final multivariate model had the smallest value for – 2 Log likelihood and the highest correctly predicted percentage. 3 RESULTS The mean need for recovery for the whole study population (n = 760) was 35.9 (SD = 32.6). 490 subjects had a low need for recovery (64.5%) and 270 subjects had a high need for recovery (35.5%). 241

Table 1.

Need for recovery (NFR) score by age group. Workers ≥ 45 y.

Workers < 45 y. Variable

n

%

NFR score* Low NFR† High NFR†

298 145

67.3 32.7

Mean (SD)

n

%

Mean (SD)

192 125

60.6 39.4

33.7 (31.1)

39.0 (34.3)

* p = 0.070; † p = 0.057 Table 2.

Psychosocial and physical work strain factors for the whole study population and by age group. Total

Workers < 45 y.

Workers ≥ 45 y.

Variable

n

Mean (SD)

n

Mean (SD)

n

Mean (SD)

Quantitative demands Emotional demands* Hiding emotions Cognitive demands* Influence at work Possibil. development* Degrees of freedom Meaning of work Workplace commitment Social support Social relations Role clarity Role conflicts (rev.)* Predictability Feedback Sense of community Quality of leadership Physical work strain

759 758 757 759 756 757 757 756 758 756 754 758 754 758 755 754 756 757

43.8 (16.8) 45.8 (26.3) 45.7 (24.8) 59.7 (20.3) 48.6 (15.2) 62.7 (19.5) 57.9 (20.3) 82.6 (17.9) 63.5 (19.4) 65.0 (21.9) 58.6 (24.6) 72.3 (18.2) 64.4 (22.8) 55.8 (24.7) 43.3 (25.9) 72.8 (22.4) 58.7 (25.5) 2.9 (2.3)

443 442 441 443 440 441 441 440 442 441 439 442 441 442 440 438 441 441

44.8 (16.2) 47.3 (24.6) 45.3 (25.1) 61.3 (18.7) 48.4 (15.1) 64.4 (18.1) 57.3 (20.2) 82.8 (17.1) 62.8 (19.0) 65.8 (21.6) 60.2 (24.2) 71.8 (17.2) 62.8 (21.8) 55.0 (23.6) 44.7 (25.9) 72.3 (22.2) 58.9 (24.5) 2.9 (2.4)

316 316 316 316 316 316 316 316 316 315 315 316 313 316 315 316 315 316

42.4 (17.7) 43.6 (28.3) 46.4 (24.5) 57.4 (22.3) 48.8 (15.4) 60.4 (21.1) 58.7 (20.4) 82.4 (19.0) 64.6 (19.8) 64.1 (22.4) 56.5 (24.9) 73.1 (19.6) 66.7 (23.9) 56.8 (26.2) 41.3 (25.7) 73.5 (22.6) 58.6 (26.8) 2.8 (2.3)

* p < 0.05

The figures for the two age groups separately are shown in Table 1. Although the differences in need for recovery between the two age groups didn’t reach statistical significance, there was a clear tendency that the older workers had a higher need for recovery. Tables 2 and 3 summarize the occupational factors for the whole study population and the two age groups. Older workers reported to have less quantitative, emotional and cognitive demands. The differences were statistical significant for emotional and cognitive demands. Older workers reported to have less possibilities for development, but experienced less role conflicts as compared to the younger workers. Both dimensions showed to be statistically different. The psychosocial exposure factors didn’t show any significant differences. Physical work strain and having an additional job was not different between the two age groups. Older workers worked significantly more part time and were less employed in shift work. Overall it could be concluded that older worker had more or less a more favourable occupational exposure profile than the younger workers. The personal, family and social life factors are summarized in Table 4. The mean stress coping score for the total study population (n = 748) was 23.2 (SD = 25.0) and 22.4 (SD = 23.9) and 24.3 (SD = 26.5) for the younger (n = 436) and older (n = 312) workers respectively. 242

Table 3.

Other relevant occupational factors for the whole study population and by age group. Total

Workers < 45 y.

Workers ≥ 45 y.

Variable

n

%

n

%

n

%

Part time work* yes no

261 496

34.5 65.5

136 305

30.8 69.2

125 191

39.6 60.4

Shift work† no yes

507 247

67.2 32.8

271 170

61.5 38.5

236 77

75.4 24.6

Other employment no yes

700 59

92.2 7.8

407 36

91.9 8.1

293 23

92.7 7.3

* p < 0.05; † p < 0.001

Table 4.

Personal and family and social factors for the whole study population and by age. Total

Workers < 45 y.

Workers ≥ 45 y.

Variable

n

%

n

%

n

%

Gender* men women

266 494

35.0 65.0

134 309

30.2 69.8

132 185

41.6 58.4

MSD† no yes

559 185

75.1 24.9

343 92

78.9 21.1

216 93

69.9 30.1

Number of children at home‡ 0 1 2 ≥3

249 208 182 53

39.9 28.2 24.7 7.2

152 114 125 40

35.2 26.5 29.0 9.3

142 94 57 13

46.4 30.7 18.6 4.2

Taking care of ill or old people‡ no yes

586 167

77.8 22.2

374 65

85.2 14.8

212 102

67.5 32.5

Contacts with close relatives* Very satisfactory Fairly satisfactory Not very satisfactory

475 254 26

62.9 33.6 3.4

297 133 9

67.7 30.3 2.1

178 121 17

56.3 38.3 5.4

Contacts with friends Very satisfactory Fairly satisfactory Not very satisfactory

395 331 29

52.3 43.8 3.8

246 177 17

55.9 40.2 3.9

149 154 12

47.3 48.9 3.8

Social contacts Very satisfactory Fairly satisfactory Not very satisfactory

344 380 32

45.5 50.3 4.2

212 212 16

48.2 48.2 3.6

132 168 16

41.8 53.2 5.1

MSD: musculoskeletal disorder; *p = 0.001; † p = 0.005; ‡ p < 0.001

243

Table 5.

Final multivariate logistic regression model for the presence of a high need for recovery in the whole study population (n = 681).

Variable

OR (95% CI)

≥45 y. (vs. < 45 y.) Quantitative demands Emotional demands Possibilities for development Predictability Sense of community Physical strain Degree of stress coping Satisfactory social contacts Not very (vs. very) Fairly (vs. very)

1.73 (1.16–2.59) 1.04 (1.03–1.06) 1.03 (1.02–1.04) 0.98 (0.96–0.99) 0.98 (0.98–0.99) 0.99 (0.98–1.00) 1.19 (1.09–1.30) 0.99 (0.98–1.00)

Table 6.

3.12 (1.13–8.65) 1.86 (1.24–2.80)

Final multivariate logistic regression models for the presence of a high need for recovery for both age groups.

Variable Quantitative demands Emotional demands Cognitive demands Possibilities for development Predictability Sense of community Physical strain Gender (women vs. men) Satisfactory social contacts Not very (vs. very) Fairly (vs. very)

<45 y. (n = 399) OR (95% CI)

≥45 y. (n = 280) OR (95% CI)

1.04 (1.02–1.06) 1.03 (1.02–1.04))

1.04 (1.02–1.07) 1.02 (1.01–1.04) 1.02 (1.00–1.05) 0.96 (0.94–0.98) 0.98 (0.97–1.00)

0.98 (0.96–0.99) 0.98 (0.97–1.00) 0.98 (0.97–1.00) 1.20 (1.05–1.36) 2.40 (1.25–4.60)

1.23 (1.07–1.41)

4.28 (1.04–17.59) 2.73 (1.56–4.79)

Table 5 shows the explorative multivariate logistic regression model for the presence of a high need for recovery in the whole study population. Table 6 shows the final multivariate logistic regression models for the presence of a high need for recovery for both age groups separately.

4 DISCUSSION In this study older workers had a higher need for recovery than younger workers, afer taking into account occupational, personal and family and social factors. This confirms the findings of our earlier research (Kiss et al., 2007). In this earlier study, however, occupational exposures were about similar in older and younger workers, while in the present study population older workers reported their occupational exposures to be more favourable than younger workers. Furthermore, the odds ratio for a high need for recovery was markedly higher than in the former study (1.73 vs. 1.56). It could therefore be concluded that, regardless the occupational exposures, older workers have a higher need for recovery than younger workers. 244

Quantitative demands had a similar impact on the presence of a high need for recovery in older and younger workers. Reducing the amount of work is beneficiary for both age groups to a similar degree. Despite the fact that older workers reported to be less exposed to emotional and cognitive demands, they experienced a negative impact on their need for recovery. For emotional demands this was to the same extent as for the younger workers (OR 1.02 vs. 1.03 respectively). Cognitive demands didn’t show to have any effect in the younger group. It is therefore justified to conclude that older workers were particularly sensitive to emotional and cognitive demands. There was a significantly larger proportion of men in the older group. This was probably the reflection of the increased participation of younger women on the labour market. The differences in number of children at home and taking care of ill or old people is a reflection of the normal life cycle evolution: when people get older, children leave the house and disabled parents make their entrance. Although certain personal and family factors were significantly different between older and younger workers, the main factors contributing to the presence of a high need for recovery were work related. The strong association of social factors with the need for recovery could be reciprocal: satisfactory social contacts could have a positive influence on one’s need for recovery, but it could also be explained by the fact that subjects with a high need for recovery are too exhausted to take part in social and leisure activities. The role of social factors still needs to be elucidated. Since data collection is still ongoing, the presented results can only be considered as preliminary. However, the confirmation of our earlier findings suggests that older workers have indeed a higher need for recovery than younger workers, whether they have similar or even more favourable occupational exposures. Exposure assessment without assessing (early) health effects is therefore inadequate to propose preventive measures. REFERENCES Checkoway H., Pearce N., and Kriebel D., (2004). Research methods in occupational epidemiology, 2nd edition, (New York: Oxford University Press). De Meester M., and Kiss P., (2002). Relationship between work stress and need for recovery. Preliminary results [abstract]. La Medicina del lavoro, 93: pp. 460 Jansen N.W.H., Kant I.J., van Amelsvoort L.G.P.M., Nijhuis F.J.N., and van den Brandt P.A., (2003). Need for recovery from work: evaluating short-term effects of working hours, patterns and schedules. Ergonomics, 46: pp. 664–680 Kiss P., De Meester M., and Braeckman L., (2007). Differences between younger and older workers in the need for recovery after work. International Archives of Occupational and Environmental Health, Jun 19 [Epub ahead of print] KristensenT.S., Hannerz H., HøghA., and BorgV., (2005). The Copenhagen Psychosocial Questionnaire (COPSOQ). A tool for the assessment and improvement of the psychosocial work environment. Scandinavian Journal of Work Environment and Health, 31: pp. 438–449 Sluiter J.K., de Croon E.M., Meijman T.F., and Frings-Dresen M.H.W., (2003). Need for recovery from work related fatigue and its role in the development and prediction of subjective health complaints. Occupational and Environmental Medicine, 60 (Suppl 1): pp. i62–i70 Sluiter J.K., Frings-Dresen M.H., van der Beek A.J., and Meijman T.F., (2001). The relation between workinduced neuroendocrine reactivity and recovery, subjective need for recovery, and health status. Journal of Psychosomatic Research, 50: pp. 29–37 Sluiter J.K., Van der Beek A.J., and Frings-Dresen M.H.W., (1999). The influence of work characteristics on the need for recovery and experienced health: a study on coach drivers. Ergonomics, 42: pp. 573–583 SPSS, 2006, SPSS for Windows, version 15.0, (Chicago: SPSS Inc.) Ursin H., and Eriksen H.R., (2004). The cognitive activation theory of stress. Psychoneuroendocrinology, 29: pp. 567–592 van Veldhoven M., and Broersen S., (2003). Measurement quality and validity of the “need for recovery scale”. Occupational and Environmental Medicine, 60 (Suppl 1): pp. i3–i9 World Health Organization (WHO), 1993, Aging and working capacity: report of a WHO study group. WHO technical report series 835, (Geneva: World Health Organization)

245

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Age differences in mental workload while performing visual search task∗ Miwa Takahara, Toshiaki Miura, Kazumitsu Shinohara and Takahiko Kimura Department of Applied Cognitive Psychology, Graduate School of Human Sciences, Osaka University, Suita, Japan

ABSTRACT: The main objective of this study is to investigate the age-related changes in the perceived mental workload during a visual search task. Nineteen older adults (M = 72.1 years) and fourteen younger adults (M = 22.9 years) participated. The participants were asked to detect a target while ignoring task-irrelevant singleton distractors and to estimate the perceived mental workload, using the NASA Task Load Index (NASA-TLX). Reaction times (RTs), sensitivity in detecting the target (d ) and response bias (β) were recorded as the performance on the visual search task. After completing the visual search task, the subjective mental workload was assessed by the NASA-TLX. The results on the visual search task showed that although the older adults’ RTs were longer than those of the younger adults, there were no age-related changes with regard to d and β. Some differences in the NASA-TLX were found between the older and younger adults; the temporal demand score increased with age, suggesting that older adults tended to feel time pressure strongly in contrast to younger adults. These findings imply that it is important to analyze performance as well as assess the perceived mental workload when designing visual tasks suitable for older adults.

1 INTRODUCTION Visual search is an important behavior in our daily life. Many a time, we have to find our target from many objects for the accomplishment of our purpose. However, there is little chance that the only thing we have to do is look for our target. While searching for our target, we often make efforts to ignore irrelevant objects. This activity involving signal selection and inhibition is called “attentional control function,” and it has been pointed out that this function deteriorates with age (Lincourt et al., 1997; Juola et al., 2000; Pratt and Bellomo, 1999). Moreover, the age-related changes in attentional control become marked when irrelevant objects have a feature that captures attention. Meanwhile, some researchers report no age-related changes in attentional control (Kramer et al., 1999). Therefore, we have to be clear about the age-related changes in attentional control. Instead of cognitive performance, perceived mental workload can be used to investigate individual effects in the performance of cognitive tasks. It is indicated that the perceived mental workload, which is measured by subjective indexes such as the NASA Task Load Index (NASA-TLX) and the Subjective Workload Assessment Technique (SWAT), increases as the performance on an attentional task decreases (Warm et al., 1996). However, Deaton and Parasuraman (1993) and Bunce ∗ This

research was supported by the Ministry of Education, Culture, Sports, Science and Technology (Project No. 16091205). Correspondence concerning this paper should be addressed to Miwa Takahara ([email protected]).

247

and Sisa (2002) reported that in older adults, the perceived mental workload increased even though age equivalence was observed in vigilance performance. These results demonstrated the possibility that older adults influence their workloads by performing attentional tasks, even if their performance does not decrease. Thus, we have to assess not only their cognitive performance but also their perceived mental workload during attentional tasks. In particular, it remains unclear whether the perceived mental workload during the performance of a visual search task changes with age. Therefore, this research was conducted to investigate the age-related changes in cognitive performance and perceived mental workload during a visual search task. A modified paradigm of Theeuwes et al. (2004) was used as the visual search task. In this task, singleton stimuli were presented as distractors, which participants had to ignore, and the participants’ attentional controls were investigated. Furthermore, the perceived mental workload was estimated by the NASA-TLX. The expected results were that unlike in the case of younger adults, in older adults, the visual search performance decrease and the perceived mental workload would increase.

2 METHOD 2.1 Participants Nineteen older adults aged 67 to 79 years (seven women), and 14 younger adults aged 18 to 24 years (eight women) formed the two age groups. The mean ages for the older and younger groups were 72.1 and 22.9 years respectively. The older participants were recruited through the National Silver Human Resources Center Association, and the younger participants were undergraduate students or graduate students of Osaka University. All the participants had normal vision or corrected normal vision. The mean educational years for the older and younger groups were 13 and 15 years respectively. 2.2 Visual search task 2.2.1 Apparatus and stimuli A Personal computer (e-Machine, J2921) and 21-inch display (MITSUBISHI, Diamondtron Flat RDF225) were used in this study. SuperLab Pro for Windows Ver. 2.0 (Cedrus) was used for the stimulus presentations and for recording the participants’ responses. The participants responded by using a keyboard. Experiments were conducted in a dark room, and participants’ heads were fixed by a chin rest. The display was arranged at a distance about 60 cm distance from the participants. The trials were conducted in four phases: the “Ready” phase, “Fixation” phase, “Search” phase and “Mask” phase. Stimuli were presented on black backgrounds. In the Ready phase, the word “Ready?” was arranged at the center of the display. In the Fixation phase, a fixation point denoted by an asterisk was presented at the center of the display (1◦ × 1◦ ) along with four circles (2.5◦ in diameter) at a distance of 5.5◦ from the fixation point. These circles were arranged in a square- or diamond-shaped configuration. Each circle contained an additional figure, which was a diamondshaped (in the simple blocks) or star octagon-shaped (in the complex blocks), and six crossover segments (1.5◦ in length). The six segments were a vertical segment, two segments tilted 20◦ to the right and left from the vertical position, a horizontal segment, and two segments tilted 20◦ upward and downward from the horizontal position. These segments crossed each other in the center. The color of the stimuli in the fixation phase was gray (16 cd/m2). Furthermore, some stimuli of the fixation phase changed in the Search phase. In the Search phase, the fixation point changed to a cross shape, and the color of all the circles, except for one, changed to red (16 cd/m2). At the same time, in the circle, only one segment remained, and the remaining five segments and additional figure disappeared. In the case of the one circle that remained gray, there was the possibility that a target would appear. The target was a vertical or horizontal line. In each red circle, a tilted line was presented as a distractor. In the Mask phase, a random dot pattern appeared in all the circles. 248

Task-irrelevant singleton distractor

*

*





Fixation Gray Red

Figure 1.

Fixation

Search [Simple block]

Search [Complex block]

Difference between fixation and search phases in each block.

2.2.2 Task Initially, the Ready phase was presented, and it immediately changed to the Fixation phase when the participants pressed a key. The Fixation phase was presented for 1000 ms. Next, the Search phase appeared on the display. The presentation time of the Search phase was determined individually to maintain a correct response rate of 70%–80% in the preparation experiments. The mean presentation times for the older and younger groups were 183 ms and 54 ms, respectively. Lastly, the Mask phase was presented for 3,000 ms. The participants were instructed to determine whether or not the target, which was a vertical or horizontal line, was present in the gray circle, and to press the “Z” or “/” key that corresponded to the target-present or target-absent as quickly and as accurately as possible, every time the Search phase appeared. There were two kinds of blocks: Simple and Complex, and the Search phase was different for every block. Each block had two conditions: control (no-distractor) and distractor. The control conditions were the same between the two blocks. However, in the distractor condition of the Simple block, one of the distractors was a diamond-shaped figure (simple task-irrelevant distractor), and in the distractor condition of the Complex block, it was a star octagon-shaped figure (complex task-irrelevant distractor). The two conditions were presented in random order. Participants were instructed to ignore the task-irrelevant distractors. Each block contained the practice blocks (24 trials) and the experimental block (96 trials). The reaction times (RTs), sensitivity in detecting the target (d ) and response bias (β) were recorded as the performance on the visual search task. 2.3 Perceived mental workload The perceived mental workload was estimated by NASA-TLX. This comprises six subordinate scales: mental demands, physical demands, temporal demands, own performance, effort, and frustration. First, the participants arranged the six subordinate scales in order of importance (arrangement) with regard to the performance of the visual search task. Next, they rated each subordinate scale from 0 (low/poor) to 100 (high/good) (rating). A weighted workload score (WWL) was calculated by the results pertaining to arrangement and rating. If frustration was perceived as more important by the participant, frustration ranked as 1 and the arrangement score was ranked as 5 (possible range 0 to 5). The WWL was calculated by multiplying the arrangement score with the rating score (possible range: 0 to 100) per subordinate scale, summing these weighted scores, and dividing it by 15 (the total arrangement score). The participants evaluated the perceived mental workload for each block in the visual search task by using the PC. 2.4 Procedure First, the participants practiced one of the two blocks in the visual search task, and then, they arranged the six subordinate scales of the NASA-TLX in order of importance with regard to performance of the block. Following the execution of the block, the participants evaluated each 249

Table 1.

Mean RTs, d , and β (SE) for the visual search task. Older adults

RTs (ms) d β

Control Distractor Control Distractor Control Distractor

Younger adults

Simple

Complex

Simple

Complex

895 (60) 874 (62) 2.01 (0.14) 2.07 (0.17) 1.64 (0.67) 1.90 (0.88)

857 (62) 861 (56) 1.83 (0.18) 1.95 (0.14) 1.49 (0.77) 2.03 (0.69)

806 (32) 825 (40) 2.12 (0.20) 2.11 (0.21) 2.52 (0.78) 2.82 (0.87)

813 (35) 830 (37) 1.96 (0.17) 2.12 (0.23) 2.20 (0.77) 2.26 (1.59)

subordinate scale from 0 (low/poor) to 100 (high/good). The same proceedings were applied to the other block. The entire session lasted for about 50 min. 3 RESULTS 3.1 Visual search task The RTs, d , and β were analyzed by a 2 (age: older and younger adults) × 2 (block: simple and complex blocks) × 2 (condition: control and task-irrelevant distractor conditions) ANOVA as the performance on the visual search task. The data is presented in Table 1. Regarding the three measurements, none of the main effects and interactions were significant. 3.2 Perceived mental workload The WWL and rating score of the six subordinate scales were analyzed by a 2 (age: older and younger adults) × 2 (block: simple and complex blocks) ANOVA as the perceived mental workload. The data is presented in Table 2. First, regarding the WWL, the main effects and interactions were not found to be statistically significant. In the subordinate scales, the main effects and interactions with regard to physical demands, own performance, and effort were found to be nonsignificant. However, regarding the other subordinate scales, the participants indicated different effects among the factors. With regard to the mental demands, the main effect for age was significant (F [1,31] = 6.43, p = 0.02). This showed that older adults perceived more perceptual load than younger adults. The main effect for block also was significant (F [1,31] = 4.34, p = 0.05), indicating that participants perceived more perceptual load in performing the complex block than the simple block. The interaction effect was not significant. In addition, although the interaction was not significant, the main effect for age and block on temporal demands was significant (F [1,31] = 4.59, p = 0.04; F [1,31] = 6.02, p = 0.02). These results indicated that time pressure was perceived more strongly as age increased, and that participants perceived time pressure more strongly in performing the complex block than the simple block. Last, a large age-related change was observed with regard to frustration (F [1,31] = 11.41, p = 0.002). Briefly, older adults perceived more frustration than younger adults. The main effect for block and interaction were not significant. 4 DISCUSSION We estimated that there would be age differences in the performance of the visual search task (RTs, d , and β). However, after having considered all the measurements, no differences among 250

Table 2.

Mean WWL and subordinate scales (SE) for perceived mental workload. Older adults

Younger adults

NASA-TLX

Simple

Complex

Simple

Complex

WWL Subordinate scales Mental demands Physical demands Temporal demands Own performance Effort Frustration

57.7 (3.5)

58.7 (3.4)

64.9 (2.9)

68.3 (3.1)

55.3 (12.7) 48.7 (11.2) 49.8 (11.4) 67.1 (15.4) 59.6 (13.7) 44.5 (10.2)

67.1 (15.4) 43.2 (9.9) 62.9 (14.4) 67.2 (15.4) 56.4 (12.9) 41.6 (9.5)

71.0 (19.0) 34.6 (9.2) 36.3 (9.7) 66.2 (17.7) 66.9 (17.9) 71.3 (19.1)

76.8 (20.5) 42.8 (11.4) 40.8 (10.9) 58.8 (15.7) 68.1 (18.2) 68.5 (18.3)

participants of different ages were found. In other words, the results showed that age, distraction from singleton stimuli, and stimulus complexity did not affect participants’performances. Although these results were unexpected, we considered that the perceptual adjustments due to the preparation experiments, in particular, contributed to task equalization among participants of different ages. Unlike in the performance of the visual search task, in some evaluations of perceived mental workload, the results indicated that there were, indeed, age differences. With regard to mental demands and frustration, the young adults’ estimations were higher than those of older adults. It seemed that the younger adults could maintain their performance by persevering through higher mental demands and frustration. However, these results were not peculiar, because the presentation times in the Search phase were controlled such that they were shorter (54 ms) for the younger adults than for the older adults (183 ms). With regard to temporal demands, we obtained the opposite outcome. The older adults reported higher temporal demands than the younger adults in performing the task, although their presentation times in the Search phase were longer than those for the younger adults and all other times in the trial were the same regardless of age. This showed that the older adults had a greater tendency to perceive time pressure than did the younger adults, even though their performance did not decrease. This result is similar to the results of Deaton and Parasuraman (1993) and Bunce and Sisa (2002), according to which in older adults, there exists a discrepancy between the performance in a vigilance task and the estimation of perceived mental workload. Although the age-related changes in cognitive function have been investigated in the past, it was not apparent how older adults perceived their mental workload. The findings of this study imply that older adults are particularly affected by time pressure when they are searching for a target object; moreover, there exists the possibility of their failure during the performance of such a visual task. Therefore, the control of older adults’temporal demands may be critical when designing visual tasks suitable for them which include visual search (e.g. Graphical User Interface (GUI) and Automated Teller Machine (ATM)). If the task complexity and temporal demands are appropriately controlled, it appears that older adults will be able to accomplish the task as well as younger adults. REFERENCES Bunce, D., and Sisa, L., (2002). Age differences in perceived workload across a short vigil. Ergonomics, 45: pp. 949–960 Deaton, J. E., and Parasuraman, R., (1993). Sensory and cognitive vigilance: Effects of age on performance and subjective workload, Human Performance, 6: pp. 71–97 Juola, J. F., Koshino, H., Warner, C. B., McMickell, M., and Peterson, M., (2000). Automatic and voluntary control of attention in young and older adults. American Journal of Psychology, 113: pp. 159–178

251

Kramer, A. F., Hahn, S., Irwin, D. E., and Theeuwes, J., (1999). Attentional capture and aging: Implications for visual search performance and oculomotor control. Psychology and Aging, 14: pp. 135–154 Lincourt, A. E., Folk, C. L., and Hoyer, W. J., (1997). Effects of aging on voluntary and involuntary shifts of attention. Aging Neuropsychology & Cognition, 4: pp. 290–303 Pratt, J., and Bellomo, C. N., (1999). Attentional capture in younger and older adults. Aging Neuropsychology and Cognition, 6: pp. 19–31 Theeuwes, J., Kramer, A. F., and Kingstone, A., (2004). Attentional capture modulates perceptual sensitivity. Psychonomic Bulletin & Review, 11: pp. 551–554 Warm, J. S., Dember, W. N., and Hancock, P. A., (1996). Vigilance and workload in automated systems. In: Automation and Human Performance: Theory andApplications, edited by Parasuraman, R. and Mouloua, M. (Mahwah, NJ: Erlbaum), pp. 183–200

252

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

A study of clinical assessment of Unilateral Spatial Neglect using a Head Mounted Display system (HMD) for elderly stroke patients in a virtual reality technology Toshiaki Tanaka, S. Ino & T. Ifukube Research Center for Advanced Science and Technology, University of Tokyo, Japan

S. Sugihara Sapporo Shuyukai Hospital, Japan

S. Shirogane Institute of Rehabilitation Science, Tokuyukai Medical Corporation, Japan

Y. Oyama Tokeidai Memorial Hospital, Japan

Y. Maeda Hokkaido University, Japan

ABSTRACT: Purpose – Unilateral spatial neglect (USN) is a common syndrome in which a stroke patient fails to report or respond to stimulation from the side of space opposite a brain lesion, where these symptoms are not due to primary sensory or motor deficits. USN is most damaging to an older stroke patient who also has a lower performance in their activities of daily living or those elderly who are still working. The purpose of this study was to analyze an evaluation process system of USN using HMD in order to understand more accurately any faults of USN operating in the object-centered co-ordinates and egocentric co-ordinates system. Participants – Nine elderly stroke patients (mean age 68.7 years old) participated in this study after gaining their informed consent and they all had Left USN as determined by clinical tests. Methods – Assessments of USN were performed by using the BIT common clinical test (the line cancellation test) and special two tests; the object-centered co-ordinates (OC) condition and the egocentric co-ordinates (EC) condition. OC condition focused the test sheet only by a CCD. EC condition was that CCD can always follow the subject’s movement. The subjects were first evaluated by the common clinical test without HMD and then two spatial tests with HMD. Result – In the OC condition, the left side was 44% and the right side was 94%. In the EC condition, the left side was 61% and the right side was 67%. In the EC condition, the subjects can rotate to the center of the test sheet. Discussion – The results showed that the assessment of USN using a technique of HMD system may indicate the disability of USN more precisely than the common clinical tests. HMD can produce an artificially versatile environment as compared to the common clinical evaluation. Keywords: Unilateral spatial neglect, head mounted display system, virtual reality, clinical assessment, elderly stroke patient

253

1 INTRODUCTION Unilateral spatial neglect (USN) is a common syndrome in which a patient fails to report or respond to stimulation from the side of space opposite a brain lesion, where these symptoms are not due to primary sensory or motor deficits (Heilman, 1979). Patients with severe neglect often collide with objects, ignore food on one side of the plate, and in general tend to rely on just one side of the body (Weistein, 1977). Patients with USN of the left hemispace require longer hospital stays and have more difficulty resuming activities of daily living (Wade, 1983). Neglect is associated with lower performance on measures of impairment, as well as on measures of disability in ADL. Recently, several studies have singled out USN as one of the major disruptive factors impeding functional recovery and rehabilitation success (Denes, 1982). From a rehabilitation perspective, the traditional assessment of USN centers on a variety of simple perceptual motor tasks. Investigations have used line crossing (Albert, 1973), cancellation task (Diller, 1977) and more recently, an indented reading test (Calpan, 1987). However, there is no single standardized battery of tests currently available for the assessment of USN. An analysis of USN can be explained with a space coordinate system theory. The boundaries of the neglected space are not constant in as much as the neglect patients’ performance is influenced by the relevant system of spatial coordinates; egocentric or allocentric co-ordinates. Egocentric co-ordinates specify locations relative to the viewer (Karnath, 1994), whereas allocentric co-ordinates code their position independent of viewpoint (Calvanio, 1987). Several sensory manipulations may be temporarily effective for improving unilateral spatial neglect. Karnath (1994) indicated the effectiveness of neck vibration. Pizzamiglio et al. (1990) also adopted an effective means of optokinetic stimulation. Rossetti et al. (1998) investigated the effect of prism adaptation on neglect symptoms, including the pathological shift of the subjective midline to the right. They reported that all patients exposed to the optical shift of the visual field to the right were improved in their manual body-midline demonstration and on their classical neuropsychological tests. However, these manipulations have not yet succeeded in bringing about a consistent improvement of neglect. Virtual Reality (VR) has many advantages over other ADL rehabilitation techniques and offers the potential to develop a human performance testing and training environment (Lee, 2003) and also a VR system for training individuals with unilateral spatial neglect to cross streets in a safe and vigilant manner (Weiss, 2003). VR can give human versatile sensory information artificially and easily for the visual, vestibular, and the somatic sensations. Recently, VR has been investigated in a few studies using devices for compensation of visual sensory. For example, there is one approach where HMD gives a patient with Parkinson’disease an emphasized visual input in order to improve a frozen gait of the patient (Prothero, 1993). HMD has a function which can focus on a certain object or to limit the surrounding environmental conditions, and to offer versatile visual information. Therefore, HMD can produce the object-centred co-ordinates for a USN patient. The purpose of this study was to analyze an evaluation process system of USN using HMD in order to understand more accurately any faults of USN operating in the object-centered co-ordinates and egocentric co-ordinates system. 2 METHODS 2.1 Subjects Nine patients who had suffered a stroke (mean age 68.7 years old) participated in this study after gaining their informed consent. The patients were tested for the presence of any neglect for activities of daily living (ADL) by two physical therapists. Two medical doctors checked the right hemisphere damage of all subjects by CT (computed tomography) or MRI (magnetic resonance imaging). Individuals with weak visual acuity, dementia, hemianopsia, apraxia or those being lefthanded were excluded. The subjects also had to be able to sit on an ordinary chair by themselves. The period from the appearance of disease to study assessment was 4–27 weeks (Table 1). 254

Table 1.

Patients characteristics.

No.

Age

Diagnosis

Lesion

Time of rehab. Onset

FIM-M

1 2 3 4 5 6 7 8 9

75 65 64 63 56 70 79 68 78

I I H H I I I I I

FPT BgFPT Th Bg PT Bg FPT BgFPT PT

6 weeks 1 week 1 week 1 week 1 week 1 week 1 week 2 weeks 1 week

30 38 61 35 85 33 86 72 42

I; Infarction, H; haemorrhage, F; frontal lobe, P; parietal lobe, T; temporal lobe, Bg; basal ganglia, Th; thalamus, FIM-M; Functional Indedence Measure Motor. ∗ all lesions were right sided. Table 2.

Checklist of Everyday Neglect Behaviors.

Does the patient 1. 2. 3. 4. 5. 6. 7.

show difficulties when talking or communicating with others neglect the left/right side of personal space? show difficulties in eating? show difficulties in grooming (self-care, washing, bathing, etc.)? show difficulties in dressing? show difficulties in body movement transferring (from a bed, to W/C, etc.)? show difficulties in locomotion 1 (the patient collides against objects and wall on the affected side and/or can not negociate a W/C between doors, kerbs, etc.)? 8. show difficulties in locomotion 2 (the patient turns toward the direction of the affected side.)? 9. show difficulties during PT exercise? 10. show difficulties during OT exercise?

2.2 Functional assessment The Functional Independence Measure (FIM) was executed as an ADL evaluation (Granger 1986, 1987). The FIM motor sub scores (FIM-M) were used for measure of disability as the best predictors of rehabilitation length of stay for stroke. Moreover, two physical therapists evaluated the patients who exhibited specific neglect behaviors in ADL using a special checklist (Table 2). A modified version of Halligan’s checklist was used (Halligan, 1991). The therapists were requested to score the checklist in terms of those behaviors they considered to be related to visual neglect, as opposed to poor performance that might be expected to follow concomitant disorders such as problems of motor coordination or initiation.

3 EVALUATION FOR USN 3.1 Common clinical test (Figure 1) To assess neglect, the widely used line and star cancellation tests as included in the Behavioral Inattention Test (BIT) were given to the subjects (Wilson, 1987). We used the BIT Japanese version which was modified by Ishiai et al. (1999). 255

Left side of test sheet Right side of test sheet 18/18 18/18

Left-left Mid-left Right-left Left-right Mid-right Right-right 8/8

11/11

8/8

8/8

11/11

8/8

Figure 1. Analysis method for line and star cancellation test.

A digital camera HMD

A digital video camera

Figure 2.

Experimental setup for the HMD (head mounted display) system.

For the line cancellation test (score range from 0 to 36 points), the subjects were presented with a single sheet of paper on which six lines in varying orientations were drawn, 18 on each side. They were instructed to make a mark through all of the lines. Left- sided neglect was indicated by a failure to mark more lines on the left side than on the right. Degree of neglect was assessed by the proportion of lines omitted relative to the total number of lines. The line cancellation test sheet was divided into right and left portions and a right and then a left correct answer rates were analyzed. 34 points were set as a cut-off value. For the star cancellation test (score range from 0 to 54 points), the A4 stimulus sheet contained 56 targets (small stars) pseudo-randomly interspersed with distracter items. The targets actually fell into six columns, with two additional targets which were located centrally. The experimenter clearly indicated the full extent of the sheet and crossed out the two central targets as an example to the subject. The subject was then asked to cancel the remaining small stars. The number of targets omitted in each lateral half of the sheet was counted. The star cancellation test sheet was divided into six areas (left-left, middle-left, right-left areas and right-right, middle-right, left-right areas) and was analyzed using the correct rate for six areas. 51 points were set as a cut-off value. 3.2 Special test with head mounted display (Figure 2) (a) Experimental apparatus The main experimental apparatus includes a digital camera, HMD (GT270, Canon Inc.), and a digital video camera. HMD is a glass type display method (270,000 pixel, effective pixel number is 99.99%, weight is 150 g) that consists of two TFT liquid crystal panels. The digital camera takes a picture of a test sheet on the desk, and HMD presents the subject from the digital camera. Moreover, the subject’s head movement was recorded by a digital video camera as a qualitative motion analysis. 256

CCD camera HMD

Real image

Modified visual input with HMD

Combined system of HMD and CCD camera

Zoom-in condition (ZI)

Zoom-out condition (ZO) (b) Special test 3

(a) Special test 1 and 2

Figure 3. Two special tests of USN with HMD.

(b) Assessments of USN with HMD (Figure 3) We attempted to find the degree that USN alters when the co-ordinate of the subject’s visual field was carried out as object-centered by HMD. Therefore, we used two different lenses of the digital camera in order to change visual field and then a HMD displayed the test sheet to the subject as the three special tests as follows; Special test 1: the zoom-in (ZI) condition which can display only the test sheet using combined HMD and a DV camera in the object-centered co-ordinate. Special test 2: the zoom-out (ZO) condition which can display 70 percent of the test sheet using combined HMD and a DV camera in the object-centered co-ordinate. Special test 3: the actual image condition (zero percent of reduction) which a combined HMD and a micro CCD camera simultaneously move to follow the subject’s movement in the egocentric co-ordinate.

3.3 Procedure The subjects sat on a wheelchair if needed or on a straight back chair sitting in an up-right position as a starting point. The test sheet was put on a desk and was placed at a midline of each subject’s body. All tasks were done without any restriction as to time. Eight subjects performed the common clinical test and special test 1 and 2. The subjects were first evaluated by a common clinical test without HMD and then two spatial tests with HMD. The line cancellation test was scored using the correct rate and then the score divided into two areas; right and left. The star cancellation test was scored using the correct rate for six areas (left-left, middle-left, right-left areas and right-right, middle-right, left-right areas) in which the test sheet was divided (Figure 1). All subjects performed in random order the common clinical test and two special tests (ZI, ZO). The examiner confirmed the HMD monitor as the display from the image of the digital camera. Moreover, one subject was performed the common clinical test and two special tests; special test 1 and 3 in both object-centered and egocentric co-ordinates.

4 DATA ANALYSIS All statistics were performed using SPSS statistical software (7.5.2J). An ANOVA or Student’s t test was used as a comparison between the common clinical test and the two special tests with HMD. Moreover, a Student’s t test or an ANOVA was used for a comparison within the line cancellation 257

100% 80% 60% 40% 20% 0% 1

2

3

4

1. Talking or communicating with others 2. Neglecting the left side of bed space 3. Eating 4. Grooming (self-care skills, washing, bathing, etc.) 5. Dressing

Figure 4.

5

6

7

8

9

10

6. Transferring (from a bed, to W/C, etc) 7. Locomotion 1: negotiating a W/C between doors, kerbs, etc. 8. Locomotion 2: the patient turns toward the direction of the affected 9. During PT exercise 10. During OT exercise

Ratio of USN symtoms in ADL.

test and the star cancellation test, respectively. Multivariate ANOVA tests were performed in each group and Shëffe post hoc tests were performed if significant differences were found at the 5% significance level. 5 RESULTS In this study, the average score of FIM-M of all subjects was more than 50 points. It indicates that the subject needs maximal or moderate assistance for achieving an adequate performance of ADL. As the common clinical test for USN, in the first evaluation of the frequency of presence of neglect for ADL (Figure 4), 75 percent of all subjects admitted a USN symptom in activities of dressing. For example, a patient with USN cannot easily put on their clothes on the left side. Moreover, 62.5 percent of the subjects admitted a USN symptom in activities of transferring, and locomotion (Figure 4). According to the motion analysis of head motion in the common clinical test, the subjects began searching from the right side in both the line and the star cancellation tests. In a normal performance, the head naturally rotated from the right to the left to follow a movement during the line cancellation test. However, the head movement to their left was insufficient for searching from the right side in both tests. For the line cancellation test under the common condition, the mean percentage of the correct answers at the left side in the test sheet was 94.4%. The right side was 100%. Nobody fell below the cut-off value (Figure 5) (Ishiai, 1999). For the star cancellation test under the common clinical test (Figure 6), the mean percentage of the correct answers at the leftleft area was 91.1%. The middle-left area was 89.2% and the right-left side was 84.4%. The mean percentage of the correct answers at the right-right was 92.9%, middle-right was 96.4%, and leftright area was 81.8%. Three subjects fell below the cut-off value as an abnormal result (Ishiai, 1999). For the special test with HMD, in the motion analysis of head motion, the subjects began searching from the right side in both the line and the star cancellation tests. However, seven subjects kept rotating only on the right side. For the line cancellation test under the ZI condition in the special test with HMD (Figure 7), the mean percentage of the correct answers at the left side in the test sheet was 61.8% while the right side was 92.4%. For the ZO condition, the mean percentage of the correct answers at the left side in the test sheet was 79.9%. The right side was 91.7%. In both ZI and ZO conditions, the left score was significantly greater than the right score (p < 0.05). There was a significant difference between the common clinical test and ZI conditions of the special test with HMD for the left side score (p < 0.05). For the star cancellation test under the ZI condition in the special test with HMD (Figure 8), the mean percentage of the correct answers at the left-left area was 60.7%. The middle-left area was 69.6% and the right-left side was 77.9%. The mean percentage 258

Mean percentage of correct answers (%)

n
8 100% 80% 60% 40% 20% 0% Left side of test sheet Right side of test sheet

Figure 5.

Mean percentage of correct answers of the line cancellation test under the common method.

140%

Mean percentage of correct answers (%)

120%

Correct answers of left-left (%)

100%

Correct answers of right-left

80% Correct answers of mid-left 60% Correct answers of mid-right 40% Correct answers of left-right

20%

Correct answers of right-right

0%

Figure 6.

Mean percentage of correct answers of the star cancellation test under the common method.

Mean percentage of correct answers (%)

n
8

Figure 7.

140% : Significant differences among the common, ZI, and ZO conditions. (p  0.05)

120% 100% 80%

Correct answers of left side (%)

60%

Correct answers of right side (%)

40% 20% 0%

Mean percentage of correct answers of the line cancellation test under the three conditions.

259

n
8

Mean percentage of correct answers (%)

120% 100% Correct answers of left-left (%)

80%

Correct answers of right-left 60% Correct answers of mid-left 40%

Correct answers of mid-right

20%

Correct answers of left-right Correct answers of right-right

0% Common

Figure 8.

ZI

ZO

Means percentage of correct answers of the star cancellation test under the three conditions.

100 80 60

Left side of the sheet

40

Right side of the sheet

20 0 Common test

Figure 9.

Special test 1

Special test 3

Percentage of correct answers of the line cancellation test in three tests; common test, special test 1, and special test 3.

of the correct answers at the right-right was 87.5%, middle-right was 92.9%, and left-right area was 87.0%. For the ZO condition, the mean percentage of the correct answers at the left-left area was 69.7%. The middle-left area was 70.8% and the right-left side was 77.9%. The mean percentage of the correct answers at the right-right was 97.9%, middle-right was 87.5%, and a left-right area was 92.4%. In the subject who performed the common test and special test 1, and 3 (Figure 9), the results showed that for the line bisection test under the common condition, the mean percentage of the correct answers at the right and left sides in the test sheet was 100%. In the OC condition, the left side was 44% and the right side was 94%. In the EC condition, the left side was 61% and the right side was 67%.

6 DISCUSSION The correct answer rate of the left space under ZI and ZO conditions was significantly lower than those in the common clinical test. Moreover, the correct answer rate which rose under the ZO condition was slightly greater than that of the ZI condition. It might be considered that the ZI 260

condition placed a greater focus on an object more than the ZO condition. These results indicated that when the patients with USN concentrated on an object, their USN symptoms were more aggravated. The subjects’ dressing, transferring, and locomotion of checklist by Halligan et al. (1991) indicated a high percentage of presence of USN symptoms. Although the common BIT did not sufficiently show USN where the correct answer rates score of the left space was more than 80%, the special test with HMD indicated USN where the correct answer rates score of the left space was about 60%. The HMD test may be able to better find a USN symptom which may not be easily discovered by the common clinical test. In our former study, the use of the HMD improved the neglect symptoms in all subjects who had right cerebral hemisphere damage (Tanaka, 2003). Rossetti et al. (1998) investigated the effect of prism adaptation on neglect symptoms, including the pathological shift of the subjective midline to the right. They reported that all patients exposed to the optical shift of the visual field to the right were improved in their manual body-midline demonstration and on their classical neuropsychological tests. Lee et al. (1999), Woo and Mandelmant (1983) also suggested the effectiveness of the Fresnel prism when placed on a spectacle lens for improving various visual–field losses. The improvement induced by the HMD indicates that a signal is given to the brain that stimulates the natural recovery process in the same manner as the prism adaptation method. Moreover, the HMD system may lead to the further correction of left neglect than a Fresnel prism placed on a spectacle lens. Since a high power Fresnel prism membrane for obtaining a wide field of view is not clear, the prism produces a distortion of a real image and has lowered capabilities of visual acuity. By contrast, the HMD has the possibility of obtaining various fields of view without deterioration of visual acuity. The HMD system may play an important role in the neuropsychological rehabilitation of unilateral spatial neglect as an evaluation device. Bowen et al. (1999) performed a systematic review of published reports. They found 17 reports which directly compared right brain damage (RBD) and left brain damage (LBD) and USN occurs more frequently after RBD than LBD as supported by a systematic review of the published data. However, an accurate estimate of the rates of occurrence and recovery after stroke could not be derived. They suggested that different USN disorders may exist, which may require type-specific rehabilitation approaches. Our system may have clinical implication for new assessment because HMD can change versatile visual input to fit each patient’s degree of USN. Because, a clinical assessment for USN may be able to use various images in HMD by a computer such as change of colors and partial enlargement or reduction of real image, and to produce suitable visual information in HMD for each patient who has USN. The result also showed that HMD evaluation could produce the condition of an object-centred allocentric co-ordinate and egocentric co-ordinate system to clarify the left neglect area which can not be easily observed in the clinical evaluation for USN. In conclusion, the results showed that the assessment of USN using an HMD system may clarify the left neglect area which can not be easily observed in the clinical evaluation for USN. Moreover, it might be hypothesized that the USN test using HMD may display a greater accuracy and be able to assess the occurrence and degree of USN to a greater degree more than the common clinical test. HMD can produce an artificially versatile environment as compared to the common clinical evaluation. ACKNOWLEDGEMENTS A part of this work was supported by the grant-in-aid for scientific research from Japan Society for the Promotion of Science, 2007–2009. REFERENCES Albert ML., (1973). Simple test of visual neglect. Neurology, 23: pp. 658–664 Bowen A, McKenma K, and Tallis C., (1999). Reasons for variability in the reported rate of occurrence of unilateral spatial neglect after stroke. Stroke, 30: pp. 1196–1202

261

Calpan B., (1987). Assessment of unilateral neglect: a new reading test. J Clinical and Experimental Neuropsychology, 9: pp. 359–364 Calvanio R, Petrone PN, and Levine DN., (1987). Left visual spatial neglect in both environment-centered and body-centered. Neurology 37: pp. 1179–1183 Denes G, Semenza C, Stoppa E, and Lis A., (1982). Unilateral spatial neglect and recovery from hemiplegia: follow-up study. Brain, 105: pp. 543–552 Diller L, and Weinberg J., (1977). Hemi-inattention in rehabilitation: evolution of rational remediation program. Adv Neurol. 18 : pp. 63–82 Granger CV, Hamilton BB, Keith RA, Zielezny M, and Sherwin FS., (1986). Advances in functional assessment for medical rehabilitation. Top Geriartr Rehabil. 1 (3): pp. 59–74 Granger CV, Hamilton BB, and Sherwin FS.The functional independence measure: a new tool for rehabilitation. In Eisenberg MG, Grzesiak RC (eds.), (1987). Advances in Clinical Rehabilitation. New York: SpringerVerlag Halligan PW, Cockburn J, and Wilson BA., (1991). The behavioral assessment of visual neglect. Neuropsychological Rehabilitation. 1 (1): pp. 5–32 Heilman KM. Neglect and related disorders. In KM Heilman and E Valenstein (Eds), (1979). Clinical Neuropsychology New York: Oxford University Press Isiai S., (1999). Behavioural inattention test, Japanese ed. Shinkoh Igaku Shuppan, Co., Ltd, Tokyo Karnath HO., (1994). Disturbed coordinate transformation in the neural representation of space as the crucial mechanism leading to neglect. Neuropsychological Rehabilitation, 4: pp. 147–150 Karnath HO., (1994). Subjective body orientation in neglect and the interactive contribution of neck muscle proprioception and vestibular stimulation. Brain, 117: pp. 1001–1012 Lee AG., and Perez, A.M., (1999). Improving awareness of peripheral visual field using sectorial prism. J. Am.Optom.Assoc. 70: pp. 624–628 Lee JH, Ku J, and Cho W., (2003). A virtual reality system for the assessment and rehabilitation of the activities of daily living. Cyberpsychol Behav. 6 (4): pp. 383–388 Pizzamiglio L, Frasca R, Guariglia C, Incoccia C, and Antonucci, G., (1990). Effects of optokinetic stimulation in patients with visual neglect. Cortex, 26: pp. 535–540 Prothero JD., (1993). The treatment of akinesia using virtual images. Master’s thesis, Industrial Engineering, University of Washington Rossetti Y, Rode G, Pisella L, Farne A, Li L, Boisson D, and Perenin M-T., (1998). Prism adaptation to a rightward optical deviation rehabilitates left hemispatial neglect. Nature 395: pp. 166–169 Tanaka T, Shirogane S, Ohyanagi T, Izumi T, Yumoto H, Ino S, and Ifukube T., (2003). Application of head mounted display system for left unilateral special neglect. 14th International Congress of the World Confederation for Physical Therapy, Proceedings, RR-PO-0982 Wade DT, Skilbeck CE, and Hewer, RL., (1983). Predicting Barthel ADL score 6 months after an acute stroke. Arch Phys Med Rehabil. 64: pp. 24–28 Weinstein EA, and Friedland RP., (1977). Hemi-inattention and hemisphere specialization: introduction and historical review. Adv Neurol. 18: pp. 1–13 Weiss PL, Naveh Y, and Katz N., (2003). Design and testing of a virtual environment to train stroke patients with unilateral spatial neglect to cross a street safely. Occup Ther Int. 10 (1): pp. 39–55 Wilson BA, Cockburn J, and Halligan PW., (1987). Behavioural inattention test. Thames Valley Test Company, England. Woo GC., (1983). Mandelman, T. Fresnel prism therapy for right hemianopia. Am J. Optom & Physiol Optics 60 (8): pp. 739–743

262

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Postural control using a vibratory feedback system for balance training in the elderly T. Tanaka, S. Ino & T. Ifukube Research Center for Advanced Science and Technology, The University of Tokyo, Japan

S. Shirogane Institute of Rehabilitation Science, Tokuyukai Medical Corporation, Japan

Y. Maeda Hokkaido University, Japan

Y. Oyama Tokeidai Memorial Hospital, Japan

S. Sugihara Sapporo Shuyukai Hospital, Japan

T. Izumi School of Engineering, Hokkaido Tokai University, Japan

ABSTRACT: Purpose – The purpose of this study was to analyze the effects of the vibratory feedback system between the force plate and vibratory stimulation device for maintaining good standing balance as a training technique. Participants – The Japanese subjects were divided into two groups, 22 young adult subjects (mean age of 24.3 years), and 60 elderly subjects (mean age of 74.6 years). Methods – The vibratory feedback system has three units, a vibratory device unit, a force plate unit, and a feedback control unit. Eight vibrators in each foot mat were activated if the trajectory of the center of pressure (COP) reaches a position, which has been determined by the examiner. The prospective target position was defined as 50% of the maximal displacement of the COP in each subject, which was obtained from the modified Cross-test. The subjects stood on the force plate. They tried to reach the target position in each direction for six seconds. Moreover, 60 elderly subjects performed a new balance training program using this feedback system. They trained three times a week during four weeks. The balance test was performed two times. The first test administered at the start of the month and the second test was given four weeks later. Results – The accuracy of the COP movement to the vibratory presentation distance was 83.8% in the elderly group, and 90.7% in young adults. The result showed that the maximal displacement of four directions (forward, backward, rightward, and leftward) of the second test was significantly greater than those of the first test. Discussion – The accuracy of the COP movement to the vibratory presentation distance showed a high accuracy that exceeded 80% in both groups. The result showed that it might be the improvement of the balance ability of the elderly group because of the new balance training program. Keywords: vibratory stimuli, standing balance, rehabilitation engineering, toes and soles, force plate, combined system. 263

1 INTRODUCTION Falling is one of the most serious problems of mobility impairment in the elderly. Death rates from falls per 100,000 persons in 1984 were 1.5 for those younger than 65 and 147.0 for those 85 or older in the United States (Lambert, 1988). The occurrence rate from all fractures for those older than 65 is 3.7 times greater than that of those younger than 65 in Japan (Matsuura, 1996). Deterioration in postural balance may be a major contributor for repeated falls, resulting in an impaired ability to correct postural disturbances experienced in everyday life. A few researchers have indicated the influence of sensory information on balance activity in the elderly and studied stability in aged persons with impairments in somatosensory, visual, and vestibular functions. Age-related loss of tactile sensation, in particular, is pervasive in the elderly and appears to correlate with impaired control of postural sway (Lord, 1991 and Kenshalo, 1986), as well as the risk of falling (Lord, 1994). Numerous studies support the important contribution of tactile sensation, from the plantar foot surface, in the control of balance (Okubo, 1982; Horak, 1990; Hamalainen, 1992; Wu, 1997). We found significant differences between young adults and an elderly group in the tactile-pressure sensation by using our new device (Tanaka, 1996). We suggest that the deterioration of toe tactile sensation of older adults might be more affected by their balance abilities than young adults (Tanaka, 1996). It might be possible to develop interventions to improve functional stability in the elderly by augmenting sensory input from the plantar foot surface (Maki, 1999). In previous studies, a simple approach was established to facilitate sensation mechanically using indentors within the footwear insole in order to reduce postural sway (Lord, 1994). Kavounoudias et al. (1998) had showed that a 100 Hz frequency vibration applied to the anterior or posterior foot zones of standing subjects caused specially oriented whole-body tilts depending on the zone stimulated. Balance testing is commonly performed using a force plate as a quantitative assessment of postural sway. The mechanical criterion for stability is maintenance of the center of gravity (COG) over the base of support (BOS) (Winter, 1987). The movement of the COG located within the BOS in a horizontal plane is related to the individual’s limit of stability and also an important index for the balance test. The changes of the BOS have been noted to reflect the postural sway and the balance strategy used to maintain stance. It is pointed out as a possible cause of the fall that an elderly person can not deal with the changes of the BOS for keeping dynamic standing balance. We developed the combined system between the force plate and our vibratory device for controlling standing balance for the elderly and disabled persons who have poor balance. The purpose of this study was to analyze the effect of the combined system between the force plate and our vibratory stimulation system for maintaining good standing balance. It is our belief that our combined system will be able to control body sway and to improve the center of pressure (COP) within the base of support (BOS). 2 METHODS 2.1 Subjects The subjects were divided into two groups, twenty two young adult subjects (mean age of 24.3 years), and 60 elderly subjects (mean age of 74.6 years). All subjects were independent in their activities of daily living. The criteria for inclusion in the study were that subjects had no history of auditory or vestibular disease. Subjects were excluded if (1) they had a chronic (orthopedic, neurological, or psychiatric) disease, (2) there was no fall reported within six months prior to testing, (3) they did not use assistive devices for standing. Informed consent was obtained from all subjects prior to data collection. 2.2 Experimental equipment The combined system has three units; (1) vibratory device unit, (2) force plate unit, (3) feedback unit. 264

Vibrator Control box

Vibratory stimulation mat

Figure 1. Vibratory stimulation device.

Figure 2.

Location of vibrators on the mat.

a) Great toe

b) First metatarsal bone head

c) Fifth metatarsal bone head

d) Calcaneal part

Figure 3. Vibrators setting positions on the foot.

2.2.1 Vibratory device unit Figure 1 shows the vibratory stimulation system which was composed of eight vibrators with sponge mats and a control unit. The vibrator used a small vibration motor (14 mm in diameter, 1.6 g in weight, FM31E which is made by the Tokyo Part Industry) which is commonly used in mobile telephone technology. The voltage of the vibrator used in this research was equal to or less than 4.5 V. The system can produce mechanical vibrations up to 100 Hz. The 8 vibrators were placed at positions under the great toes, the head of the 1st metatarsal bones, the head of the 5th metatarsal bones, and the calcaneal bones between the sponge mats (30 mm × 30 mm) as shown in Figure 2 and Figure 3. In the study, four local areas (forefeet, heels, right side of right foot, and left side of left foot) were stimulated by the static vibratory stimuli (Figure 4). 2.2.2 Force plate unit The instrument was a force plate (ECG-1010DSA1B, Kyowa Electronic Instruments Inc., Japan) for measuring body sway parameters (Figure 5). The force plate contained four transducers, one in each corner of the force plate, for the measurement of vertical forces to follow the postural sway. The signal from the transducers was amplified, and then digitized by a computer. The computer program provided several indices of the body sway calculated from the center of pressure (COP), the chief body sway indices are as follows: 1) The mean and maximal displacements (mm) of the left-right (X) and anterior-posterior coordinates (Y) of the COP. 2) The mean standard deviation 265

Vibration presentation area

Center of pressure

Forward: Both great toes, first metatarsal bone heads, fifth metatarsal boneheads; Right side: Right great toes, 1st and 5th metatarsal bone head, and calcaneal part; Left side: Left great toes, 1st and 5th metatarsal bone head, and calcaneal part; Backward: Both calcaneal parts

Figure 4. Vibratory stimulation patterns and corresponding directions.

VTR

F-scan system

Vibratory stimulation mat Signal

Signal

Control box Computer (feedback program) Force plate

Figure 5.

COP information

Composition of experiment equipment.

of lateral body (SDX) and anteroposterior (SDY) coordinates of the COP. The accuracy and hysteresis of the force plate are ±1% and ±0.1%, respectively. The sampling rate of the force plate was 100 Hz. 2.2.3 Feedback control unit The combined system needs the feedback for the subjects to inform their condition of standing balance. The COP data was recorded in real time in the display of the force plate. If the trajectory of the COP reaches a position of forefeet which has been determined by the examiner, the vibrators on forefeet will then be activated. 266

2.3 Evaluation The prospective target position was defined as 50% of the maximal displacement of the COP in each subject which was obtained from a modified Cross-test (Ishikawa, 1992 and Maeda, 2002). The modified Cross-test was executed for analyzing the four maximum displacements (forward, backward, leftward, and rightward) in order to obtain the stability limit of each subject. The modified Cross-test is deeply related to the BOS which is defined as the maximal displacement of the center of gravity. BS is an important index of the capacity of balance. The combined system between the force plate and the vibratory device could provide information on the subject the prospective target position by the vibratory stimuli. The actual position was obtained from the COP data which the force plate measured until the subjects moved closer and reached the prospective target position. The difference between the actual target position and the prospective target position known by vibratory stimuli was then analyzed. The subjects stood on the force plate for seven seconds as the starting position and then they tried to reach the maximal displacement in each direction for six seconds and finally they kept at the maximal position for seven seconds. We analyzed the data of starting position from two to six seconds and the data of maximal position from two to six seconds. 2.4 Study protocol At the first stage, 22 young adults and 24 elderly subjects were randomly selected. They stood relaxed on the force plate, and their COP was measured for five seconds as a starting position. Second, the Modified Cross-test was executed for analyzing the four maximum displacements (forward, backward, leftward, and rightward) in order to obtain the stability limit of each subject. The prospective target position was defined as 50% of the maximal displacement of the COP in each subject. Third, they were asked to move the COP in each direction to the limited line as close as they could without getting over it. When the COP is over it, vibration was given. For example, when the subject nearly reached 50% of the maximal forward displacement, the vibrator stimulation was given to his/her forefoot. During all trials, their arms were folded across the chest and their eyes were closed. They were asked to move the COP using only the ankle joint motion. Moreover, an additional 60 elderly subjects performed a new balance training program using this feedback system. The subjects performed the modified Cross-test which reached a maximal displacement of the COP in each direction as balance training. They trained for the three times a week during four weeks. The maximal displacement of the COP was evaluated two times. The first evaluation was administered at the start of the month and the second was given at the end of the four weeks later. 2.5 Statistical analysis All statistics were performed using SPSS statistical software (7.5.2J). The body sway parameters for young adults and the elderly group were compared using the Student’s t-test or Welch’s t-test and Mann-Whiney test at the 5% significance level. For the body sway parameters in the two periods; prior to the body tilting and post body tilting were all analyzed under four tilting directions. In the elderly group, the maximal displacement for each direction for the first and second evaluation was also compared. 3 RESULTS 3.1 COP position adjustment by vibratory stimulation presentation The vibration presentation distance used by the actual experiment was decided by 50% of the body sway amplitude as defined by the modified Cross-Test (Table 1). This distance shows the distance from the starting point (this position where the rest standing upright is averaged COP). How it moved accurately to the target position presented by the vibratory stimulation at an actual COP position was shown as “Accuracy”. The result was shown in Figure 7. 267

Table 1. Vibratory stimulation presentation distance in direction of each movement.

Elderly Group Young Adult Group ∗∗∗ :

Forward

Backward

Right side

Left side

36.1 ± 10.8∗∗∗

30.8 ± 7.2 37.0 ± 7.8

36.1 ± 8.2 37.0 ± 6.7

35.1 ± 8.4 38.7 ± 6.5

51.4 ± 6.6

Significant difference between the two groups at 0.001. 140.0 Elderly group 120.0

80.8

87.0

84.3

83.2

(%)

100.0 80.0 60.0 40.0 20.0 0.0 140.0

Young adult group 99.4

120.0

(%)

100.0

87.7

82.4

93.3

80.0 60.0 40.0 20.0 0.0

Forward

Backward

Right side

Left side

Figure 6. Accuracy of COP movement to vibratory stimulation target.

In the elderly group, the mean accuracy of four directions was 83.8% (Figure 6). In the young adult group, the mean accuracy of four directions was 90.7%. The elderly, as compared with the young adults, showed significantly lower ability to control COP using vibratory feedback given on foot soles. By the posture information feedback that they used foot base vibratory stimulation for, both groups were able to control the COP position together in an aim range. The result also showed that the maximal displacement of four directions (forward, backward, rightward, and leftward) of the second test was significantly greater than those of the first test in Figure 7.

4 DISCUSSION: POSTURE ADJUSTMENT BY THE VIBRATORY STIMULATION It is known that the vibratory stimulation influences the body sways from some researches. Kavounoudias, et al. (1998) were reported that the vibratory stimulation of 100 Hz is done to both forefoot and both calcaneal parts for three seconds in healthy young adults, and the COP moved 40–50 mm on the average. In other reports of Kavounoudias, et al. (1999), it is suggested the COP moves to the other side with the stimulation part in the lateral side. Moreover, Shirogane et al. (2004) showed that the vibratory stimulation was presented in the calcaneal parts of the young adults and the elderly people with the same vibrators as this research, and COP was displaced forward by about 5.4 mm. That is to say, presenting the 268

(mm)

Pre-training

Figure 7.

100 90 80 70 60 50 40 30 20 10 0

Post-training * *

Forward

* *

* *

* *

Backward Rightward Leftward

Differences between pre and post balance training data for the means of maximal displacement of the modified Cross test (n = 60) ∗ : Significant difference between pre and post training at 0.01.

vibratory stimulation to foot soles when standing upright causes the COP displacement and it has the possibility with constant directionality. The directionality is a direction opposite to the vibratory stimulation part (for example, when the vibratory stimulation is done to the both calcaneal part, the COP displaced forward). Besides, the physiological mechanism remains unanswered. The accuracy of the COP movement to the vibratory presentation distance was 80.8–87.0% in elderly group, and 82.4–99.4% in young adults. This result showed high accuracy that exceeded both of the 80% by two groups. It means that this vibratory stimulation is effective as the posture information presentation. Thus, there is a possibility that this device can be used as balance training equipment. In the comparison between two groups, the accuracy of the elderly group in forward and backward directions was more significantly lower than the young adults group. And other two directions, there were no significantly differences, but the elderly group was lower than the young adults. It revealed that the task accomplishment was difficult in the elderly people. From the results of the body characteristics, it is clear that the elderly group has a sensory-motor coordination, muscular power and the sensitivity dysfunctions, compared with the young adults. Consequently, it was considered that they had influenced this task accomplishment. The result also showed that the maximal displacement of four directions of the second test was significantly greater than those of the first test. It might be evidence of an improvement of balance ability which relates to their increment of BOS. In conclusion, being able to maintain posture within the arbitrary range by the vibratory stimulation became clear. It is possible to use it as balance training equipment by inducing COP to an arbitrary position. Moreover, this range is set in the stability limit, and there is a possibility that the danger of the fall can be avoided. ACKNOWLEDGEMENTS A part of this work was supported by the grant-in-aid for scientific research from Japan Society for the Promotion of Science, 2005–2006. REFERENCES Hamalainen, H., Kekoni, J., Rautio, J., Matikainen, E., and Juntunen, J., (1992). Effect of unilateral sensory impairment of the sole of the foot on postural in man: Implications for the role of mechanoreception in postural control. Hum Movement Sci, 11: pp. 549–561 Horak, F.B., Nashner, L.M., and Diener, H.C., (1990). Postural strategies associated with somatosensory and vestibular loss. Exp Brain Res, 82: pp. 167–177

269

Ishikawa, A et al.: Cross-test to Measure standing ability by voluntary movements. J Phys Ther Sci. 4: pp. 35–44, 1992 Kavounoudias, A., Roll, R., and Roll, J.P., (1998). The plantar sole is a “dynamometric map” for human balance control. Neuroreport, 9: pp. 3247–3252 Kavounoudia, A., Roll, R., and Roll, J.P., (1999). Specific whole-body shifts induced by frequency-modulated vibrations of human plantar soles. Neurosci Lett, 266: pp. 181–184 Kavounoudias, A., (1999): From balance regulation to body orientation: two goals for muscle proprioception information processing? Exp Brain Res. Jan; 124(1): pp. 80–8, 1999 Kenshalo, D.R., (1986). Somethetic sensitivity in young and elderly humans. J Gerontol, 41: pp. 732–742 Lambert, D.A., and Sattin, RW., (1988). Deaths from falls, 1978-1984. MMWR CDC Surveill Summ, 37: pp. 21–26 Lord, S.R., Clak, R.D., and Webste, I.W., (1991). Postural stability and associated physiological factors in a population of aged persons. J Gerontol, 46, M69–M76. Lord, S.R., Ward, J.A., Wiilams, P., and Anstey, K.I., (1994). Physiological factors associated with falls in order community-dwelling women. J Am Geriatr Soc, 42: pp. 1110–1117 Maeda, Y et al.: Static and dynamic balance activity in elderly adults. Bulletin of School of Health Sciences, Sapporo Medical University, 5: pp. 79–85, 2002 Maki, B.E., Mcllroy, W.E., Perry, S.D., and Norrie, R.G., (1999). Effect of facilitation of sensation from plantar foot-surface boundaries on postural stabilization in young and older adults. J Gerontol, 54A, M281–M287. Matsuura, J., (1996). A trend of national welfare. Health and Welfare Statistics Association. 43(12): pp. 4–29 Okubo, J., Watanabe, I., and Baron, J.B., (1981). Study on influence of the plantar mechanoreceptor on body sway. Aggressologie, 21: pp. 61–69 Shirogane S, Tanaka, T., Ino, S., Ifukube, T. et al.: A basic experiment of a standing balance control device using vibratory stimulation on soles. Mechanical Engineering Congress, 04: pp. 215–216, 2004 Tanaka, T., Noriyasu, S., Ino, S., Ifukube, T., and Nakata, M., (1996). Influence of the toe pressure and tactile sense of the great toe on dynamic standing balance in healthy subjects. IEEE Trans. Rehab. Eng., 4: pp. 84–90 Winter, DA. Sagittal plane balance and posture in human walking. IEEE Trans.Med.Biol., 9: pp. 8–11, 1987 Wu, G., and Chiang, J.H., (1997). The significance of somatosensory stimulations to the human foot in the control of postural reflexes. Exp Brain Res, 114: pp. 163–169

270

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

A study of static and dynamic balance abilities in the elderly Yusuke Maeda Hokkaido University, Sapporo, Hokkaido, Japan

Toshiaki Tanaka, Shuichi Ino & Tohru Ifukube Research Center for Advanced Science and Technology, The University of Tokyo, Japan

Satoshi Shirogane Institute of Rehabilitation Science, Tokuyukai Medical Corporation, Japan

Yohei Oyama Tokeidai Memorial Hospital, Japan

ABSTRACT: Purpose – One of the most serious problems of mobility impairment is the tendency of the elderly population to fall. A coordination of sensory and motor function is also indispensable for static and dynamic standing balance. The purpose of this study was to analyze the limitation of the base of support in relation to the center of gravity and to clinically assess the influence of sensory interaction of standing balance with aging. Methods – The subjects were divided into two groups, the young group (17 subjects, average age 22.5 years), and elderly group (10 subjects, average age 73.4 years). All subjects were independent in their activities of daily living. The study used two experiments. First experiment: After the subjects were asked to stand as relaxed as possible, they shifted from the starting position to the four directions of sway; anterior, posterior, right and left. They were asked to maintain their standing position at the maximal distance for each body sway as possible for 20 seconds. Second experiment: All subjects were tested by using the Clinical Test for Sensory Interaction in Balance (CTSIB). The subjects were asked to maintain standing balance for 30 seconds under the six different sensory conditions. Results – The data of the young group for the base of support were greater than that of the elderly group, especially during the anterior sway. According to the CTSIB, the body sway data of the elderly group were greater than those of the young group when both visual and somatosensory information were disturbed. Conclusion – The results suggest that evaluating postural stability which includes a few taskoriented approaches and assessments of the influences of sensory conditions could be made available as a clinical balance test and provide specific physical therapy treatment in the elderly. Keywords:

elderly adult, balance, base of support

1 INTRODUCTION The incidence of falls is an important social and health issue among the aged population. The fracture often makes their mobility decrease even if it isn’t fatal. It is essential to maintain and improve the balance ability to prevent falls among the elderly. Postural control needs to integrate motor and 271

sensory function. The three most important sensory information responsible for avoiding falling are visual, the vestibular, and the somatosensroy information. Age-related loss of their sensory information is associated with an increased likelihood of falling (Judge et al., 1995, Lord et al., 1994). Therefore, balance ability should be versatilely evaluated both sensory and motor functions for effective balance training. Elderly persons have a lowered physical ability as compared with the younger adults due to muscle weakness especially related to the ankle joint. Mistaking one’s balance strategy, and a diminished the limitation of stability are examples of lowered physical ability. We should examine these parameters and investigate balance from many points of view in order to evaluate the elderly’s range of mobility in detail. However quantifying balance ability needs to be proceeded by improving the specific measurement equipment. Research in the relation of evaluation-treatment-effect judgment has been very limited thus far. Therefore we must attempt to make a standardized evaluation scale that includes both static and dynamic situations in future research. The purpose of this study was to investigate the evaluation and treatment about static and dynamic upright stance posture of healthy people including both motor and sensory functions and to discuss the relationship between balance ability and physical condition. These analyses could show the characteristics of the base of support (BOS) of elderly persons compared to those of young adults. We would be able to indicate a useful evaluation and a special treatment method for the elderly or frail individuals in the physical therapy field. 2 METHODS The subjects were divided into two age related groups, the young adult group (17 subjects, average age 22.5 years), and the elderly group (10 subjects, average age 73.4 years). All subjects were independent in their activities of daily living. We examined the range of motion and muscle strength of the lower extremity (Richard et al., 1986), with a two-point discrimination analysis on the foot sole for sensory evaluation (Mackinnon and Dellon, 1985), as well as measurements of lengths for the upper and lower limbs in each subject. The maximum power with isometric contraction of each muscle was measured twice and the best value was adopted. The static and dynamic two-point discrimination tests were measured for the planter of the great toe, the ball of the great toe, and the heel as a sensory assessment. Next, we measured two kinds of quantitative balance experiments. First experiment: After the subjects were asked to stand as relaxed as possible, they shifted from the starting position to the four directions of sway; anterior, posterior, right and left. They were asked to maintain their standing position at the maximal distance for each body sway for 20 seconds or as well as possible. They were asked to hold their arms in front of their chest, and not to move their trunk, hip, and knee during the trial. If their foot left the floor or the subjects lost their equilibrium, the trial was retried. The maximum distance of COP sway was calculated by posturography. Furthermore, to know the distribution of foot pressure on the sole, we divided the foot into three different areas, the planter great toe, the ball of great toe, and the heel. The pressure of each area was measured when the subjects were shifting in their standing position by using the Foot Scan System. Second experiment: All subjects were tested by using the Clinical Test for Sensory Interaction in Balance (CTSIB) (Shumway-Cook and Horak, 1986). It is used by taking combinations of the three visual and the two support-surface conditions. Visual conditions included the use of a blindfold for eliminating any visual input and visual-conflict dome-headgear for purposely producing inaccurate input (Figure 1). The domeheadgear had a yellow tinge coloration and a lattice pattern design on the lens surface in order to disturb their visual acuity. The support-surface conditions included the use of a soft cushion that reduced the accuracy of the orientation information. Condition 1: Condition 2: Condition 3:

normal standing, standing on a solid surface. No sensory perturbation. standing on a solid surface and no visual input by use of a blindfold. standing on a solid surface and inaccurate visual input purposely given by use of the dome-headgear. 272

1

2

3

4

5

6

Vision

Normal

Absent

Inaccurate

Normal

Absent

Inaccurate

Surface

Normal

Normal

Normal

Inaccurate

Inaccurate

Inaccurate

Figure 1. The illustration of CTSIB.

Condition 4: Condition 5: Condition 6:

standing on a soft cushion and normal visual input. standing on a soft cushion and no visual input by a blindfold. standing on a soft cushion and inaccurate visual input purposely given by wearing the dome-headgear.

The subjects were asked to maintain a standing balance position for 30 seconds preceding the six different sensory conditions. The total trajectory body sway was calculated for each of the six conditions. 3 RESULTS The lower extremity muscle strength of the young adult group was 1.3–1.8 times stronger than the elderly group, especially so in the great toe area which was 2.1–2.4 times greater. And the total results of muscle strength were statistically significant. 2-point discrimination distance of the young adults was approximately 1.5 times as sensitive as the elderly for the planter of the great toe, the ball of the great toe, and the heel. The results of the first experiment show that elderly groups’ sway displacement was shorter in all four directions, especially anterior direction and the foot pressure at the sway side was more decreased than those of young adults (Figure 2). According to the CTSIB, the body sway data of the elderly group were greater than those of the young adult group when both visual and somatosensory information were disturbed in the Condition 5 and 6 (Figure 3). 4 DISCUSSION The results of the first experiment showed that the data of the young adults was significantly greater than that of the elderly in the anterior direction. In the other three directions, the data of young adults was only slightly greater than those of the elderly. All subjects mainly used the ankle strategy in the first experiment. However, the movement of dorsiflexion of the elderly was less than those of the young adults. Therefore, in the anterior direction, the data of the young adults was significantly greater than that of the elderly. The result of CTSIB showed that the amount of sway increased in both groups because of the limitation of sensory inputs. In the case of the elderly group, this tendency became remarkable as a result of aging. It can be assumed that elderly persons have problems with their ability to adjust in these situations that were changed by sensory inputs. 273

10 * 8 6 (cm)

Young adult Older adult

4 2 0

Anterior

Posterior

Right

* : p  0.05

Left

Figure 2. The sway displacement in each direction. *

3000

*

2500

(mm)

2000 Young adult Older adult

1500 1000 500 0

1

2

3

4

5

6

* : p  0.05

Figure 3. Total trajectory of the body sway (mm).

We should examine the body based on postural control to assess balance ability of healthy older adults. Consideration should be given as to the elderly motion, and a muscle strength test should be performed not only for isometric muscle activity but also for eccentric muscle activity in their lower extremities. Additionally, flexibility of joints and sensitivity of sensory systems especially in the foot sole should be examined. Importance should be given to measuring postural sway both in static as well as dynamic situations by posturograohy under various conditions. In conclusion, both static and dynamic balance assessments should be needed for improving poor balance and activities of daily living in the elderly. REFERENCES Judge, J. O., King, M. B., Whipple, R., Clive, J., and Wolfson, L. I., (1995). Dynamic balance in older persons: effects of reduced visual and proprioceptive input. J Gerontol A Biol Sci Med Sci, 50, M263–270 Lord, S. R., Ward, J. A., Williams, P., and Anstey, K. J., (1994). Physiological factors associated with falls in older community-dwelling women. J Am Geriatr Soc, 42: pp. 1110–1117 Mackinnon, S.E., and Dellon, A.L., (1985). Two-point discrimination tester. J Hand Ther. 10A, 6, Part.1: pp. 906–907 Richard W. Bohannon (1986): Test-retest reliability of hand-held dynamometry during a single session of strength assessment. Phys Ther. 66, 2: pp. 206–209 Shumway-Cook, A., and Horak, Fay B., (1986). Assessing the influence of sensory interaction on balance. Phys Ther. 66, 10: pp. 1548–1550

274

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Differences of visual information processing between younger and aged person Kimihiro Yamanaka Department of System Design, Tokyo Metropolitan University, Hino-shi Tokyo, Japan

Yuka Nakanishi Graduate School of System Design, Tokyo Metropolitan University, Hino-shi Tokyo, Japan

Mitsuyuki Kawakami Department of System Design, Tokyo Metropolitan University, Hino-shi Tokyo, Japan

ABSTRACT: The aim of this study is to clarify the cause of decreased visual function with aging, by considering the difference in visual information processing between younger and aged persons. In this study, the fixation point and saccadic movement were treated as evaluation indices and measured to examine visual information processing by an experimental approach. The results show the following. (1) Saccadic latency in shifting the gaze point from the fixation point to the target was longer in aged subjects than in young subjects, and this prolongation also occurred in aged subjects when shifting to closer gaze points. In addition, correction saccade probability was higher in aged than in young subjects, indicating that saccade precision diminishes as ageing advances. (2) Aged subjects were slower than young subjects in recognizing the characteristics of the target. A difference was also seen in recognition time according to the target type, irrespective of age.

1 INTRODUCTION In 2005, Japan’s aged population (aged 65 and above) reached a historic high of 20,560,000, or 20.04% of the total population, exceeding 20% for the first time (Ministry of Health, Labour and Welfare, 2006). Therefore, the problem of an ageing workforce and the influence on society in general cannot be ignored. In particular, the detrimental effect of ageing on visual information has led to a rapid rise in the number of traffic accidents involving aged drivers (Uno, 2004). The present study addresses these issues by comparing visual information processing in the young and aged, to examine the processing steps most affected by ageing and to elucidate the causes of reduced visual perception in the aged. 2 VISUAL INFORMATION PROCESSING To investigate the characteristics of visual information processing (Card et al., 1983 and Rasmussen, 1983) in the young and aged, we divided the sequence of processing steps involved in ‘detecting the target’, ‘shifting the gaze point’ and ‘recognizing’ into four steps, as shown in Figure 1. Step 1. Can the target (stimulus) be found? Step 2. Preparation time for recognition stage. Step 3. Accuracy of eye movement. Step 4. Time needed for recognition. 275

Peripheral visual field

Step1: Step 1:Measurement Measurementof of peripheralvisual visualfield field peripheral

Fixation point Step2: Step 2:Measurement Measurementof of latency of of saccadic saccadic latency

Step4: Step 4:Measurement Measurementof of recognitiontime time recognition

Useful visual field Visual stimuli (Japanese syllabary)

Figure 1.

Step3: Step 3:Measurement Measurementof of correctionofofsaccadic saccadic correction

Steps of visual information processing.

Liquid -crystal display Liquid-crystal display PC for stimulus presentation 60 (cm) 60(cm)

Reaction switch

Subject

Eye mark recorder (Nac)

Figure 2.

Operator Digital recorder (Teac)

Layout of experiment.

The goal of analysis was to determine which of these processing steps was the most influenced by ageing. Steps 1–4 can be divided into visual function and information processing. In this study, we analyzed steps 2–4, which involve visual information processing, in order to investigate perception, cognition and instant judgement in the aged. 3 EXPERIMENT 3.1 Outline of experiment 3.1.1 Method and condition of experiment The layout of the experiment is shown in Figure 2. The subject was seated, with the head held steady by a chinrest, and asked to gaze at a black fixation point displayed at an optical angle of 1◦ on an LCD screen (FUJITSU, VL-202VH) placed 60 cm in front of the subject. The display was a white background on which black stimuli were presented. The subject was fitted with an eyemovement recorder (NAC, EMR-8B) and asked to gaze at a fixation point in the upper right region of the display. When the subject’s gaze on the circular fixation point had been confirmed, a display program that presented stimuli according to the experimental paradigm was initiated. Visual targets to be recognized by the subject were presented by the program at random points on the display within a predefined range of timings. The subjects detected the target in the peripheral field without any eye movement, rapidly shifting their gaze to the target when the fixation point was removed, and pressing a button at the moment of recognition of the target’s nature (its form or signification), before orally describing the target. When the subject pressed the button, the signal was converted by a switchbox to a mouse signal and inputted to a PC that recorded the time of fixation-point removal, the coordinates of the presented target and the time of button depression. This allowed 276

us to calculate the time required to perform the sequence of visual information processes from the removal of the fixation point to the recognition of the target. In addition, any incorrect recognitions indicated by the subject’s oral report were rated as inaccurate information processing and excluded from the analysis. Three types of visual targets were used in the experiment: written characters, shapes and arrows, with eight different items in each condition. The display range was set in a preliminary test, which determined that stimuli would be presented within a range of 18.6–28.2◦ from the fixation point. In terms of the number of experiment repetitions, for young subjects, one session of 60 presentations was conducted for each condition, making a total of three sessions, whereas consideration of visual fatigue in the aged subjects meant that one session consisted of 20 presentations, with three sessions per condition making a total of nine sessions. In addition, the target display area was divided into three regions of increasing distance from the fixation point for use as comparison parameters. The average illuminance of the experimental environment was 373.3 lx, while the average luminance of the display was 69.9 cd/m2 . 3.1.2 Subjects The subjects were a young group, consisting of nine people aged 21–24 years (all males) and an aged group, consisting of five people aged 64–71 years (three males, two females). No subjects had any abnormalities of visual function, and all were able to recognize with the naked eye or corrected vision a visual target presented at an optical angle of 1◦ on a display 60 cm in front of them. 3.2 Evaluation indices All the evaluation indicators used in this study were derived from analysis of gaze-point data obtained from the measurement of eye movements. With regard to measurement precision, the time resolution was 60 Hz and the resolution of eye-movement detection was 0.1◦ . In order to differentiate and examine the sequence of processing steps, we detected the indicators described below. 3.2.1 Saccadic latency Saccadic latency in this study was defined as the time from removal of the fixation point to commencement of eye movement above a rotational speed of 35◦ /s as the subject began to shift their gaze to the visual target (Shimoda, 2005). 3.2.2 Saccade time The saccade time was defined as the time taken for the subject’s gaze to shift to the visual target after leaving the fixation point. Note that this saccade time included not only saccadic movement but also tracking and suspension of movement due to correction saccades. 3.2.3 Correction saccade probability The correction saccade probability was the probability of the target not being apprehended with a single saccade and therefore requiring a second or third saccadic movement (Latour, 1962). 3.2.4 Recognition time Recognition time was the time from visual apprehension of the target to recognition of its nature, as indicated by pressing a button. 4 RESULTS AND DISCUSSION Differences between young and aged subjects in each of the processing steps 2–4 are presented in Figures 3–6, with the x-axis showing the distance from the fixation point to the target and the y-axis showing the time taken or the probability. Figure 3 shows the results for saccadic latency, measured as preparation time for the recognition stage. As the graph shows, there was no change in saccadic latency in young subjects, even for 277

Saccadic latency (ms)

600 ** : p  0.01 300 Younger person Aged person 0

20.3

23.6

26.7

Distance from fixation point to visual stimuli (deg.)

Figure 3.

Result of saccadic latency.

Saccadic time (ms)

200

100 Younger person Aged person 0

20.3

23.6

26.7

Distance from fixation point to visual stimuli (deg.)

Result of saccadic time.

Correction saccadic probability (%)

Figure 4.

10 ** : p  0.01 5 Younger person Aged person 0

20.3 23.6 26.7 Distance from fixation point to visual stimuli (deg.)

Result of correction saccade probability.

Recognition time (ms)

Figure 5.

600

** : p  0.05

300

Younger person Aged person 0

20.3 23.6 26.7 Distance from fixation point to visual stimuli (deg.)

Figure 6.

Result of recognition time.

278

Recognition time (ms)

600

300 Younger person Aged person 0

Character

Object

Allow

Kinds of visual stimuli

Figure 7.

Relation between recognition time and kinds of visual stimuli.

targets far from the fixation point; whereas in aged subjects saccadic latency increased in the 26.7◦ condition, the largest displacement between target and fixation point. It is known (Taylar, 1957) that, in general, saccadic latency increases as saccadic movement rises above 20–30◦ ; however, in this case, it can be seen that aged subjects begin to show increased saccadic latency for saccades of shorter range than in younger subjects. Furthermore, in all conditions of displacement between the fixation point and the target, saccadic latency was longer in the aged than the young subjects, with a t-test confirming a significant difference between the two groups (error level = 1%). These results therefore show an increase in saccadic latency among aged subjects. Figure 4 shows the results of saccade time measurements used for evaluating the accuracy of eye movements. This figure shows that saccade time increases in both young and aged subjects as the distance from the fixation point to the target increases. There was no large difference between the two groups, and the absence of significant difference was confirmed by t-test. However, the results of measurements for correction saccade probability presented in Figure 5 indicate a large difference between young and aged subjects, and a significant difference was confirmed by t-test (error level = 1%). These results appear to indicate a reduction in saccade precision, due to ageing. In addition, in young subjects, the probability of failing to apprehend the target by the first saccade and of performing a correction saccade increased as the distance between the target location and the fixation point increased, suggesting that saccade precision is influenced by displacement of the target. However, in aged subjects, the correction saccade probability was also high for targets comparatively close to the fixation point. This clearly demonstrates the decrease in saccade precision, due to ageing. In Figure 6, although the target location and distance from the fixation point had no discernible influence on recognition times within either group, there is clearly a large difference in recognition times between groups. A significant difference was confirmed by t-test (error level = 1%). On the basis of these results, there appears to be a difference in recognition time between the young and aged subjects. Differences in recognition times within and between groups were also seen according to the type of target, as shown in Figure 7. In both young and aged subjects, recognition times were longest for shapes, followed by hiragana (Japanese syllabary) and arrows. An ANOVA (analysis of variance) was performed with the target types as factors; however, no significant differences were found in both young and aged subjects. Thus, the results of this study do not allow us to identify which targets were more easily recognized, but they do indicate that recognition times differ according to type of target. 5 CONCLUSIONS The main results obtained from this study are summarized below. 1) Saccadic latency in shifting the gaze point from the fixation point to the target was greater in aged subjects than in young subjects, and this prolongation also occurred in aged subjects when 279

shifting to closer gaze points. In addition, correction saccade probability was higher in aged than in young subjects, indicating that saccade precision diminishes as ageing advances. 2) Aged subjects were slower than young subjects in recognizing the characteristics of the target. A difference was also seen in recognition time according to the target type, irrespective of age. REFERENCES Card S. K., Moran T. P., and Newell. A., (1983). The Psychology of Human Computer Interaction, Lawrence Erlbaum Associates. Latour P. L., (1962). Visual threshold during eye movements. Vision Research, 2: pp. 261–262 Ministry of Health, Labour and Welfare, 2006, Population Survey Report, http://www.mhlw.go.jp/toukei/ Rasmussen, J., (1983). Skills, Rules, Knowledge: signals, signs and symbols and other distinctions in human performance models, IEEE Trans. on SMC, SMC-13, 3: pp. 257–267 Shimoda, H., (2005). Introduction to Physiological Experiments (2) Measurement of Ocular Indexes. Journal of Human Interface Society, Vol. 7, 2: pp. 139–144 Taylor E. A., (1957). The spans: perception, apprehension, and recognition as related to reading and speed reading, Am J Ophthalmol, Vol. 44, 4: pp. 501–507 Uno, H., (2004). Characteristics of an Elder Driver in Ordinary Driving Situation. JSAE Review, Vol. 25, 1: pp. 83–90

280

Chapter 6 Nurse

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Effects of disturbed sleep on work ability and well-being among European nurses D. Camerino, P.M. Conway, S. Sartori & G. Costa Fondazione Ospedale Maggiore, Policlinico Mangiagalli e Regina Elena – Department of Occupational and Environmental Health, University of Milan, Milan, Italy

1 INTRODUCTION Nurses’ shortage in Europe has drawn considerable socio-political attention and has urged studies from important institutions such as the Joint Programme for Working Life Research in Europe (SALTSA) and the UE Commission on the themes of working conditions, health and well-being of nurses, with the aim to find effective ways to repress premature exit from the profession (Hasselhorn et al., 2005). Norms and regulations about shiftwork arrangements have been also established to help safeguarding this profession. However, personnel shortage and nurses’striving for increased salary often imply that standards concerning work schedule are difficult to met or put into practice (Lee et al., 2007). This happens mostly because attempts to guarantee adequate coverage of vacant posts often lead to an increase in working hours, a reduction of rest time between shifts, too many consecutive night shifts or week-ends worked, low work time predictability, etc. Unfavourable work schedule may have relevant consequences on nurses’ performance, with an impact mainly on increased likelihood of accidents, serious mistakes, lowered quality of patient care (Reid et al., 1993; Todd et al., 1989; Vik and MacKay, 1982) and higher rates of sick leave (Panton and Eitzen, 1997, Reid et al., 1997). Negative effects on safety and performance during nightwork (that implies prolonged wakefulness, partial sleep deprivation and subsequent sleepiness) derive mainly from a desynchronization of the cycardian rhythms and an alteration in the homeostasis of normal body mechanisms (Åkerstedt, 2007; Rogers and Dingers, 2003). Furthermore, inappropriate sleep stage at awakening (such as that occurring during slow wave sleep) is an important factor in amplifying sleep inertia (Cavallaro and Versace, 2003). While association between shiftwork and chronic fatigue may be weak as the latter is not entirely attributable to ‘objective’ shiftwork conditions (Ruggero, 2003), the relationship between shiftwork and disturbed sleep is more consistent (Åkerstedt, 2006). Inter-individual differences still exist (Roger and Dinger 2003; Violani et al., 2006) and are mainly ascribable to personal ‘arousability’ (i.e. the individual level of activation to internal and external stimuli). Shiftwork may aggravate fatigue and increase likelihood of errors (Costa, 1997; Rogers AE et al., 2004; Vila B, 2006; Folkard et al., 2006; Caruso et al., 2006; Dorrian et al., 2006). Moreover shift working nurses are often exposed to cumulative sleep debt and this implies increased fatigue (Muecke, 2005; Gold et al., 1992; Lee, 1992; Tasto et al., 1978) that flows into sleepiness (Axelsson et al., 2005) and poor alertness. Perkins (2001) has illustrated the phenomenon of micro-sleep, in which a person may uncontrollably experience altered states of consciousness for periods up to 30 seconds. During micro-sleeps, there are phenomena such as concentration loss and heightened non-responsiveness to environmental stimuli. These states may arise during any activity, with their frequency being directly related to the duration of sleep deprivation. Recent evidence points to a rather strong relation between impaired sleep, safety and health, also as a result of factors such as work stress, poor diet and physical exercise regimes associated with the 283

social upheaval brought about by irregular working hours. In particular, poor sleep may be related to stress-related complaints such as cardiovascular diseases, diabetes and low mental health, including burnout (Åkerstedt, 2006; Caruso CC et al., 2006). Ulsugi et al. (2005) have found that occupational stress according to the Karasek’s Demand-Control (DCM) and the Siegrist’s Effort-RewardImbalance (ERI) was associated with sleep disturbances and the authors speculated about a possible hypothalamic-pituitary adrenal axis activation to explain this relationship (Ulsugi et al., 2005). Åkerstedt (2006), who regards good sleep as a stress antagonist, tested the hypothesis of a stress-sleep interaction by investigating immune, metabolic and endocrine physiological correlates of disturbed sleep. In keeping with Åkerstedt, Axelsson (2005) listed the essential functions of sleep: homeostatic, restorative, energy saving, thermoregulatory, host responses to infections, learning and memory consolidation. He found low levels of morning testosterone (i.e. an anabolic hormone known to be associated with fatigue and/or a sense of being unsuccessful) among workers dissatisfied with their shift system. This may indicate a physiological basis for tolerance to shiftwork. Further studies have been conducted on the reduction in performance among shift working nurses since the classic reports in this area by Bjerner (Bjerner, 1967; Fisher et al., 2002; Rogers AE et al., 2003). Rogers and Dinges (2003) reported lapses in attention and inability to stay focused, reduced motivation, compromised problem-solving, confusion or bewilderment, irritability or hostility, unusual tenseness or anxiety, memory lapses (particularly in short-term memory), impaired communication, faulty and/or slowed information processing and judgment, diminished ability to detect, and recognize the significance of subtle changes in a patient’s health, diminished reaction time, inability to deal with the unexpected, indifference and loss of empathy (Rogers et al., 2004). In general, areas of performance turned out to be differentially affected by circadian rhythms and sleep loss (Åkerstedt, 2007) and also to have a dose-response relationship with sleep (Wilkinson et al., 1982; Jewett et al., 1999; Van Dongen et al., 2003) even if individual differences in performance impairment remain an important issue (Åkerstedt, 2007). Hence, against this background and considering that the work ability index is designed to assess self-evaluation of on-the-job performance, health, well-being and also absenteeism, we speculate that impaired sleep in shift working nurses could affect the perception of their work ability. Moreover, in this study we also aimed at testing the sensitivity of the work ability index to poor sleep compared to other indexes of well-being, i.e. burnout, job satisfaction and thinking of quitting nursing.

2 METHODS 2.1 Sample and procedure The present study is based on data collected within the framework of the longitudinal Nurses’ Early Exit Study (NEXT-study), an investigation carried out in several European countries between 2002 and 2004 with the aim to identify reasons and circumstances of premature departure from the nursing profession. Among several personal and work-related characteristics enquired in the investigation, part of the data from the NEXT-study provided information also on work ability and other well-being parameters, work schedule and sleep. The NEXT-study was approved centrally by the University of Wuppertal in Germany and also locally in each national centre. Detailed objectives and procedure of the NEXT-Study have been yet described elsewhere (Hasselhorn et al., 2003; Hasselhorn et al., 2006). In each country involved in the NEXT study, selection of participants was conducted by means of a stratified sampling procedure, so as to reflect the national distribution of nursing staff by type of institution (hospital, nursing home, home care service), geographical spread, and ownership (public or private). A prospective questionnaire-based design was used for study conduction. Participants were assured about anonymity of data collection, directly by the research team or the contact person 284

and also by means of an informed consent. The first assessment was carried out between October 2002 and June 2003, depending on countries’ study planning, while in each case the follow-up assessment was conducted one year after baseline. In most countries, questionnaires were sent to participants via the institution’s internal mailing system. In few occasions, direct posting to the participants’ home addresses was used, with a return to the research institutions by means of a prepaid envelop. The self-administered questionnaires distributed at baseline (Time 0) and follow-up (Time 1) comprised several questions on socio demographic factors, the work environment (psychosocial and physical characteristics), attitudes, individual traits, health and well-being. The sample included in the present study consisted of registered nurses only (i.e. including staff nurses, head nurses and nurse directors) employed in hospitals (both state-owned and private) from eight European countries (Belgium, Germany, Finland, France, Italy, the Netherlands, Poland and Slovakia) that participated in both measurement occasions of the NEXT-Study. This sub sample selection was done in order to increase homogeneity in working conditions. In the countries included, a total of 10,301 hospital staff nurses responded to either the baseline or the follow-up questionnaires and were then considered for the present analysis. Sample size decreased in the multivariate models owing to the presence of missing values on the variables considered. 2.2 Measures 2.2.1 Sleep Sleep was assessed using four items originally developed by the NEXT-study group. The questions asked the respondents to assess their ‘sleep quantity’ (two items) and ‘sleep quality’ (two items). An overall ‘sleep’ scale was created combining the four items. The combined scale ranged from 1 to 5, with higher scores indicating better sleep. One missing item per participant was tolerated for scale construction, with the individual mean calculated on the data provided used to replace missing values. The ‘Sleep quality and quantity’ scale obtained a Cronbach’s alpha of 0.78. For the present study, a three-level exposure sleep variable was formed on the basis of tertiles of the sample distribution (with the lower tertile corresponding to a more disturbed sleep). 2.2.2 Outcomes Emotional exhaustion was measured using the six-item scale from the Copenhagen Burnout Inventory (CBI - Borritz Kristensen, 2001). Participants had to indicate on a five-point scale how often they ‘feel tired’, ‘physically exhausted’, ‘emotionally exhausted’, and how often they think: ‘I cannot take it anymore’, ‘feel worn out’, and ‘feel weak and susceptible to illness’. The answering categories ranged from ‘never/almost never’ to ‘(almost) every day’. Cronbach’s alpha was .89. Job dissatisfaction was measured using a four-item scale developed by Kristensen et al. (2005) to assess the extent to which workers are satisfied with their working conditions. Responses were given on a four-point scale ranging from ‘very unsatisfied’ to ‘very satisfied’. Cronbach’s alpha was .74. Thinking of quitting nursing was measured by a single item: ‘How often do you think about giving up nursing completely?’ Response categories ranged from ‘everyday’ to ‘never’. Emotional exhaustion, job dissatisfaction and thinking of quitting nursing have been dichotomized using scores above the 90th percentile of the respective scale distribution as the cut-off point for identification of disease cases. This was done to minimize the bias potentially occurring when odds ratios are used to estimate relative risks in cases where disease incidence is higher than 10%. Work ability was measured at follow-up by the Work Ability Index (WAI, Ilmarinen and Tuomi, 2004). The total score is calculated by summing the scores obtained at the seven items of the WAI (Tuomi et al., 1998). The range for the total score is 7–49 points, with higher scores indicating higher perceived work ability. Internal validity has been previously demonstrated (Eskelinen et al., 1991; Nygård et al., 1991) and the instrument turned out to be stable according to test-retest reliability analysis (De Zwart et al., 2002). In this study, we adopted a short version of the WAI. 285

This version differs from the original WAI instrument in that item no. 3 contains only 15 grouped medical conditions summarising the 51 conditions detailed in the usual version, thus improving ‘face validity’. Nübling et al. (2004) has developed an algorithm to allow comparability of the data obtained by the two versions and he found a good convergent validity. In this analysis, ‘poor work ability’ was used to identify disease cases. 2.2.3 Confounders The following confounding factors were included: work schedule (‘day work’, ‘irregular day working hours’, ‘shiftwork without nights’, ‘shiftwork with nights’, ‘night shift only’), gender, age, country, type of employment contract (fixed/temporary), occupational position (staff nurse, head nurse and nurse director), family status (‘living alone’, ‘living as the only adult with child/children’, ‘living with another adult’, ‘living with another adult and child/children’), health status (measured by means of item 3 of the WAI), work-to-family conflict (Netemeyer et al., 1996) and influence at work (measure developed by the NEXT Group). 2.3 Statistical analysis Baseline prevalence scores of impaired sleep were calculated according to sample characteristics and associations were tested by means of chi-square or Pearson’s correlations depending on variable type. Baseline associations between sleep and the outcomes were then evaluated in each shift pattern. Multivariate logistic regression analyses were conducted to assess the adjusted relationships between sleep at baseline and the one-year incidence of emotional exhaustion, job dissatisfaction, poor work ability and high thinking of quitting nursing. Odd ratios and 95% confidence intervals were obtained. Analyses were conducted using SPSS version 14.0.

3 RESULTS At baseline 87.6% were women, with age ranging from 19 to 63 years (X = 37.19, SD = ±8.7) and work seniority in nursing from 1 to 43 years (X = 15.6, SD = ±8.7); 88.2% were staff nurses, 91.7% had an unlimited work contract and 61.8% were shiftworkers (of whom 3.8% were permanent nightworkers). As reported in table 1, the highest baseline prevalence of subjects in the poor sleep tertile was observed among those aged 30–39 years old (43%) and over 59 (43.5%), those working on night shift only and on shiftwork with nights (51.7% and 41.9% respectively), while no differences were found for gender. A poorer sleep (continuous variable) was significantly associated with work-to-family conflict, influence at work and the number of diseased diagnosed by a physician (Pearson’s correlation coefficients of −.37, .13, −.19 respectively; p < .01 all). Irrespective of work schedule, a poorer sleep was significantly associated with emotional exhaustion, job dissatisfaction, frequent thinking of quitting nursing and poor work ability at baseline (not shown). Moreover, we did not find any significant interactive effects of work schedule and sleep on the outcomes measure at baseline (p > .10). Table 2 shows the result of the prospective analysis. After controlling for several confounders, being in the poor sleep tertile (compared to the good one) was associated with a higher disease incidence across all outcomes considered (Odds Ratios ranging from 1.34 to 2.88). Finally, being in the medium compared to the good sleep tertile was significantly associated with an increased incidence of poor work ability and job dissatisfaction (Odds Ratios of 1.84 and 1.44 respectively).

4 DISCUSSION Results of this study provide evidence about a sleep effect on job dissatisfaction, emotional exhaustion, work ability and thinking of quitting the profession among European nurses. These effects may be a direct consequence of reduced restorative benefits resulting from disturbed sleep, but 286

Table 1.

Prevalence of subjects by sleep and confounders.

N (%)

Poor sleep

Moderate sleep

Good sleep

Country Netherlands Belgium Germany Finland France Italy Poland Slovakia

743 (7.89) 662 (7.03%) 1,179 (12.51) 1,648 (17.49%) 987 (10.48%) 2,433 (25.83%) 1,228 (13.03%) 541 (5.74%)

19.25 46.06 31.55 34.65 40.02 46.03 40.80 43.07

40.11 35.95 33.33 30.04 28.37 34.81 28.91 27.54

40.65 17.98 35.11 35.32 31.61 19.15 30.29 29.39

Gender Male Female

1,164 (12.36%) 8,250 (87.64%)

37.03 38.85

37.46 31.72

25.52 29.43

Age <29 30–39 40–49 50–59 >59

1,551 (16.55%) 3,468 (37.0%) 3,141 (33.51%) 1,191 (12.71%) 23 (0.25%)

34.04 42.91 37.09 36.10 43.48

33.85 33.04 32.12 29.81 21.74

32.11 24.05 30.79 34.09 34.78

Employment contract Fixed Temporary

8,588 (91.71%) 776 (8.29%)

39.12 33.89

32.46 31.19

28.41 39.42

Occupational position Nurse director Head nurse Staff nurse

488 (5.20%) 621 (6.62%) 8,272 (88.18%)

27.46 33.33 39.75

32.79 31.08 32.48

39.75 35.59 27.77

Work schedule Day work Irregular day working hours Shiftwork without nights Shiftwork with nights Night shift only

1,183 (12.79%) 292 (3.16%) 2,059 (22.26%) 5,367 (58.03%) 348 (3.76%)

27.56 31.85 35.75 41.89 51.72

33.47 34.93 34.73 31.66 24.71

38.97 33.22 29.53 26.46 23.56

1,303 (13.00 %) 605 (6.04%)

35.00 46.12

33.46 28.43

31.54 25.45

2,608 (26.02%) 5,507 (54.94%)

35.01 42.49

31.90 32.29

33.09 25.22

Family status Living alone Living as the only adult with child/children Living with another adult Living with another adult and child/children”

Chi- square

365.42 (df 14) p < .0001

16.62 (df 2) p < .0001

82.63(df 8) p < .0001

15.81 (df 2) p < .0001

56.44 (df 4) p < .0001

145.35 (df 8) p < .0001

88.41 (df 6) p < .0001

they may be also indirectly related to negative after-effects such as sleepiness, inadequate eating habits, impaired social relationships, etc. As concerns restorative effects, it is known for example that slow wave sleep induces secretion of the growth hormone (Orr, 2001) and strengthens learning and memory processes (Gais et al., 2002; Huber et al., 2004). The perception of impaired work ability was significantly associated with poor or moderate sleep as speculated, yet not resulting as the more sensitive parameter. Instead, job dissatisfaction, followed 287

288

1.56 1.56 1.67 4.11 1.44 1.72 1.64 1.79 1.78 1.38 1.25

1.32 1.42 3.07 1.64

0.89 0.87 0.75 1.25 0.33 0.73 0.71 0.49 0.65 1.13 1.11

0.74 0.96 1.83 0.93

Nd = Not included since it is part of the Work ability Ind

6.13 5.56 5.08 7.32 4.22 19.01 9.34 2.13

95% CI

1.53 1.41 1.18 1.91 0.91 5.07 2.08 1.10

>3,83 (1–5) O.R.

Emotional exhaustion

1.34 1.19

0.96 1.09

2.06 1.71 1.32 1.09

1.06 0.86 1.16 0.99

0.72 0.69 0.56

1.78 1.53 1.60 0.88 0.66 1.34 1.52 0.76

1.06 0.94

0.69 0.90

1.08 0.96 1.20 1.03

0.51 0.43 0.66 0.52

0.56 0.53 0.39

1.17 1.00 1.00 0.57 0.38 0.86 0.88 0.58

1.70 1.51

1.33 1.32

3.91 3.05 1.46 1.16

2.23 1.74 2.05 1.88

0.91 0.89 0.80

2.72 2.34 2.55 1.35 1.14 2.07 2.65 0.98

>3 (range 1–5) O.R. 95% CI

Thought to give up health care job

2.88 1.44

0.98 1.17

0.94 1.13 1.44 1.24

1.79 0.77 1.20 1.06

1.17 1.23 1.29

3.91 3.26 2.65 3.48 2.33 9.29 5.35 1.87

2.26 1.11

0.73 0.99

0.54 0.75 1.33 1.18

1.07 0.43 0.85 0.75

0.91 0.95 0.92

2.05 1.71 1.35 1.84 1.16 4.95 2.69 1.35

3.67 1.88

1.32 1.39

1.59 1.70 1.58 1.30

3.00 1.35 1.70 1.48

1.50 1.60 1.82

7.46 6.21 5.21 6.58 4.65 17.43 10.62 2.57

<1,75 (range 1–4) O.R. 95% CI

Job Dissatisfaction

Multivariate Logistic Regression analyses for assessing the risk of becoming a new disease case one year later

COUNTRY vs The Netherland Germany 3.07 Finland 2.80 France 2.44 Italy 2.37 Belgium 1.96 Poland 9.91 Slovakia 4.41 GENDER 1.53 AGE BAND vs 16–29 yrs 30–39 yrs 1.18 40–49 yrs 1.16 50–69 yrs 1.12 SHIFT WORK vs day work regular hours day work other 2.27 only night shift 0.70 shift without night 1.12 shift with nights 1.08 POSITION vs “sister/charge nurse” deputy sister 0.95 other nursing staff 1.07 Work – family conflict 1.25 no. of current diseases 1.18 diagnosed by a physician UNLIMITED WORK CONTRACT vs limited 0.99 INFLUENCE AT WORK 1.17 SLEEP DISORDERS vs good sleep poor sleep 2.37 moderate sleep 1.24

Cut off

Table 2.

2.07 1.84

0.54 1.26

1.09 1.28 1.25 nd

0.64 1.65 1.27 1.14

1.11 1.61 3.00

3.49 1.32 2.01 1.34 0.92 4.49 0.74 1.05

1.34 1.17

0.26 0.92

0.43 0.61 1.07 nd

0.17 0.74 0.68 0.60

0.67 0.98 1.67

1.44 0.52 0.77 0.53 0.31 1.87 0.19 0.63

3.20 2.88

1.14 1.73

2.73 2.63 1.46 nd

2.31 3.73 2.39 2.14

1.82 2.64 5.38

8.43 3.34 5.25 3.42 2.72 10.78 2.71 1.74

<3 (poor category) O.R. 95% CI

Work Ability Index

by emotional exhaustion, seem to be the parameters with the highest sensitivity to poor sleep. Ekstedt et al. (2006) reported evidence concerning the relationship between sleep and burnout. The authors suggested that impaired sleep may play an important role in the development of fatigue or exhaustion in the burnout syndrome, via the activation of the hypothalamo-pituitary-adrenal stress axis (Ekstedt et al., 2006). As an explanation to such differences in sensitivity, it seems plausible that job dissatisfaction and emotional exhaustion represent a more immediate and general feeling of discomfort, whereas work ability is a more cognitively elaborated and long-lasting end-point. In fact, as a rule individuals must experience numerous occasions of unsuccessful behaviour to reach a state in which a perception of impoverished work ability takes place. Thinking of quitting is a more distal outcome and it is highly sensitive to external factor such as the availability of job alternatives (Hasselhorn et al., 2006). As for the limitations, our study did not take in consideration that the causal effects could also be reverse, i.e. that burnout and low work ability may also be determinant of poor sleep. For example, it is know that subjects with burnout usually report lower sleep efficiency, increased arousal and sleep fragmentation, spend more time awake and in stage-1 sleep, have less slow wave sleep and rapid eye-movement sleep, and also a lower delta power density in non rapid eye-movement sleep. Moreover, during follow-up there may have been subjects whose sleep had improved and other whose sleep had decreased and this may have reduced effect estimate effects. Unfortunately, lack of measurement taken on sleep at follow-up does not allow us to clearly determine the extent of these biases. In conclusion, in our study conducted on a huge and representative sample of nurses throughout Europe we have found a dose-response relationship between lower sleep and poor work ability. While shiftwork is a risk factor for disturbed sleep, it is poor sleep itself that mainly affects work ability. Therefore, our study provides additional support for an important role played by individual components in tolerance to shiftwork.

REFERENCES Akerstedt, T., (2006). Psychosocial stress and impaired sleep. Scandinavian Journal of Work Environmental Health, 32: pp. 493–501 Akerstedt, T., (2007). Altered sleep/wake patterns and mental performance. Physiology and Behavior, 90: pp. 209–218 Axelsson, J., (2005). Long shifts, short rests and vulnerability to shift work. (Stockholm: Reproprint AB). Axelsson, J., Ingre, M., Akerstedt, T., and Holmback, U., (2005). Effects of acutely displaced sleep on testosterone. The Journal of Clinical Endocrinology Metabolism, 90: pp. 4530–4535 Bjerner, B., (1967). The nurse & rehabilitation. Tidskrift for Sveriges Sjukskoterskor, 34: pp. 136–138 Borritz, M., and Kristensen, T. S., (2001). Copenhagen Burnout Inventory: Normative data from a representative Danish population on Personal Burnout and results from the PUMA study on Personal Burnout, Work Burnout, and Client Burnout. (Copenhagen: National Institute of Occupational Health). Caruso, C. C., Bushnell, T., Eggerth, D., Heitmann, A., Kojola, B., and Newman, K. et al., (2006). Long Working Hours, Safety, and Health: Toward a National Research Agenda. American Journal of Industrial Medicine, 49: pp. 930–942 Cavallaro, C., and Versace, F., (2003). Stage at awakening, sleep inertia and performance. Sleep Research Online, 5: pp. 89–97 Costa, G., (1997). The problem: shift work. Chronobiology International, 14: pp. 89–98 De Zwart, B. D., Frings-Dresen, M. H., and Van Duivenbooden, J. C., (2002). Test-retest reliability of the Work Ability Index questionnaire. Occupational Medicine, 52: pp. 177–181 Dorrian, J., Lamond, N., Heuvel, v. d., Pincombe, J., Rogers, A., and Dawson, D., (2006). A pilot study of the safety implication of australian nurses’sleep and work hours. Chronobiology International, 23: pp. 1149–1163 Ekstedt, M., Soderstrom, M., Akerstedt, T., Nilsson, J., Sondergaard, H. P., and Aleksander, P., (2006). Disturbed sleep and fatigue in occupational burnout. Scandinavian Journal of Work Environment and Health, 32: pp. 121–131

289

Eskelinen, L., Kohvakka, A., Merisalo, T., Hurri, H., and Wägar, G., (1991). Relationship between the self-assessment and clinical assessment of health status and work ability. Scandinavian Journal of Work Environment and Health, 17: pp. 40–47 Fischer, F. M., Rotenberg, L., Latorre, M., Soares, N. S., Rosa, P. L., and Teixeira, L. R. et al., (2006). Work ability of health care shift workers. What matters? Chronobiology Int. 64: pp. 1165–1179 Folkard, S., Lombardi, D. A., and Spencer, M. B., (2006). Estimating the circadian rhythm in the risk of occupational injuries and accidents. Chronobiology International, 23: pp. 1181–1192 Gais, S., Molle, M., Helms, K., and Bom, J., (2002). Learning-dependent increases in sleep spindle density. The Journal of Neuroscience, 22: pp. 6830–6834 Gold, D. R., Rogacz, S., Bock, S., Tosteson, N., Baum, T. M., and Speizer, F. E. et al., (1992). Rotating shift work, sleep, and accidents related to sleepiness in hospital nurses. American Journal of Public Health, 82: pp. 1011-1014 Hasselhorn, H. M., Tackenberg, P., Kuemmerling, A., Wittenberg, J., Simon, M., and Bertazzi, P. A. et al., (2006). Nurses’ health, age and the wish to leave the profession–findings from the European NEXT-Study. La Medicina del Lavoro, 97: pp. 207–214 Hasselhorn, H. M., Tackenberg, P., and Mueller, B., (2003). Working conditions and intent to leave the profession among nursing staff in Europe. (Stockholm: National Institute for Working Life) Hasselhorn, H., Tackenberg, P., Buescher, A., Simon, M., Kuemmerling, A., and Mueller, B. H., (2005). Work and health of nurses in Europe, results from the NEXT-Study. (Witten/Herdecke: Universitat) Huber, R., Gilardi, M. F., Massimini, M., and Tononi, G., (2004). Local sleep & learning. Nature, 430: pp. 78–81 Ilmarinen, J., and Tuomi, K., (2004). Past, present and future of work ability. (Helsinki: Finnish Institute of Occupational Health) Jewett, M.E., Wyatt, J.K., Ritz-De Cecco, A., Khalsa, S.B., Dijk, D.J., and Czeisler, C.A., (1999). Time course of sleep inertia dissipation in human performance and alertness. Journal of Sleep Research, 8: pp. 1–8 Kristensen, T. S., Hannerz, H., Hogh, A., and Borg, V., (2005). The Copenhagen Psychosocial Questionnaire – a tool for the assessment and improvement of the psychosocial work environment. Scandinavian Journal of Work, Environment and Health, 31: pp. 438–449 Lee, K. A., (1992). Self-reported sleep disturbances in employed women. Sleep, 15: pp. 493–498 Lee, S., McCann, D., and Messenger, J. C., (2007). Working time around the world: trends in working hours, laws, and policies in a global comparative perspective, (London and Geneva: Routledge and ILO) Muecke, S., (2005). Effects of rotating night shifts: literature review. Journal of Advanced Nursing, 50: pp. 433–439 Netemeyer, R. G., Boles, J. S., and McMurrian, R., (1996). Development and validation of work and family conflict and family-work conflict scales. J. of Applied Psychology, 81: pp. 400–410 Nübling, M., Hasselhorn, H. M., Seitsamo, J., and Ilmarinen, J., (2004). Comparing the use of the short and the long disease list in the Work Ability Index Questionnaire. In: Assessment and Promotion of Work Ability, Health and Well-Being of Ageing Workers edited by Costa G., Goedhard W., Ilmarinen J., (Amsterdam: Elsevier), pp. 292–295 Nygård, C. H., Eskelinen, L., Suvanto, S., and Ilmarinen, J., (1991). Association between functional capacity and work ability among elderly municipal employees. Scandinavian Journal of Work Environment and Health, 17: pp. 122–127 Orr, W., (2001). Sleep and Health. In: International Encyclopedia of the Social and Behavioral Sciences, edited by Smelser, N. J., Baltes, P.,B., (St Louis: Elsevier), pp. 14162–14166 Panton, D., and Eitzen, G., (1997). Dynamic roster: addressing fatigue and recovery issues in the workforce. Journal of Occupational Health and Safety, 13 (5): pp. 487–492 Perkins, L., (2001). Is the night shift worth the risk? RN, 64: pp. 65–66 Reid, N., Robinson, G., and Todd, C., (1993). The quantity of nursing care on wards working 8- and 12- hour shifts. International Journal of Nursing Studies, 30: pp. 403–413 Reyd, K., Roberts, T., and Dawson, D., (1997). Improvising shiftwork management: II: shiftwork and health. Journal of Occupational Health and Safety: pp. 439–450 Rogers, A.E., Hwang, W.T., Scott, L., Aiken, L., and Dinges, D., (2004). The working hours of hospital staff nurses and patient safety. Health affairs, 23: pp. 202–212 Rogers, N. L., and Dinges, D. F., (2003). Subjective surrogates of performance during night work. Sleep, 26: pp. 790–791 Ruggero, J., (2003). Correlates of fatigue in critical care nursing. Research in Nursing and Health, 26: pp. 434–444

290

Tasto, D. L., Colligan, M. J., Skjei, E.W., and Polly, S.J., (1978). Health consequences of shiftwork. (Cincinnati: National Institute for Occupational Safety and Health) Todd, C., Robinson, G., and Reid, N., (1989). 12-hour shifts: job satisfaction of nurses. Journal of Nursing Management, 1: pp. 215–220 Tuomi, K., Ilmarinen, J., Jahkola A, A., Katajarinne, L., and Tulkki, A., (1998). Work Ability Index (2nd ed.). (Helsinki: Finnish Institute of Occupational Health. Helsinki) Ulsugi, M., Saijo, Y., Yoshioka, E., Horikawa, N., Sato, T., and Gong, Y. et al., (2005). Relationships of occupational stress to insomnia and short sleep in Japanese workers. Sleep, 28: pp. 728–735 Van Dongen, H. P., Maislin, G., Mullington, D. F., and Dinges, D. F., (2003). The Cumulative Cost of Additional Wakefulness: Dose-Response Effects on Neurobehavioral Functions and Sleep Physiology From Chronic Sleep Restriction & Total Sleep Deprivation. Sleep, 26: pp.117–126 Vik, A. G., and MacKay, C., (1982). How does the 12-hour shift affect patient care? The Journal of Nursing Administration, 12: pp. 11–14 Vila, B., (2006). Impact of long hours on police officers and the communities they serve. American Journal of Industrial Medicine, 49: pp. 972–980 Violani, C., Lombardo, C., Lucidi, F., Devoto, A., and Mallia, L., (2006). Validità ed attendibilità di una scala italiana di arousability. In: Proceeding of XI Riunione annuale della Società Italiana di ricerca sul Sonno (Milano: direttivo SIRS), www.ricercasulsonno.it, pp. 25–27 Wilkinson, RT., and Houghton, D., (1982). Field test of arousal: a portable reaction timer with data storage. Human Factors, 24: pp. 487–493

291

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Physical/mental recovery and work ability of nurses in Japan Chizuko Miyamoto, Yuri Suenaga, Mizuho Ando & Natsuko Noda Management in Nursing, Tokyo Healthcare University, Tokyo, Japan

Yukiko Okui & Mayuri Hashimoto Management in Nursing, Gifu College of Nursing, Gifu, Japan

ABSTRACT: The working circumstances around Japanese nurses are not so good. In this research, our aim is to investigate both physical and mental recovery and the relation between recovery and work ability to build a magnetic hospital and to discuss effective methods for improving the current situation. Subjects are 208 nurses working on full time base of one university hospital with 750 beds in central part of Japan. We measured physical/mental recovery respectively by frequency on five degree scales. Self-estimated work ability was measured by Work Ability Index. Mean age of the subjects was 35.3 years old (range 22–59). By WAI categories, the percentage classified excellent was 5.3%, good 38.5%, moderate 50.0%, poor 6.3%. The subjects who felt physical recovery “almost every day” were 17.2%, “often” 33.0%, and 8.6% felt recovery “very seldom”. About mental recovery, 20.1% were felt recovery “almost every day”, 30.6% “often”, and 10.5% “very seldom”. At both physical/mental recovery, mean total WAI scores of subjects who could recover almost every day were significantly higher. Among WAI items, mental demand of the job, estimated work impairment due to diseases, own prognosis of work ability two years from now, and mental resources significantly related with both physical/mental recovery. At both physical/mental recovery, mean age of subjects who could recover almost every day were the highest. However weak and inverse correlation was identified between total-WAI score and age; this indicated older nurses’ work ability was lower than that of younger nurses. Among WAI items, the score of diagnosed diseases showed inverse relation with age. This meant older nurses had more diseases; while work ability with respect to mental demands of the job showed direct relation with age. Both physical/mental recovery had positive relation with WAI. This indicated that recovery could be a target of intervention for improvement of work ability. As regards the relation among recovery, age, and WAI items, the older nurses were supposed to cope with their diseases by utilizing their mental ability effectively. It was suggested that older nurses might have some good ways both to respond to the demands of their jobs and to recover well from their daily work without falling into tired out condition. Interventions to inform the nurses how important recovery is, and to convey the method to recover especially from older nurses to younger nurses, will be needed in the future to facilitate physical/mental recovery for nurses in any age and to build up a magnetic hospital.

1 INTRODUCTION The working circumstances around Japanese nurses are not so good. The nurses are always stressed physically and mentally. Under these situations, the number of nurses leaving jobs tends to increase gradually; especially one of 10 newly employed left the job within one year after they completed nursing school programs. 293

In 2001, we started a project using the Work Ability Model to collaborate with a university hospital nurses to build a magnetic hospital attracting both patients and nurses. At first, we explained our intervention by the word “to create margins”, which meant creating time for enjoying both nursing career and private life. But some nurses did not accept this concept of “margins”. The reason was that they thought they were working almost to death, thus it was impossible to create time for something more. From their reactions, we recognized that we needed to use another acceptable word for explanation of our concept. Then, we selected the word “recovery” instead of margins, and planned to investigate the nurses’ recovery in 2006. In this research, our aim is to investigate both physical and mental recovery and the relation between recovery and work ability, to examine the property to use the word “recovery”. This concept is used as the intervention target to build a magnetic hospital and to discuss effective methods for improving the current situation.

2 METHOD 2.1 Subject Nurses working on full time base of one university hospital with 750 beds in central part of Japan.

2.2 Questionnaire We measured physical/mental recovery respectively by frequency on five degree scales. Subjects were asked to choose one answer by how frequently they used to feel physical and mental recovery. 1 point was put to “I feel physical/mental recovery at the beginning of day-shift almost every day.” 2-point was put to the feeling to recover “often”, this meant “two times of three day-shifts”. 3-point to “sometimes” meant “once two day-shifts”, 4-point to “not frequent” meant “once three or four day-shifts”, and 5-point was put to recover “very seldom” meant “once five day-shifts or below”. Self-estimated work ability was measured by Work Ability Index (Tuomi et al., 1998). WAI Questionnaire composed of eight items, WAI-total score ranged from 49 to 7, classified: poor 7–27, moderate 28–36, good 37–43, and excellent 44–49.

2.3 Procedure The questionnaire was distributed to each subject through nursing organization with written explanation including the aim of this survey, and was collected personally by dropping it into a slit of set box so that anonymity was guaranteed.

2.4 Analysis The data were analyzed with SPSS for Windows. The one-way ANOVA was used to compare mean of groups and Pearson’s Correlation Coefficient was used to examine correlation between two variables.

3 RESULTS 3.1 Subject Response rate was 67.2% or 208 subjects (male 11 or 5.3%). Mean age of them was 35.3 years old (range 22–59), 55.5% of them were unmarried, and 80.0% were staff nurses. 294

3.2 WAI score Mean total-WAI score was 35.4 (range 15–48). By WAI categories, the percentage classified excellent was 5.3%, good 38.5%, moderate 50.0%, poor 6.3%. Weak and inverse correlation was identified between total-WAI score and age; this indicated older nurses’ work ability was lower than that of younger nurses. Among WAI items, the score of diagnosed diseases showed inverse relation with age. This meant older nurses had more diseases; while work ability with respect to mental demands of the job showed direct relation with age (Table 1).

3.3 Physical/mental recovery The subjects who felt physical recovery “almost every day” were 17.2%, “often” 33.0%, “sometimes” 17.7%, “not frequent” 23.0%, and 8.6% of nurses felt recovery “very seldom”. Mean physical recovery was 2.71 ± 1.25 point. About mental recovery, 20.1% were felt recovery “almost every day”, 30.6% “often”, 21.5% “sometimes”, 16.8% “not frequent”, and 10.5% “very seldom”. Mean mental recovery was 2.66 ± 1.28 point.

3.4 Relation between physical/mental recovery and age At both physical/mental recovery, mean age of subjects who could recover almost every day were the highest, and the mean age of subjects with very seldom recovery were the lowest. These relations were significant statistically by one-way ANOVA (Table 2).

Table 1.

Peason’s correlation coefficient among age and WAI scores. WAI items score

Age and WAI scores

Total-WAI score

1†

2†

3†

4†

5†

6†

7†

8†

Age

−.184**

.020

.048

.213***

−.240***

−.098

−.110

.001

−.034

** <0.01, *** <0.001 † :1: current work ability compared with the lifetime best, 2: physical demands of the job, 3: mental demands of the job, 4: number of current diseases, 5: estimated work impairment due to diseases, 6: sick leave during the past 12 months, 7: own prognosis of work ability two years from now, 8: mental resources

Table 2.

Relation between physical/mental recovery and age. Mean age

Frequency of recovery

Physical recovery

Mental recovery

Almost every day Often Sometimes Not frequent Very seldom

42.1 36.5 31.2 33.9 29.4

39.9 36.8 34.0 33.1 28.6

F = 7.280***

F = 5.070***

*** <0.001

295

Table 3.

Relation between physical/mental recovery and total-WAI score. Mean total-WAI score

Frequency of recovery

Physical recovery

Mental recovery

Almost every day Often Sometimes Not frequent Very seldom

37.4 36.4 35.3 33.5 32.3

37.1 37.3 34.7 31.8 33.3

F = 4.96***

F = 8.73***

*** < 0.001

Table 4.

Relation between physical/mental recovery and total-WAI score. WAI-items score : F value and significance

Frequency of recovery

1†

2†

3†

4†

5†

6†

7†

8†

Physical recovery Mental recovery

1.16 ns 2.33 ns

0.54 ns 2.31 ns

2.68* 4.87***

2.14 ns 0.42 ns

5.27*** 5.50***

0.38 ns 1.64 ns

7.98*** 11.00***

5.79*** 11.03***

∗ <0.05,

*** <0.001 current work ability compared with the lifetime best, 2: physical demands of the job, 3: mental demands of the job, 4: number of current diseases, 5: estimated work impairment due to diseases, 6: sick leave during the past 12 months, 7: own prognosis of work ability two years from now, 8: mental resources † :1:

3.5 Relation between physical/mental recovery and total-WAI score Physical/mental recovery related with total-WAI score was significant statistically. At both physical/ mental recovery, mean total WAI scores of subjects who could recover almost every day were higher, and subjects who could recover not frequently or very seldom showed lower WAI scores (Table 3). 3.6 Relation between physical/mental recovery and WAI items Among WAI items, as the following Table 4 shows, (a) mental demand of the job, (b) estimated work impairment due to diseases, (c) own prognosis of work ability two years from now, and (d) mental resources significantly related with both physical/mental recovery by ANOVA (Table 4).

4 CONCLUSION Both physical/mental recovery had positive relation with WAI. This indicated that recovery could be a target of intervention for improvement of work ability. As regards the frequency of subjects’ recoveries, less than 20% of nurses could recover almost every day. It was estimated that many nurses were working without sufficient recovery until weekend or off days. However, the younger nurses who had fewer diseases than older ones did not show better work ability with respect to physical demands of job, and they showed worse recovery. And the relation between recovery and WAI items showed that insufficient work ability for mental demands and low mental resources 296

might be the factors of bad recovery seen among younger nurses. From these results, it can be affirmed that the younger nurses were over-loaded physically and mentally. Then the younger nurses perceived that it was too difficult to continue the work under the same situation. On the other hand, older nurses were supposed to cope with their diseases by utilizing their mental ability effectively. It was suggested that older nurses might have some good ways both to respond to the demands of their jobs and to recover well from their daily work without falling into tired out condition. Interventions to inform the nurses how important recovery is, and to convey the method to recover especially from older nurses to younger nurses, will be needed in the future to facilitate physical/mental recovery for nurses in any age and to build up a magnetic hospital. REFERENCES Japanese Nursing Association. http://www.nurse.or.jp/toukei/index.html Miyamoto, C. et al., (2007). Work Ability among Nurses in Japan. In: 2nd International Occupational Health Nursing Conference, Bangkok. Tuomi, K. et al., (1998). Work Ability Index, 2nd edition, Finnish Institute of Occupational Health.

297

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Relationship between depersonalization syndrome and medical malpractice among Japanese nurses Miyuki Sugiura Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Masataka Hirosawa Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

Yasuyuki Yamada & Yasunobu Nishi Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Sumio Tanaka & Motoki Mizuno Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

ABSTRACT: Depersonalization syndrome, which includes symptoms such as loss of feelings of agency and alienation from the perceptual world, may be one of the predicting factors of medical malpractice among nurses. However, there have been few studies which have shown a relationship between depersonalization syndrome and medical malpractice. The purpose of this study was to investigate the possible relationship between depersonalization syndrome and medical malpractice. A total of 627 Japanese female nurses completed questionnaires which included the Cambridge Depersonalization Scale (CDS) and questions regarding frequency of medical incidents and accidents. We divided participants into non-malpractice, incident and accident groups according to whether or not they had experience of incidents and accidents and compared scores of total CDS and its subscales among these groups. These results showed that total CDS scores in incident and accident groups were higher than in non-malpractice group. Score of Anomalous Body Experience (CDS subscale) in incident group was higher than in non-malpractice group. Score of Alienation from Surroundings (CDS subscale) in accident group was higher than in non-malpractice group. These results suggest that depersonalization syndrome (or tendency) is possibly one of the predicting factors of medical incidents and accidents. Therefore, we need to support nurses who experience this syndrome before malpractice occurs. Furthermore, it is pointed out that depersonalization syndrome is accompanied by distress and exhaustion. Therefore, it is important that we make efforts to establish ideal work management systems to prevent depersonalization.

1 INTRODUCTION Medical malpractice causes patients serious risks and makes healthcare costs increase. Additionally, malpractice negatively affects nurses in terms of professional and personal status, confidence and practice. These results cause a lowering of quality in healthcare service. Thus, the reduction of malpractice has become a major concern for society. In recent years, the mental and physical condition of nurses, involving such problems as depression, obsession (Yoshida et al., 2004) and fatigue (Carlton and Blegen, 2006) have also been regarded as factors related to malpractice. However, the nursing profession involves various 299

job-related stresses such as time pressure, understaffing, interpersonal relationships and so on. Therefore, various types of mental problems are likely to be seen among nurses. Depersonalization syndrome seems to be one of the mental problems that appears commonly among nurses. It is well known that depersonalization syndrome accompanies most mental disease, emotional stress, somatic disease and exhaustion (Nuller, 1982). Females are said to experience this syndrome at least two times more than men (DSM-IV-TR) and in particular, those who are called “intellectual” experience much more depersonalization (Tateyama, 1998). The nursing profession involves very stressful work and it is one of the common careers for intellectually-inclined women. Therefore, the nursing profession may be one of the professions which is closely related to the depersonalization syndrome. Depersonalization is defined by DSM-IV-TR as “a feeling of detachment or estrangement from one’s self. The individual may feel like an automaton, or as if he or she is living in a dream or movie. There may be a sensation of being an outside observer of one’s mental processes, one’s body, or parts of one’s body. Various types of sensory anesthesia, lack of affective responses, and a sensation of lacking control of one’s action, including speech, are often present. During the depersonalization experience, patients have enough ability to recognize the real world”. Popular symptoms of depersonalization are visual depersonalization, altered body experience, emotional numbing, loss of agency feeling, and changes in subjective experiencing (Sierra and Berrios, 2001). These symptoms may prevent sufferers from accomplishing their day-to-day business and may lead to malpractice, related to patients’ lives. However, until now there has been no study which has verified the relationship between depersonalization syndrome and malpractice. Therefore, the purpose of this study was to investigate the relationship between depersonalization syndrome and malpractice.

2 METHODS 2.1 Participants Participants were 627 female nurses from two university hospitals. Their mean age was 26.23 years (SD = 4.23). The study was approved by the ethical committee of the university hospital system. Nurses participated in this investigation after informed consent was obtained. The investigation was performed in 2007. 2.2 Measures We used the following self-report instruments in this study: 1. The Cambridge Depersonalization scale (CDS). The CDS comprehensive instrument contained 29 items addressing the complaints classically associated with depersonalization syndrome. (Sierra and Berrios, 2000). It has four subscales (Anomalous Body Experience, Emotional Numbing, Anomalous Subjective Recall, Alienation from Surroundings) (Sierra et al., 2005). 2. Malpractice In order to assess experience of malpractice, we asked participants about frequencies of medical incidents and accidents in the past six months. 2.3 Statistical analysis First, we divided participants into non-malpractice, incident and accident groups according to whether or not they had experience of incidents and accidents. The non-malpractice group included nurses who had no experience of either incidents or accidents. The incident group included nurses who had experienced only incidents. The accident group included nurses who had experienced accidents. Secondly, Kruskal Wallis test and Scheffe’s test were carried out in order to compare 300

Table 1.

Comparison of scores of total CDS and its subscales among three groups. 1. nonMalpractice N = 217

2. Incident N = 251

3. Accident N = 159

M

M

M

SD

SD

SD

χ2

p

Sheffe 1<2

Anomaious Body Experience Emotional Nurmbing Anomalous Subjective Recall Alienation from Surroundings

1.59

5.37

2.70

6.67

2.33

5.91

7.20

∗

1.63 1.70

4.43 3.60

2.14 2.37

4.96 4.64

2.49 2.68

5.40 5.67

3.62 4.35

n.s. n.s.

1.83

3.81

2.35

4.35

3.02

5.63

5.94

†

1<3

Total CDS

7.32

18.13

10.65

21.71

11.22

22.12

8.51

∗

1 < 2, 1 < 3

† p < 0.1, ∗ p < 0.05

scores of the CDS and its subscales among these groups. Statistical analysis was carried out with the SPSS version 15.0. 3 RESULTS Score of total CDS in incident group was significantly higher than in non-malpractice group (p < .05) and in accident group it showed a higher tendency than in non-malpractice group (p < .1). Score of Anomalous Body Experience in incident group was higher than in non-malpractice group (p < .05). Score of Alienation from Surroundings in accident group showed a higher tendency than in non-malpractice group (p < .1). There were no significant differences between three groups in scores of Emotional Numbing and Anomalous Subjective Recall (Table 1). 4 DISCUSSION 4.1 Relationship between depersonalization syndrome, medical incidents and accidents Nurses in incident and accident groups showed higher depersonalization tendency than nurses in non-malpractice group (Table 1). This suggests the possibility that depersonalization tendency is one of the predicting factors of malpractice. Therefore, when we find some depersonalization syndrome in nurses we may need to transfer them to a secure work place in order to prevent malpractice. When serious depersonalization appears in nurses we need to exhort them to take a leave of absence from duty in order to protect them from malpractice. On the other hand, depersonalization syndrome accompanies emotional stress, somatic disease and exhaustion (Nuller, 1982). So, problems in day to day practice such as time pressures, understaffing, and excessive work load may lead to depersonalization. Problems in hospital systems may be one of the causative factors of depersonalization. It seems that problems in working management systems relate indirectly to malpractice caused by depersonalization. Therefore, it is important to pay attention to work systems which predict depersonalization. 4.2 Relationship between symptoms of depersonalization and medical incidents and accidents Score of “Anomalous Body Experience” was higher in incident group, and score of “Alienation from Surrounding” was higher in accident group than in non-malpractice group. 301

“Anomalous Body Experience” was composed of items regarding feeling of out of body experiences, loss of agency feeling, and automatic and mechanical body feeling. “Alienation from Surroundings” was composed of items regarding feeling of not being real, being cut off from the world, and having a veil between oneself and the outside world. It may be very difficult to express to others feelings of “Anomalous Body Experience” and “Alienation from Surrounding”, so the nurse’s colleagues may not be able to notice such problems. However, in the case of “Anomalous Body Experience”, the unusual behaviour used to cope with this experience (induced compulsiveness to make sure of one’s real body) makes it possible to draw colleagues’ attention to the fact that the nurse has some trouble in her mind. On the other hand, in the case of “Alienation from Surrounding”, if colleagues turn attention to the nurse’s trouble and give her some notice, their word of notice will not likely reach her mind. This indicates that both the nurses with alienation and their colleagues need much effort to prevent malpractice and in order to prevent accidents. Therefore, support systems can often function adequately to deal with the experience of “Anomalous Body Experience”, but cannot deal with “Alienation from Surrounding” before accidents occur.

5 LIMITATIONS AND FUTURE STUDY In this study, we used a self-reported questionnaire to assess depersonalization syndrome and malpractice. Because of this, we may not have been able to evaluate depersonalization syndrome and malpractice perfectly. Additionally, we focused on depersonalization as a factor related to malpractice. However, since malpractice can be caused by various factors, we should consider influences from other factors related to malpractice in future study.

6 CONCLUSION Depersonalization symptoms (or tendency) may be one of the predicting factors of malpractice. Therefore, we need to help nurses who have these problems deal with their symptoms (or tendencies) and improve working conditions before malpractice occurs. REFERENCES American Psychiatric Association, (2000). Diagnostic and statistical manual of mental disorder (4th ed.). (Wahington, DC) Carlton, G. and Blegen, M.A., (2006). Medication-related errors: a literature review of incidence and antecedents. Annual Review of Nursing Research, 24: pp. 19–38 Nuller, YL., (1982). Depersonalization-symptoms, meaning, therapy. Acta Psychiatrica Scandinavica, 66 (6): pp. 451–458 Sierra, M. and Berrios, G.E., (2000). The Cambridge Depersonalization Scale: a new instrument for the measurement of depersonalization. Psychiatry Research, 93 (2): pp. 153–164 Sierra, M. and Berrios, G.E., (2001). The phenomenological stability of depersonalization: comparing the old with the new. The Journal of Nervous & Mental Disease, 189 (9) Sierra, M., Baker, D., Medford, N. and David, A.S., (2005). Unpacking the depersonalization syndrome: an exploratory factor analysis on the Cambridge Depersonalization Scale. Psychological Medicine, 35 (10): pp. 1523–1532 Tateyama, M., (1998). Depersonalization. In: Encyclopedia of clinical psychiatry Vol.1, edited by Marushita, M., (Tokyo: Nakayama-Shoten), pp. 196–207, (In Japanese) pp. 629–636 Yoshida,Y., Otsubo, T., Takenaka, K., Ikawa, T., Owashi, T., Takashio, O., Koda, R., Aoyama, H., Matumura, K., and Kamizima, K., (2004). The factors related to accident proneness among hospital nurses. Clinical Psychiatry, 46 (7): pp. 723–730 (In Japanese)

302

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Effect of continuous exercise on job stress among Japanese nurses Yujiro Kawata, Yasuyuki Yamada, Miyuki Sugiura & Yasunobu Nishi Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Motoki Mizuno, Masataka Hirosawa & Sumio Tanaka Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University school of Health and Sports Science, Chiba, Japan

ABSTRACT: The purpose of this study is to examine the effect of continuous exercise on job stress among Japanese nurses based on the stress model of Lazarus and Folkman (1984). The survey was administered to female Japanese nurses in 2007. The total number of valid respondents was 221. The questionnaire consisted of the Nursing Job Stressor Scale (NJSS), the Self-Rating Depression Scale (SDS) and questions regarding frequency of exercise in a week. The subjects were divided into continuous exercise group (CE) and non-continuous exercise group (non-CE). As the results of t-test, there was no significant difference of NJSS between CE and non-CE. Moreover there was no significant difference of SDS between CE and non-CE. In the correlation analysis, job stressors were not related to depression in CE. Meanwhile, job stressors related to human relations and work loads were associated with depression only in non-CE. The results of this study indicated that although there were small differences in the scores of NJSS and SDS between CE and non-CE, there seemed to be a relationship between job stressors and depression only in non-CE. Therefore, from the results of this study, we argue that continuous exercise has a modifying effect on job stress among Japanese nurses. 1 INTRODUCTION During the last two decades, psychological stress for nurses has been paid much attention in Japan. The long duration of stress in nurses can bring much disadvantage to patients and hospitals as well as to the nurses themselves. Thus, it is extremely important to verify the source of their job stress and to consider a plan to improve their circumstances. In this study, we focus on the role of continuous exercise as a stress coping as the first attempt to provide stress management for nurses. If exercise can be shown to have a role of positive stress coping, which can diminish negative stress responses such as depression, we could gain much informative knowledge regarding stress management among workers. Psychological stress has been understood according to the relationship between stressor and stress response which was cited in the stress model of Lazarus and Folkman (1984). In this model, stress coping has been given particular attention and much studied as a key concept to approach psychological stress problems. In this study, we verify the relationship which may exist among continuous exercise, perception of job stressors and depression, which is regarded as a result of stress response among Japanese nurses, and we discuss the possible effect of continuous exercise on job stress among Japanese nurses. 2 METHODS 2.1 Participants The questionnaire investigation was administered to 716 female Japanese nurses in 2007. The questionnaire was taken to a university hospital in the capital area of Japan. The total number of 303

valid respondents was 221. The average age was 25.1 (SD = 3.21) years old. Most of the participants were not assigned to a managerial position and were not married. 2.2 Measure 2.2.1 Nursing Job Stressor Scale (NJSS) The subjects’ job stressor were assessed by means of the Nursing Job Stressor Scale (NJSS; Higasiguchi and Nakagawa, 2003) This scale was developed to assess perceived job stressors and was composed of following seven subscales; “conflict with other nursing staffs”, “nursing role conflict”, “conflict with physicians/autonomy”, “dealing with death and dying”, “qualitative work load”, “quantitative work load” and “conflict with patients”. Each response was reported on a 5-point Likert-scale ranging from 0 (strongly disagree) to 4 (strongly agree). The scores of each subscale are gained by summing the scores of each item. 2.2.2 Self-Rating Depression Scale (SDS) The Self-Rating Depression Scale (SDS; Zung, 1965) was designed to assess an individual’s depression. This scale is composed of 20 items including the major mental and partially physical symptoms of depression. The validity and reliability of the SDS for Japanese have been reported (e.g. Kawada et al., 1999). The score of the SDS is the sum of 20 items. Each response was reported on a 4-point Likert-scale ranging from 1 (strongly disagree) to 4 (strongly agree). 2.2.3 Continuous exercise The item used for assessment was the question “how many times do you have exercise in a week?” Respondents were asked the number of times they exercised in a week. In this study, we regarded the nurses who exercise continuously (CE) as those who answered that they exercised continuously at least once a week. 2.3 Data analysis The subjects were divided into continuous exercise group (CE) and non-continuous exercise group (non-CE). For the analysis, first, exploratory factor analysis was carried out to examine validity and reliability of NJSS, then t-test was carried out to compare the scores of NJSS and SDS in each group. Next, we examined the relation between NJSS and SDS using correlation analysis (Pearson) in each group. The statistical software SPSS Ver.10.5 was used to carry out statistical analysis.

3 RESULTS AND DISCUSSION 3.1 Categorization of continuous exercise group (CE) and non-continuous exercise group (non-CE) Continuous exercise group (CE) and non-continuous exercise group (non-CE) were categorized on the basis of the frequency of exercise; more than once a week or no exercise a week. As the results, 87 nurses were classified into CE group, and 134 nurses were classified into non-CE group. 3.2 Factor analysis of Nursing Job Stressor Scale (NJSS) In order to examine the validity and the reliability of Nursing Job Stressor Scale, factor analysis was carried out. The result was shown to have seven factor structures, which is the same as was found in the precedent study (Higasiguchi and Nakagawa, 2003). All seven factor of Cronbach’s α coefficient were over .70. Therefore, all items were used in the following analysis. 304

Table 1.

Correlation coefficients between the NJSS and the SDS in CE group and non-CE group. SDS

NJSS

Conflict with other nursing staffs Nursing role conflict Conflict with physicians/Autonomy Dealing with death and dying Qualitative work load Quantitative work load Conflict with patients

CE (n = 87)

non-CE (n = 134)

0.19 0.19 0.14 0.15 0.09 0.05 −0.02

0.17∗ 0.24∗∗ 0.24∗∗ 0.15 0.27∗∗ 0.29∗∗ 0.16

∗ p < .05, ∗∗ p < .01

3.3 Comparison of two groups in stressor and depression variables (t-test) In order to compare the scores of the NJSS in CE and non-CE group, t-test was carried out. As the result, there were no significant differences of the scores of job stressors between both groups. In order to compare the scores of the SDS in CE and non-CE group, t-test was carried out. As the consequence, there was no significant difference of the scores of SDS between both groups. These findings suggest that CE and non-CE perceived approximately the same amount of job stressors and CE and non-CE fall into approximately the same level of depression. 3.4 Relation between the NJSS and the SDS in each group A correlation analysis (Pearson) was carried out to examine the relation between NJSS and SDS in CE group and non-CE group (Table 1). There was no significant correlation between job stressors and depression in CE. On the other hand, job stressors related to human relations with nursing stuff members and autonomy, role conflict and work loads have significantly positive correlation with depression in non-CE. These findings suggest that continuous exercise might have a role of a positive stress coping. This agrees with the comments by Mizuno et al. (2006) which pointed out the important role of human relations on nurses’ health condition and the necessity of positive stress coping strategies toward job stressors related to human relations.

4 CONCLUSION AND FUTURE STUDIES This study examined the effect of continuous exercise on job stress among Japanese nurses. The results of this study indicated that there were few differences in the scores of NJSS between CE and non-CE. Moreover, there were not any differences in the score of SDS between CE and non-CE. Job stressors were not related to depression in CE. On the other hand, job stressors related to human relations, nursing role conflicts and work loads were associated with depression in non-CE. This suggests, although there were only small differences in the scores of NJSS and SDS between CE and non-CE, that there seemed to be some relationship between job stressors and depression in non-CE. Therefore, we argue that continuous exercise has a modifying effect of diminishing job stress among Japanese nurses. In this study, although it was shown that continuous exercise might have a role of positive stress coping, effective kinds of exercise and exercise intensity and time are still unclear. Thus, these factors should be considered in the near future. Moreover, exercise itself may become a stressor for 305

someone who is very tired physically and psychologically or who is not good at exercise. Therefore the individual’s health condition and personality should also be considered in order to explain in detail how continuous exercise plays a role as positive stress coping among Japanese nurses. REFERENCES Higasiguchi, K. (Kitaoka), and Nakagawa, H., (2003). Job strain, coping and burnout among Japanese nurses. Japanese Journal of Health & Human Ecology, Vol. 69, 3, pp. 66–79 Kawada, T., Suzuki, S., Kubota, F., Ohnishi, N., and Satoh, K., (1999). Content and cross validity of the Todai Health Index Depression Scale in relation to the Center for Epidemiologic Studies Depression Scale and the Zung Self-rating Depression Scale. Journal of Occupational Health, 41: pp. 154–159 Lazarus, R. S., and Folkman, S., (1984). Stress, appraisal and coping. (New York: Springer Publishing Company) Mizuno, M., Yamada, Y., Mizuno, Y., Matsuda, F., Koizumi, T., and Sakai, K., (2006). A study on work stress and work-family balance of nurses in Japan. Proceedings IEA 2006 Congress, CD-ROM, Elsevier. Zung, W.W.K., (1965). A self-rating depression scale. Achieves of General Psychiatry, 12: pp. 63–70

306

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Relation between Typus Melancholicus and burnout among Japanese nurses Yasuyuki Yamada Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Masataka Hirosawa Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

Miyuki Sugiura & Yasunobu Nishi Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Sumio Tanaka & Motoki Mizuno Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

ABSTRACT: The purpose of this study was to examine the relation between Typus Melancholicus (TM) and burnout in order to clarify whether TM has a role in inducing a personality of burnout or not. This study verifies the hypothesis that “burnout scores of TM nurses are higher than those of non-TM ones”. A questionnaire investigation was carried out with female Japanese nurses in 2007. The total number of valid participants was 701. The questionnaire was composed of the Kasahara’s Typus Melancholicus Scale (KS) and Maslach’s Burnout Inventory (MBI). At first, in this study, a factor analysis of KS was carried out and the result revealed there were two factors of “sthenic in work” and “asthenic in human relation” traits in KS. Then, the results of the 3 (level of sthenic) × 3 (level of asthenic) ANOVA, confirmed the significant main effects of sthenic trait (Low > Middle > High) and asthenic trait (Low < Middle < High) on burnout (p < .001). Hence, the hypothesis of this study was not supported. However, it was shown that the burnout score reflected the asthenic (lack of sthenic) nurses’ burnout more than TM, sthenic (lack of asthenic), and anti-TM nurses’ ones. Therefore, clarifying some different subtypes of burnout according to individuals’personality may provide an effective approach to understanding burnout in detail.

1 INTRODUCTION Based on Maslach and Jackson’s (1984) definition that “a psychological syndrome of emotional exhaustion, depersonalization, and reduced personal accomplishment that can occur among individuals who work with people in some capacity”, three symptoms of “emotional exhaustion”, “depersonalization”, and “reduced personal accomplishment” can be regarded as characterizing burnout syndrome (burnout). However, the internal developmental process of these symptoms still remains unclear, and it still remains difficult to answer the question, “what do these symptoms mean for an individual?” In order to answer this question, clarifying the causality between the individual and the three burnout symptoms, by a case study, is necessary. However, as Perlman and Hartman (1982) pointed out, individual difference variables have received less empirical attention than situational factors in 307

burnout research. One of the reasons is that a collective intervention to the working environment is more practical for the worksite. However, such a trend has left us unable to determine what burnout means to the individual and what kind of help he/she needs from others, because such empirical studies have tended to regard burnout as a phenomenon which can be measured by a certain score reflecting some universal and common phenomenon beyond various occupations and nationalities. Under these circumstances, a personality study of burnout is needed in order to understand burnout in individuals. This study focuses on Tellenbach’s melancholic type of personality (in German, Typus Melancholicus; TM) as one of the personality traits which has been assumed to have a strong relation to burnout and which has symbolized Japanese nurses’ personality. The features of TM include perfectionism, thoroughness, and earnestness in working place. In the score of human relations, the features of TM include devotion to others, being sensitive to evaluation from others, and avoidance of human conflict. These are regarded as expressions of a fundamental feature of TM; “marked adherence to order”. TM was originally noted as a premorbid personality of depression, however, TM features overlap with perfectionism and type-A behaviour, which have been reported as individual factors of burnout (Kobasa, et al., 1983). Hence, this study examines the relation between TM and burnout in order to clarify whether TM has a role of inducing a personality of burnout or not. Specifically, this study verifies the hypothesis that “burnout scores of TM nurses are higher than those of non-TM ones”. 2 METHODS Participants were Japanese female nurses from one university hospital. This research was carried out in 2007. Valid data were obtained from 693 nurses. Their mean age was 27.4 (SD = 6.7). This study was approved by the ethical committee of the university hospital system. Nurses participated in this investigation after informed consent was obtained. The questionnaire was composed of Kasahara’s Typus Melancholicus Scale (KS, Kasahara, 1984), and Maslach’s burnout inventory (Maslach and Jackson, 1981).

3 RESULTS 3.1 Factor analysis of Kasahara’s Typus Melancholicus Scale At first we examined the factor structure of KS among the subjects. A factor analysis was conducted using promax rotation (principal factor method) for all items (Table 3.1). As results, two factors exhibited eigen values of greater than 1.0 in the initial extraction of factors. Two items (Q12 and Q13) were excluded from the two factor structure of KS because these items showed low factor loadings (<.30). Moreover, internal consistencies of both factors were high (α > .73). Accordingly, in this study, we judged that two factor structures were best for KS. The two factors were named as “sthenic in work (sthenic)” trait and “asthenic in human relation (asthenic)” trait by the meaning of both factors’ items. 3.2 Comparison of the scores of MBI between the different level groups of TM We divided the subjects into nine groups based on the two factors of TM; 3 (levels of sthenic) × 3 (levels of asthenic). Then, in order to compare the burnout score in each group, the 3 (levels of sthenic) × 3 (levels of asthenic) ANOVA was carried out to MBI score (Table 3.2). As for the results, regarding MBI score, significantly main effects of sthenic trait (Low > Middle > High) and asthenic trait (Low < Middle < High) were confirmed (p < .001). Namely, although the main effect of asthenic trait supported the hypothesis of this study, that of sthenic trait didn’t support it. Moreover, it was proved that sthenic (lack of asthenic) nurses showed the lowest score of burnout, and asthenic (lack of sthenic) nurses showed the highest score. 308

Table 1.

Factor analysis of Kasahara’s Typus Melancholicus Scale. Factor Loadings

TM factors

Items

Sthenic (α = .73)

2 When I start something, I always finish it thoroughly 3 I have a strong sense of responsibility. 15 I am neat. 14 I like to arrange my belongings. 4 I give importance to my social duty. 1 I like to work.

0.67 0.63 0.60 0.57 0.47 0.43

−0.08 −0.02 −0.08 −0.06 0.13 −0.09

13 I am rather cheerful. 12 I sometimes get excited easily.

0.27 0.18

0.03 0.02

−0.21 −0.06 −0.02 −0.13 0.13 0.26 0.38

0.67 0.59 0.57 0.52 0.52 0.46 0.41

3.22 16.96

2.52 12.42

Asthenic (α = .74)

F1

7 I am rather timid 6 I would rather avoid confrontation with somebody 8 I am nervous about what other people think of me 11 I do not like to be conspicuous. 10 I would not do something extreme. 5 I cannot say no, when someone asks me to do something. 9 I give importance to common sense. Eigenvalue (First Solution) Precent of Variance

Table 2.

F2

Comparison of total score of MBI between different level groups of TM using a two-way ANOVA. Sthenic Levels Low

Middle

High

Asthenic Levels

Mean

SD

Mean

SD

Mean

Low Middle High Total sum

50.9 50.6 62.7 54.6

26.5 24.3 22.1 24.8

39.7 45.2 52.0 46.3

25.2 24.2 22.4 24.2

43.8 41.5 46.6 44.0

Total sum

Two-way ANOVA (F-value)

SD

Mean

SD

Sthenic

Asthenic

INT

22.7 24.7 25.0 24.4

44.4 45.7 53.2 48.1

25.5 24.5 23.8 24.8

10.4*

8.4*

1.0

*p < .001

4 DISCUSSION As the results of this study show, those nurses who had strong asthenic traits gained higher scores of MBI, however, those nurses who had strong stenic traits did not gain higher scores of MBI. Therefore, the hypothesis of this study, “burnout scores of TM nurses are higher than those of non-TM ones”, was rejected because only the nurses who have both personality traits are regarded as TM one. In this study, the nurses with higher levels of sthenic trait showed lower burnout scores. This result is contrary to conventional personality studies of burnout which dealt with type-A behaviour pattern and perfectionism. However, when we take into consideration the high stress situation of the nursing job, this result may be able to be understood. Namely, when we disregard the responsibility and hardiness which are demanded by the nursing profession, sthenic nurses may show higher burnout scores because they always place demands on themselves to perform a lot of work by themselves. Moreover, anti-sthenic nurses showed lower burnout scores because they don’t place demands on themselves to complete their work perfectly. In this case, this hypothesis 309

may be supported. However, most of the nurses are always obligated to perform a huge quantitative and qualitative work load in Japan regardless of having sthenic trait or not. In this case, sthenic nurses may be able to adapt to the nursing job more easily than asthenic nurses because the sthenic personality trait matches the environmental obligation. Moreover, in this study, nurses with higher levels of asthenic trait showed higher burnout scores. Asthenic nurses do things for others in order to keep expected interpersonal orderliness. Therefore, they tend to be sensitive to evaluation from others, show concern for others and avoid human conflict and they often find it difficult to say “no.” According to these features, the behaviour of asthenic nurses, just like that of type-A, is often motivated by external demand (Kobasa et al., 1983). When we consider the type-A behaviour pattern theory, asthenic nurse’s excessive work, which is brought about by external motivation, can be seen as a cause of burnout. Furthermore, asthenic nurses are more sensitive to job stressors than anti-asthenic ones (Yamada et al., 2007). Therefore, it is said that asthenic nurses tend to experience burnout more than anti-asthenic nurses. Considering the combination of both TM factors, sthenic (lack of asthenic) nurses showed the lowest scores of burnout, and asthenic (lack of sthenic) nurses showed the highest scores. Although these results were remarkable, such personality variables of nurses have not been dealt with in conventional studies. Thus we believe that clarifying the burnout mechanisms in each of the personality types among nurses (TM, sthenic (lack of asthenic), asthenic (lack of sthenic), and anti-TM) may provide an effective approach to understanding burnout in detail.

5 CONCLUSIONS In this study, the hypothesis that “burnout scores of TM nurses are higher than those of non-TM ones” was not supported. However, it was shown that the burnout score reflected the asthenic (lack of sthenic) nurses’ burnout more than TM, sthenic (lack of asthenic), and anti-TM nurses’ ones. Therefore, clarifying some different subtypes of burnout according to individuals’ personality may provide an effective approach to understanding burnout in detail. REFERENCES Kobasa, S. C., Maddi, S. R., and Zola, M. A., (1983). Type A and Hardiness. Journal of Behavioral Medicine, Vol. 6, No. 1: pp. 41–51 Kasahara, Y., (1984). Depression in general practice. Japan J. Psychosom Med, Vol. 24: pp. 6–14 (in Japanese) Maslach, C., and Jackson, S. E., (1981). The measurement of experienced burnout. Journal of Occupational Behavior, 2: pp. 99–113 Maslach, C., and Jackson, S. E., (1984). Burnout in Organizational Settings. In: S. Oskamp (Ed.), Appried social psychology annual: Applications in organizational settings (Vol. 5, pp.133–135). Beverly Hills, CA: Sage. Perlman, B., and Hartman, E. A., (1982). Burnout: summary and future research. Human Relations, 35: pp. 283–305 Tellenbach, H., (1961). Melancholie. Springer, Berlin. (In Germany) Yamada, Y., Sugiura, M., Nishi, Y., Hirosawa, M., Tanaka, S., and Mizuno, M., (2007). Effect of Typus Melancholicus on Cognition of Job Stressor among Japanese Nurses. The 18th Japan- China-Korea Joint Conference on Occupational Health, Program and abstracts: pp. 131–132

310

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Relationship between depression and depersonalization among Japanese nurses Sumio Tanaka Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

Miyuki Sugiura Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Masataka Hirosawa Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

Yasuyuki Yamada & Yasunobu Nishi Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Motoki Mizuno Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

ABSTRACT: The purpose of this study is to examine the relationship of depression and depersonalization among Japanese nurses. The survey was conducted to female Japanese nurses in 2007. The total number of valid respondents was 1058. The questionnaire consisted of the Cambridge Depersonalization Scale (CDS) and Zung Self-Rating Depression Scale (SDS). We translated the English version of CDS into Japanese. The subjects were divided into three groups depending upon the depth of depression (nondepression, slight depression, moderate depression). We compared scores of depersonalization among the above-mentioned three groups. The result showed that the scores of total CDS and all subscales of the moderate depression group were significantly higher than those of the nondepression and the slight depression groups. These results suggested that the two symptoms overlapped each other. Therefore, if we find one of two symptoms in nurses, we should suspect they may also have the other one. We must support them in order to prevent their burnout syndrome and malpractices.

1 INTRODUCTION It has been well known that depersonalization syndrome accompanies most mental diseases, emotional stress in somatic diseases and exhaustion (Nuller, 1982), also in mentally healthy people. Depression is a very common disease and also one of the symptoms which are related to depersonalization. However, there are only a few empirical studies which have investigated the relationship between depression and depersonalization. Zung Self-Rating Depression Scale (SDS) has been certified its validity but another Cambridge Depersonalization Scale (CDS) is not popular in Japan. Therefore, to standardize the CDS scale will be one of the purposes of this study. The main purpose of this study was to clarify the relationship between depression and depersonalization. Japanese female nurses completed questionnaires composed by the above-mentioned 311

two scales. Following this study, we will discuss any other aspects related to depersonalization and other disorders among Japanese nurses.

2 METHODS 2.1 Participants The questionnaire survey was conducted to 1,058 female Japanese nurses working at two university hospitals in the metropolitan area of Japan. Their average age was 27.6 (SD = 6.23) years old. Most of the participants were not assigned to a managerial position and were not married. This survey was conducted from January to April in 2007. 2.2 Measure 2.2.1 The Cambridge Depersonalization Scale (CDS) The CDS is a comprehensive instrument which consists of 29 items addressing the complaints classically associated with depersonalization syndrome. Each item is rated on two Likert scales for frequency and duration of experience. The question is “How often have you had these experiences over the last six months and how long is their approximate duration?” The sum of these two scores generates an index of item intensity (the range of frequency; 0–4, the range of duration; 1–6, the range of one item; 0–10). The global score of the scale is the arithmetic sum of all items (range; 0–290). This scale has high internal consistency and validity (Cronbach’s co-efficient α and spirit half reliability of 0.89 and 0.92). In a clinical scene, the cut-off point of depersonalization is 70. It has a sensitivity of 75.5% and specificity of 87.2% (Sierra and Berrios, 2000). In addition, the CDS has four subscales ((1) Anomalous Body Experiences, (2) Emotional Numbing, (3) Anomalous Subjective Recall, (4)Alienation from Surroundings). 2.2.2 Self-Rating Depression Scale (SDS) The Self-Rating Depression Scale (SDS; Zung, 1965) was designed to assess an individual’s depression. This scale is composed of 20 items including the major mental and partially physical symptoms of depression. The validity and reliability of the SDS for Japanese have been reported (e.g. Kawada et al., 1999). The score of the SDS is the sum of 20 items. Each response was reported on a 4-point Likert-scale ranging from 1 (strongly disagree) to 4 (strongly agree). In general, the total scores of the sum of 20 items in SDS have four states. The depth of depression is divided into next four grades: the range of the non-depression is 20–39, that of the slight depression is 40–49, that of the moderate depression is 50–59, that of the severe depression is 60–80. 2.3 Data analysis Based on a previous study (Sugiura, 2007), the participants were divided into four groups (nondepression, slight depression, moderate depression, and severe depression). Finally we compared the scores of total CDS and four subscales among four groups by Kruskal-Wallis test and Scheffe’s test. The statistical software SPSS Ver.14.0 was used to carry out statistic analysis.

3 RESULTS AND DISCUSSIONS 3.1 Distribution of depression scale (SDS) These results showed that the number of participants in the non-depression group was 557, that of the slight depression group was 461, that of the moderate depression group was 37, and that of the severe depression group was only three (Table 1). Over half of the participants included in 312

Table 1.

Distribution of depression scale (SDS). N

Non-depression Slight depression Moderate depression Severe depression

557 461 37 3

Total participants

Table 2.

%

1058

52.6 43.6 3.5 0.3 100

Comparison of CDS scores between three groups (SDS).

CDS Anomalous Body Experience Emotional Nurmbing Anomalous Subjective Recall Alienation from Surroundings

Non-depression N = 557

Slight depression N = 461

Moderate depression N = 40

M

SD

M

SD

M

SD

χ2 p

7.83 1.70 1.59 1.94 1.92

16.94 4.88 4.34 3.87 3.77

9.92 2.29 1.95 2.39 2.55

18.47 5.72 4.26 4.28 4.80

29.25 8.03 6.23 6.25 5.30

48.35 14.42 10.13 10.08 8.20

13.05∗∗ 19.77∗∗∗ 18.88∗∗∗ 14.34∗∗∗ 12.81∗∗

∗∗ p < .01 ∗∗∗ p < .001

the non-depression group. 43.6% of the participants were included in the slight depression group. If they receive social supports from their surroundings, they might have no problems because the slight depression is a very common condition which is easy to recover by adequate supports. We must pay attention to participants included in the moderate and severe depression groups. In this study, the participants included in the severe depression group were very few (only 0.3%), therefore we put the participants in the severe depression group into the moderate depression group in the following analysis. 3.2 Comparison of the CDS scores among three groups of depression states. In order to examine significant differences in scores of the CDS among three groups depending upon the depth of depression (SDS), the Kruskal-Wallis test confirmed significant differences among three groups in scores of total CDS and four subscales. Additionally, Scheffe’s test showed that the scores of total CDS and all subscales of the moderate depression group were significantly higher than those of the non-depression and the slight depression groups (Table 2). These results suggested that two symptoms of depression and depersonarization overlapped each other. Therefore, when we observe one of two symptoms in nurses, we should suspect they may also have the other one. Of course, almost of all the nurses do not suffer from depersonalization syndrome. However, even though the total scores of a person are under 70 (i.e., the cut-off point), if the score of “alienation from surroundings” is relatively high, we must take care of that type of persons. This is because we think “alienation from surroundings” is most correlated to strong stress and medical malpractice.

4 CONCLUSIONS The main purpose of this study is to clarify the relationship between depression and depersonalization among Japanese nurses. The results of this study indicated that the scores of total CDS and four 313

subscales of moderate depression group were significantly higher than those of the non-depression group and the slight depression group. This result suggested that the two symptoms overlapped each other. If we find one of two symptoms in nurses, we should suspect they may also have the other one. In this study, almost of all the participants do not suffer from depersonalization disorder. However, even if total scores were under the cut-off point of depersonalization, we must take care of the person attentively. It is because in the view of clinical psychology and psychiatry, the dissociative reactions have been increasing in present day. In DSM-IV-TR, depersonalization disorder is a kind of dissociation disorder. Those who have dissociation disorder often lose the real sense; they have an integrated self, which will make them easy to experience medical mistakes and malpractices. We must research the relationship between the dissociation disorder and the CDS. Of course, the depersonalization itself may become clearer. In any case, we think depersonalization syndrome may be one of the predicting factors of medical malpractices. Following this study, we will discuss any other aspects related to depersonalization and other disorders among Japanese nurses and manage working condition of nurses in order to prevent their burnout and malpractices. The final purpose is of course to understand the conditions that nurses can maintain a good mental condition and can prevent from medical malpractice. Furthermore, we intend to present a viewpoint of the development of working systems for the effective stress management on the job of nurses.

ACKNOWLEDGEMENTS We are most grateful to the nurses working at two university hospitals for their support and the contribution to the data collection of this study. We also appreciate helpful comments, suggestions and constant supports from our colleagues at Juntendo University on this study. REFERENCES Kawada, T., and Suzuki, S., (1992). Factor structure of self rating depression scale by Zung and prevalence of depressive state of night shift workers. Sangyo-Igaku, 34: pp. 131–136 Kawada, T., Suzuki, S., Kubota, F., Ohnishi, N., and Satoh, K., (1999). Content and cross validity of the Todai Health Index Depression Scale in relation to the Center for Epidemiologic Studies Depression Scale and the Zung Self-rating Depression Scale. J. Occup. Health, 41: pp. 154–159 Nuller, Y.L., (1982). Depersonalization-symptoms, meaning, therapy. Acta Psychiatrica Scandinavica, 66 (6): pp. 451–458 Sierra, M., and Berrios, G.E. The Cambridge Depersonalization Scale: a new instrument for the measurement of depersonalization. Psychiatry Research 93 2000: pp. 153–164 Sugiura, M. et al., (2007). Relationship between Depersonalization and Obsessive-Compulsive Disorder in Japanese Nurses. The 8th Pan-Pacific Conference on Occupational Ergonomics. Zung, W.W.K., (1965). A self-rating depression scale. Arch. Gen. Psychiatry, 12: pp. 63–70

314

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Effect of obsessive personality traits and impulsiveness on obsessive-compulsive disorder and eating disorders among hospital nurses Yasunobu Nishi Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Masataka Hirosawa Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

Miyuki Sugiura & Yasuyuki Yamada Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Sumio Tanaka & Motoki Mizuno Juntendo University Graduate School of Health and Sports Science, Chiba, Japan Juntendo University School of Health and Sports Science, Chiba, Japan

ABSTRACT: In this study, we examine (1) the intensity of impulsiveness in nurses with highly and mildly obsessive personality traits and (2) how an obsessive personality and impulsiveness are involved in obsessive-compulsive disorder and eating disorders, which are frequently found among Japanese nurses. This study demonstrated that in hospital nurses, those with a higher number of obsessive personality traits are more impulsive. Moreover, it was clarified that there is a greater tendency toward obsessive-compulsive disorder among those with a higher number of obsessive personality traits and that eating disorders tend to worsen with increasing impulsiveness.

1 INTRODUCTION Obsessive-compulsive disorder and eating disorders can be frequently found among hospital nurses. These disorders often have negative results not only with respect their mental and physical conditions but also with respect to their performances of various tasks. While this simple statement might not be disputed, in order to develop a more appropriate health care system in hospital nurses, we should reconsider several problems. First, some studies have pointed out that the common feature between these two disorders is that they are both somehow related to having an obsessive personality. Moreover, this personality is often viewed as a preferable characteristic for those working in the field of medical care. Second, recently, another empirical study has indicated that these two disorders have little relation with an obsessive personality; rather, they are more closely associated with impulsiveness. This can be supported by clinical findings such as the fact that impulsiveness can usually be found in compulsive acts and that obsessive-compulsive disorder and eating disorders are often associated with impulse control disorders. Impulsiveness may obstruct medical care, which requires a calm approach. Although there are many studies that dealt with the comorbidity of obsessive-compulsive disorder and eating disorders, few empirical studies have dealt with the relationship between these disorders and personality traits such as an obsessive personality and impulsiveness. Under these 315

conditions, this paper reveals (1) the relation between the two traits, namely, an obsessive personality and impulsiveness, and (2) the importance of the abovementioned disorders and the personality traits that cause them in terms of nurses’ mental and physical health. In this study, we examine (1) the intensity of impulsiveness in nurses with highly and mildly obsessive personality traits and (2) how an obsessive personality and impulsiveness are involved in obsessive-compulsive disorder and eating disorders among hospital nurses.

2 METHODS Questionnaires were administered to 716 nurses working at University Hospital in Japan. This research was carried out in 2007. The valid respondent rate was 89% (639). The participants’ mean age was 27.1 (SD = 5.9). This study was approved by the ethical committee of the University Hospital system. Informed consent was obtained from the nurses who participated in this investigation. The questionnaire package comprised the Japanese versions of the Maudsley Obsessional Compulsive Inventory (MOCI), the Eating Attitude Test-26 (EAT-26), the Barrat Impulsiveness Scale-11 (BIS-11), and the Obsessive Personality Trait Scale (OPTS).

3 RESULTS 3.1 Prevalence of obsessive-compulsive disorder and eating disorders In order to investigate the prevalence of obsessive-compulsive disorder and eating disorders among all the participants, we divided them into clinical and non-clinical groups on the basis of their MOCI and EAT-26 scores. We adopted 13 points and 20 points as the cut-off points for the MOCI and EAT-26, respectively. The results were as follows: the percentage of participants with obsessivecompulsive disorder was 13.0% (87) and the percentage with eating disorders was 3.1% (20). The number of participants with both symptoms was 8 (1.2%).

3.2 The intensity of impulsiveness in three obsessive personality trait groups In order to examine the effect of obsessive personality traits on impulsiveness, we divided the participants into three groups (high, middle, low) on the basis of their OPTS scores (Table1). These three groups of obsessive personality traits were categorized according to three levels: below 33%, between 33% and 67%, and above 67% of the whole sample. A one-way ANOVA confirmed that significant differences exist among the groups based on the total BIS-11 scores (F = 14.98, p < .001). Multiple comparison tests demonstrated that the total BIS-11 score was significantly higher in the high group than in the middle and low groups (p < .001). Moreover, the middle group had a higher total BIS-11 score than did the low group (p < .01).

Table 1.

Scores of BIS-11 in three obsessive personality trait groups

Group

n

M

SD

High Middle Low

183 228 228

61.19 58.12 56.71

8.56 8.14 8.37

316

3.3 Verifying the relation between personality traits and disorders using a two-way ANOVA In order to examine the effects of obsessive personality traits and impulsiveness on the MOCI and EAT-26 scores, a 3 × 3 (level of obsessive personality trait × level of impulsiveness) ANOVA was conducted. The MOCI scores (F = 12.43, p < .001) confirmed the positive main effects of obsessive personality traits and the EAT-26 scores (F = 16.95, p < .001) confirmed the positive main effects of impulsiveness. 4 DISCUSSION 4.1 Prevalence of obsessive-compulsive disorder and eating disorders In this study, the percentage of nurses with eating disorders was 3.1%; this result revealed a somewhat lower prevalence than was demonstrated in the previous studies that used the EAT-26. However, as Sugiura et al. (2007) remarks, the existence of eating disorders among hospital nurses is a serious problem, even if only a small number of nurses suffer from such disorders. Further, in this study, the percentage of nurses with obsessive-compulsive disorder was 13.6%; this result is approximately consistent with that of a previous study, which reported that the percentage of nurses with obsessive-compulsive disorder is 16.7% (Harada et al., 2007). On the other hand, there is an indication that with regard to the general public, the percentage of people with obsessive-compulsive disorder is 2–3% (DSM-IV). Undoubtedly, there are methodological or national differences between the Japanese studies and the DSM-IV. However, these results suggest that Japanese hospital nurses have a significantly higher risk of suffering from obsessive-compulsive disorder. 4.2 Obsessive personality traits and impulsiveness In this paper, the results of the analysis indicate that among hospital nurses, those with a higher number of obsessive personality traits are also more impulsive. In light of Hollander’s theory of obsessive-compulsive spectrum disorders (1995), which states that impulsiveness is the opposite of compulsiveness, this result requires further explanation. Indeed, certain features of an obsessive personality, such as perfectionism or carefulness, may not be concomitant with those of impulsiveness, such as a lack of attention or arbitrary behavior. In the case of hospital nurses, however, we should take into consideration the fact that they are hard-working people. Those nurses who have obsessive personalities and who attempt to complete their work perfectly may become too fatigued to control their impulsiveness while at work, which can elevate their BIS-11 scores. 4.3 Obsessive personality traits and obsessive-compulsive disorder With regard to obsessive-compulsive disorder, those with a higher number of obsessive personality traits were revealed to have a tendency to suffer from obsessive-compulsive disorder. Although several recent studies have rejected the relation between obsessive-compulsive disorder and obsessive personalities, our result demonstrated that an obsessive personality has a certain clinical meaning with respect to hospital nurses. Furthermore, considering the relation between obsessive personality traits and impulsiveness, if hospital nurses suffer from obsessive-compulsive disorder, we should pay attention to their hidden uncontrollable impulsiveness. 4.4 Impulsiveness and eating disorders With respect to eating disorders, those who are more impulsive were revealed to have a tendency to suffer from such disorders. Many clinicians consider these disorders in relation with obsessive personality traits. Moreover, Hollander (1995), in his theory regarding obsessive-compulsive spectrum disorders, considered eating disorders as belonging to the obsessive-compulsive group. However, our result suggests that eating disorders among hospital nurses are not necessarily related with whether or not they have an obsessive personality. If hospital nurses suffer from eating disorders, we should pay attention to their uncontrollable impulsiveness directly. 317

5 CONCLUSION This study demonstrated that among hospital nurses, those with a higher number of obsessive personality traits are more impulsive. Moreover, it clarified that there is a greater tendency to suffer from obsessive-compulsive disorder among those with a higher number of obsessive personality traits and that eating disorders tend to worsen with increasing impulsiveness. Therefore, when hospital nurses suffer from obsessive-compulsive disorder or eating disorders, it is important to pay attention to their uncontrollable impulsiveness, in order to sustain their mental and physical health. REFERENCES Harada, T., Nakamura, A., Tomotake, M., and Ohmori, T., (2007). The influence of obsessive-compulsive symptoms on quality of life among the female nursing staffs. Japanese Journal of Psychosomatic Medicine, 47(1): pp. 33–40 Hollander, E., and Wong, C., (1995). Obsessive-compulsive spectrum disorders. Journal of Clinical Psychiatry, 56: pp. 3–6 Matsumura, K., (1996). Alexithymia and obsessive-compulsive personality; a measurement based approach. Seishin Igaku, 38(10): pp. 1055–1063 Nakai, Y., (2003). Validity of Eating Attitude Test. Seishin Igaku, 45 (2): pp. 161–165 Someya, T., Sakado, K., Seki, T., Kojima, M., Reist, C., Tang, S., and Takahashi, S., (2001). The Japanese version of the Barratt Impulsiveness Scale, 11th version (BIS-11): Its reliability and validity. Psychiatry and Clinical Neurosciences, 55: pp. 111–114 Sugiura, M., Nishi, Y., Yamada, Y., Hirosawa, M., Tanaka, S., and Mizuno, M., (2007). Prevalence of eating disorder among Japanese nurses. The 18th Japan-China-Korea joint conference on occupational health, pp. 209 Yoshida, M., Kiriike, N., Nagata, T., and Matsunaga, H., (1995). Clinical usefulness of Japanese version of Maudsley Obsessional Compulsive Inventory in obsessive-compulsive disorders. Seishin Igaku, 37 (3): pp. 291–296

318

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Work-family balance and stressors among Japanese administrative nurses Motoki Mizuno Juntendo University School of Health and Sports Science, Chiba, Japan Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Yasuyuki Yamada Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Masataka Hirosawa Juntendo University School of Health and Sports Science, Chiba, Japan Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Miyuki Sugiura, Yasunobu Nishi & Yujiro Kawata Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

Sumio Tanaka Juntendo University School of Health and Sports Science, Chiba, Japan Juntendo University Graduate School of Health and Sports Science, Chiba, Japan

ABSTRACT: This study focuses on the work-family conflict (WFC) concept as one of the indicators of work-family balance and clarifies the job stressor types which can lead to WFC. Furthermore, this study also identifies the administrative class nurses’ (AC) stress conditions through a comparison with non-administrative class nurses (non-AC). For this purpose, we conducted a questionnaire investigation concerning work-family balance and stressors of female nurses in Japan in 2007. The total number of valid respondents was 206. The questionnaire was composed of Work-Family Conflict Scale (WFCS), Nursing Job Stressor Scale (NJSS), and the attributes of individuals. From the results it was clarified that some job stressors were significantly correlated to WFC. Furthermore, many scores of WFCS and NJSS in AC were higher than those of non-AC. Therefore, it was concluded that more attention needs to be paid to nursing staff members, and in particular to nurses in the administrative class, who play important roles in university hospitals, in order to manage their stress conditions and improve their work-family balance.

1 INTRODUCTION Recently, in Japan, to help Japanese nurses balance work with family life, it has become an urgent matter to consider maternity and parental leave, family-care leave, and reduced working hours for childcare and family care. However, little consideration has been paid either to working conditions or to the work-family balance as parts of the essential human resource strategies involved in the medical field. In particular, it is very difficult for administrative class nurses to balance work and family responsibilities nowadays. Therefore, this study focuses on the work-family conflict (WFC) concept as one of the indicators of work-family balance and clarifies the job stressor types which can cause bad WFC. Furthermore, this study also identifies administrative class nurses’ stress conditions though a comparison with non- administrative class nurses. 319

Table 1.

Comparison on the features of attribution between AC and non-AC using t-test. non-AC

AC

Factors

Mean

SD

Mean

SD

t-value

Age Service Years Total Number of Nightshifts in a Month Total Number of Holidays in a Year Total Members of Familiy

29.49 6.04 4.48 7.19 3.16

6.27 5.07 5.15 2.92 1.19

36.24 13.69 3.06 6.59 3.22

8.20 7.27 3.12 2.77 1.15

−5.03 *** −7.34 *** 2.30 * 1.32 −0.29

*p < .05, **p < .01, ***p < .001

2 METHODS This study, based on a questionnaire investigation concerning work-family balance and stressors of female nurses in Japan, was carried out in 2007. Data was gathered from two university hospitals in Japan. The total number of valid respondents was 206. The questionnaire was composed of Kanai and Wakabayashi’s (1998) Work-Family Conflict Scale (WFCS), Kitaoka-Higashiguchi and Nakagawa’s (2003) Nursing Job Stressor Scale (NJSS) and the attributes of individuals. WFCS has a four factor structure, including “work to family conflict”, “family to work conflict”, “time conflict”, and “support for solving conflict”. NJSS has a seven factor structure, including “conflict with other nursing staffs”, “nursing role conflict”, “conflict with physicians/autonomy”, “dealing with death and dying”, “qualitative work load”, “quantitative work load”, and “conflict with patients”. For the analysis, first we categorized the administrative class group (AC) and non-administrative class group (non-AC). Then some attributions were compared among both groups using a t-test in order to clarify the attribution features of both groups before the comparison of WFC and job stressors. Secondly, the coefficient of correlation was drawn from WFCS and NJSS in order to examine the relation between work-family conflict and job stressors. Finally, in order to clarify the work-family conflict and job stressor conditions among both groups, we compared the scores of WFCS and NJSS between AC and non-AC. 3 RESULTS 3.1 Features of attribution among AC and non-AC nurses In this study, the subjects were divided into AC group (n = 60) and non-AC group (n = 146). Before the comparison of work-family conflict and job stressor conditions, we compared some attributions of AC and non-AC using a t-test. This was because it is necessary to clarify whether or not the features of attribution which had a possibility to influence both concepts were meaningful. The results were shown in Table 1. In this analysis, AC showed significantly higher scores than non-AC on the items regarding age, service years, and total number of nightshifts in a month. Moreover, no significant differences were confirmed regarding the total number of holidays in a year and the total number of family members. Therefore, we should evaluate the results of the following analysis after taking these differences of attribution into consideration. 3.2 Correlation between WFCS and NJSS In order to clarify the job stressors which can cause bad WFC conditions, we examined the relation between WFCS and NJSS using calculating Pearson’s correlation coefficient. The results of this analysis are shown in Table 2. As the results, “conflict with other nursing staff members”, “nursing 320

Table 2.

Correlation between WFCS and NJSS NJSS Factors

WFCS Factors Work to Family Conflict Family to Work Conflict Time Conflict Support for Solving Conflict Total Score of WFCS † p < .1,

Table 3.

Conflict with Nursing Conflict with Dealing with Conflict Other Nursing Role Physicians/ Death and Qualitative Quantitative with Staff Members Conflict Autonomy Dying Work Load Work Load Patients 0.23 **

0.23 *** 0.26 ***

0.10

0.23 ***

0.04

0.05

0.10

0.19 **

0.16 *

0.14 *

0.11

0.00

0.00

0.19 ** 0.01

0.22 ** 0.02

0.22 ** 0.03

0.11 0.06

0.20 ** 0.08

0.11 −0.01

−0.03 −0.03

0.20 **

0.24 *** 0.25 ***

0.13†

0.22 **

0.05

−0.01

*p < .05, **p < .01, ***p < .001

Comparison of WFBS and NJSS scores between AC and non-AC non-AC

AC

Scales

Factors

Mean

SD

Mean

SD

t-value

WFCS

Work to Family Conflict Family to Work Conflict Time Conflict Support for Solving Conflict Total Score of WFCS

2.09 1.59 2.54 2.25 2.06

1.05 0.70 1.26 1.00 0.77

2.47 1.59 2.78 2.41 2.26

1.03 0.70 1.19 0.82 0.73

−2.38* −0.06 −1.28 −1.17 −1.78†

NJSS

Conflict with Other Nursing Staff Members Nursing Role Conflict Conflict with Physicians/Autonomy Dealing with Death and Dying Qualitative Work Load Quantitative Work Load Conflict with Patients

2.59

0.90

2.89

0.79

−2.36 *

2.52 2.53 2.18 2.74 3.12 2.80

0.72 0.92 1.02 0.72 0.65 1.03

2.93 2.91 2.44 2.96 3.19 2.92

0.74 0.87 0.95 0.58 0.58 0.83

−3.57 *** −2.78 ** −1.76† −2.25 * −0.67 −0.81

† p < .1,

*p < .05, **p < .01, ***p < .001

role conflict”, “conflict with physicians/autonomy”, “dealing with death and dying”, and “qualitative work load” showed significantly positive correlations to the total score of WFCS. Therefore, it is suggested that intervention regarding job stressors provides an effective method to keep a good work-family balance among Japanese nurses.

3.3 Comparison of WFCS and NJSS scores between AC and non-AC In order to identify the features of job stress and WFC conditions, the t-test was carried out. The results of this analysis are shown in Table 3. AC’s scores of “work to family conflict”, “total score of WFCS”, “conflict with other nursing staffs”, “nursing role conflict”, “conflict with physicians/autonomy”, “dealing with death and dying”, and “qualitative work load” were higher than 321

those of non-AC. Therefore, AC nurses were more stressed and found it more difficult to keep a good work-family balance than non-AC nurses in study.

4 DISCUSSION The findings suggest that the nurses in the administrative class should be adequately managed in consideration of the factors mentioned in this study in order to keep a good work-family balance. That is to say, intervention strategies toward job stressors are very effective methods among Japanese nurses, and especially among nurses of the administrative class. Moreover, this study implies that intervention, as one of the organizational designs used to reduce job stress and enhance the health conditions, can improve the work-family balance. A great concern has been raised about these matters in business organizations and particularly among factory workers. Similarly, reduction of the job stress also appears to be helpful for work motivation of nurses in the administrative class in Japan. Consequently, it is expected that the Japanese medical society may improve if such measures are promoted.

5 CONCLUSIONS In this study it was clarified that the nurse respondents from university hospitals had difficulty in balancing work and family due to various stressors. In general, work-family balance has a significantly positive influence on health conditions and a significantly negative influence on work stress (e.g. Mizuno et al., 2007). In the near future, constructive conflict management as work-family balance will require a sophisticated assessment of the conflict situation, including the perceptions of conflict participants. Conflict management literature defines collaboration as the attempt to work with the other party to find a solution that fully satisfies the concerns of all stakeholders (Sportsman and Hamilton, 2007). Collaboration between nurses, in the context of work-family balance, requires a commitment of time and interpersonal energy to be effective. Therefore, institutional and organizational support systems in medical scenes are necessary to promote competitive advantages in a severe environment. Hence, more attention needs to be paid to the stressors and work-family balance of nursing staff members, and in particular to that of nurses in the administrative class, who play important roles in university hospitals. REFERENCES Kanai, A., and Wakabayashi, M., (1998). Work-family conflict part-time female workers (In Japanese). Association of industrial/Organizational Psychology Journal, Vol.11, No.2: pp. 107–22 Kitaoka-Higashiguchi, K., and Nakagawa, H., (2003). Job strain, coping, and burnout among apanese nurses. Journal of Health and Human Ecology, Vol.69, No.3: pp. 66–79 Mizuno, M. et al. (2003). An empirical study on human resource development in nursing organization: from the view point of motivators to medical staffs. Proceedings of the XVth Triennial Congress of the International Ergonomics Association, Vol.6, pp. 645–648 Mizuno, M. et al. (2003). An analysis of motivators and learning reinforcers of medical staff: A human resource development approach. Asian Journal of Ergonomics, Pan-Pacific Council on Occupational Ergonomics, Vol.4, No.2: pp. 121–135 Mizuno, M., Yamada, Y., Mizuno, Y., Matsuda, F., Koizumi, T., and Sakai, K., (2007). An empirical study on work stress and health condition of Japanese nurses. Journal of Health and Sports, Science Juntendo University, No.10: pp. 58–63 Sportsman, S., and Hamilton, P., (2007). Conflict management styles in the health professions. Journal of Professional Nursing, Vol.23, No.3 pp.157–166

322

Chapter 7 Others

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Time constraints and age: Health impact on musculoskeletal problems and perceived health Barbini Norma Epidemiological Center, INRCA, Ancona, Italy

Squadroni Rosa University of Marche, Ancona, Italy

Sera Francesco CSPO, Firenze, Italy

ABSTRACT: Aims – To analyse the exposure to high working rhythms in women and men and to highlight how this working constraint is tolerated with ageing. To investigate the relationships between high working rhythms and some health-illnesses, such as Musculoskeletal Disorders (MSD) and Health Self-Assessment. Methods – A population of 1,195 Italian workers (297 women and 898 men) was investigated. The relationships between the presence of MSD, poor health self-assessment and working under time pressure were then explored by univariate and multivariate analysis. Results – female workers were more exposed to repetitive work with tight deadlines and to time pressure. Analyzing the occupational exposure by cohorts, a decreasing exposure frequency may be observed for men in the oldest cohorts, while the opposite was observed for women, who complained about these constraints as particularly difficult with ageing. Working under time pressure appeared to be the least tolerated constraint for women, who had a significantly higher Odds Ratio than men in all cohorts of age, with a greatest risk in the 52 years cohort. The exposure to high work rhythms was associated with worst health conditions, both for MSD and health self-assessment, but in different ways for the two sexes. In women the interaction between repetitive work with high deadlines and MSD, showed a statistically significantly association both for upper limbs (p < 0.005) and for multiple MSD (p < 0.05). The multivariate analysis showed an increasing risk with age for women (O.R. 1.79), while in men repetitive work with tight deadlines was associated with a poorly perceived health (p < 0.001). When analyzing interactions between repetitive work with tight deadlines and poor health-assessment, a progressive increased risk was observed from the 42 to 52 year cohorts for men, and in the 47 years cohort for women. Conclusions – the higher exposure of women to time constraints was confirmed, as well as the possibility for oldest men in avoiding the more demanding, or less tolerated, work. The negative health impact due to high working rhythms reflected a gender differences in the difficulty of coping with time constraints with age.

1 INTRODUCTION A growing body of surveys and studies carried out by the European Foundation for the Improvement of Living and Working Conditions highlighted that time constraints not only continue to be common in the European working population, but are increasing. 325

In order to understand the different type of time constraints, indices normally used include: high working rhythms, repetitive work under frequent deadlines and not enough time to finish work. The reduced tolerance to time pressure in relationship to aging and the age-selective role of time constraints are well known and documented. Instead, as several studies carried out in the ergonomic field have shown, the difficulty, due to age, in coping with time pressure also depends on the possibility for older workers to elaborate their own strategies and skills in order to face working constraints (Pueyo and Volkoff, 2004; Gaudart, 2000). The difficulty arises when work organisation interferes with the individual capacity to manage time constraints and workers are no longer able to take advantage of their experience and workstyle built during their working life (Volkoff and Pueyo, 2005). When analysing working exposure by sex, women are more concerned with the difficulty in facing time pressure with age and it seems that they have less chances, compared to older men, in avoiding less tolerated working conditions (Barbini et al., 2005; Aittomäki et al., 2005). Regarding the consequences of time constraints on health, interactions between work pace, especially when associated with low job discretion, and health complaints were found for musculoskeletal disorders (Nahit et al., 2001; Bernard et al., 1994; Houtman et al., 1994), cardiovascular apparatus (Möller et al., 2005) and general health. Perceived health may be a predictor of mortality as reported in several studies and it may be a useful indicator also to highlight socioeconomic inequalities in health among the working population (Schrijvers et al., 1998). What is currently lacking in terms of studies is what the impact of a “forced” exposure until old age to time pressure – considered as an unavoidable constraint – would have on women’s and men’s health. The aims of this study were: a) first, to analyse the trend of time constraints in women and men with ageing b) second, to investigate the relationships between high working rhythms, age and some health-illnesses, such as Musculoskeletal Complaints (MSC) and Self-Reported Health (SRH).

2 METHODS Analyses were performed on data from the cross-sectional phase of the longitudinal Italian Survey on Health Work and Aging. Data concerned a population of 1,195 Italian workers (297 women and 898 men) aged 32, 37, 42, 47 and 52 at the time of the first data collection (2000) and included work history, working conditions, medical history, perceived health and socio-demographic variables. In the present study ill health status was measured by two indicators, one concerned musculoskeletal complaints (MSC) and the other was based on self-reported health (SRH). Musculoskeletal indicator was defined as reported pain, for over six months, in one of nine local sites of musculoskeletal apparatus: three for upper limbs, three for the spine, three for lower limbs. The presence of multiple pains was also considered, by using the cut point of at least two musculoskeletal complaints. SRH was assessed by the question “How do you rate your current health status?” on a Lickert scale ranging from 0 (very poor) to 10 (very good). Self-reported poor health was classified as belonging to ≤ 5. Work pace was investigated by the variables “repetitive work with frequent deadlines” and “working under time pressure”. The age difficulty in coping with time pressure was investigated in exposed workers by questions “Is it particularly difficult?” and, if affirmative, “Is it particularly difficult with age?” Besides working constraints, housework, measured with number of hours per week, was also considered. Bivariate analysis was used to examine the distribution of health complaints, both musculoskeletal and perceived health, by gender and among cohorts. 326

Pearson’s χ2 test was applied to determine whether differences were significant. Several logistic regression models were fitted to assess associations between health complaints and selected working conditions. Age was considered in the models as a categorical variable, with each category corresponding to five years increment in age. In the first model, age and gender were included as covariates, while the second model was stratified for gender. First, interactions of health complaints and age were directly explored in order to understand the role of age; then time pressure, repetitive work under deadline and perceived difficulty, due to age, in working under time pressure, were also included in the model in order to investigate the relationships between health complaints and working variables. Finally, the difficulty, with age, in working under time pressure was also analysed as a dependent variable of health and job characteristics. Associations were expressed in terms of odds ratios with 95% confidence intervals. All the statistical analyses were performed using STATA 7 software. 3 RESULTS 3.1 Job characteristics Working under time pressure concerned 61.3% of workers. Women reported significantly higher exposure to this constraint (70.8%; p < 0.0001) than men (58.1%). Repetitive work with tight deadlines was reported by 21.9% of workers, and once again with a higher prevalence for women (26.6; p < 0.05). Analyzing the trend of occupational exposure by cohorts, a decreasing frequency may be observed for men in the oldest cohorts, both for time pressure and repetitive work under deadline, while the opposite was observed for women. When analysing the difficulty to working under time pressure with aging, as expected, this condition appeared to be less tolerated by women, who had a significantly higher Odds Ratio than men in all cohorts of age, with a greatest risk in the 52 years cohort. 3.2 Health complaints Regarding the occurrence of health complaints, a progressive increase may be observed up to the 47 year old cohort for MSC. According to the literature, women were more likely to report a higher prevalence of musculoskeletal pain, in all age cohorts. This gap became less evident for perceived health, which showed a similar trend with age in male and female workers. Each of the identified working conditions was positively associated with health complaints, both for MSC and SRH, but in different ways for the two sexes (Table 1). The interaction, by gender, between MSC and working under deadline showed a statistically significantly association, both for upper limbs (p < 0.001) and for multiple MSC (p < 0.05) in women, while in men repetitive work under deadline was associated with poor perceived health (p < 0.001). No interaction was found between family demands and MSC or SRH. The results of logistic regression analyses are shown in Tables 2a and 2b. For women, there was no significant association between health complaints and aging, except for perceived health in the 42 and 47 age cohort. For men, health complaints, both musculoskeletal and SRH, were significantly associated with age, particularly in the older cohort. When analyzing interactions between high working rhythms and health, significant associations between working under time pressure and musculoskeletal complaints, (both upper limbs and multiple pains), were found in women. Working under frequent deadline was significantly associated with poor SRH, for men, and with musculoskeletal complaints for women. 327

Table 1.

Prevalence of health complaints and exposure to time constraints, by gender (%). Women

Men

Multiple Musculoskeletal Complaints Repetitive work under frequent deadline Working under time pressure Perceived difficulty, due to age, to time pressure

45.2* 40.7* 41.9

28.3 26.5 26.8

Upper limbs complaints Repetitive work under frequent deadline Working under time pressure Perceived difficulty, due to age, to time pressure

42.3*** 32.8* 34.0

20.4 17.9 16.6

Self-Rated Health Repetitive work under frequent deadline Working under time pressure Perceived difficulty, due to age, to time pressure

30.1 35.4 46.7***

37.7*** 27.7 33.5***

*p < 0.05 **p < 0.01 ***p < 0.001 Table 2a.

Multivariate analyses by logistic regression. Odds Ratio for health complaints and work constraints’ adjusted for age (women). Multiple Musculoskeletal complaints

Upper limbs complaints

Self-reported Health

O.R.

C.I. (95%)

O.R.

C.I. (95%)

O.R.

C.I. (95%)

Age (32years) 37 years 42 years 47 years 52 years

1 0.81 0.98 1.83 0.76

(0.39–1.69) (0.48–2.01) (0.88–3.80) (0.32–1.78)

1 1.16 1.50 1.34 1.25

(0.51–2.63) (0.68–3.30) (0.58–3.06) (0.49–3.15)

1 1.46 2.32 3.00 1.78

(0.65–3.29) (1.06–5.05) (1.35–6.69) (0.72–4.38)

Working constraints Working under time pressure Working under deadline Perceived difficulty, due to age, to time pressure

1.93 (1.10–3.39) 1.67 (0.96–2.91) 1.44 (0.72–2.86)

2.23 (1.18–4.22) 2.57 (1.42–4.63) 1.16 (0.56–2.41)

1.53 0.90 4.62

(0.86–2.70) (0.50–1.62) (2.04–10.50)

Table 2b.

Multivariate analyses by Logistic regression. Odds Ratio for health complaints and work constraints, adjusted for age and family demands (men).

Age (32 years) 37 years 42 years 47 years 52 years Working constraints Working under time pressure Working under deadline Perceived difficulty, due to age, to time pressure

Multiple Musculoskeletal complaints

Upper limbs complaints

Self-reported Health

O.R.

C.I. (95%)

O.R.

C.I. (95%)

O.R.

C.I. (95%)

1 0.84 1.16 1.83 1.60

(0.49–1.43) (0.69–1.98) (1.11–3.00) (0.99–2.60)

1 1.18 1.05 1.53 2.01

(0.64–2.15) (0.55–2.00) (0.84–2.79) (1.14–3.52)

1 1.32 1.78 1.57 1.90

(0.79–2.18) (1.07–2.97) (0.94–2.61) (1.17–3.08)

(0.95–1.78) (0.92–2.03) (0.73–1.74)

1.17 1.46 0.75

(0.81–1.68) (0.93–2.30) (0.45–1.24)

1.03 1.97 1.66

(0.76–1.40) (1.35–2.87) (1.08–2.53)

1.30 1.13

328

The perceived difficulty, due to age, to work under time pressure was significantly associated with poor perceived health, for the two sexes, with a greater risk for female workers. When the difficulty due to age in coping with time pressure was analyzed as a dependent variable, a significant interaction was found with intellectual discretion (O.R. 1.68). The risk rose to 82% when, in addition to low job control, there was also a lack of responsibility. Strong interaction was found between the difficulty with age in coping with time pressure and poor perceived health (O.R. 2.08) in the 42 and 52 age cohorts, with a greater risk for women (O.R. 1.77). 4 DICUSSION Gender differences in working under time pressure and repetitive work under frequent deadline were observed in this study. Most differences concerned a declining trend for working under time pressure under deadline by age for men, but not for women. Women resulted as more exposed to these working constraints, at all cohorts of age. In addition, the difficulty, due to age, in working under time pressure was greater for women, who had significantly higher odds ratio at all ages. The results suggest that older women had more difficulty to move from adverse working conditions, as well as the possibility for older men in avoiding the more demanding, or less tolerated, work. In other words: time pressure appeared as an age-selective working constraint for men and as an unavoidable working constraint for women, even to the detriment of one’s health. Even though in this study male and female workers were employed in the same working sector, it is difficult to understand if the variable “working under time pressure” had the same meaning or covered the same semantic field for men and women. Assignment of tasks may noticeably differ between the genders within the same occupation. Age was significantly associated with poor perceived health, the risk increased for men exposed to repetitive work under deadline. For the other investigated health indicators, time pressure was significantly associated both with upper limbs and multiple musculoskeletal complaints in women. There are significant differences between sexes in reporting health complaints. As in other studies, women generally have been found to report more musculoskeletal disorders, especially with increasing age, and not all can be explained by the work-family conflict – even if household workload must play a relevant role – or by hormonal changes, since it was observed that for some complaints the association between time constraints and health complaints was adverse also for men. Women were more likely to report symptoms in several areas, which may suggest the concentration of women in jobs involving more risk factors and/or a complex interaction between work, family and psycho-physiological factors, which need to be investigated by further studies. Although time pressure was less tolerated with aging, there were no consistent associations in the expected direction between the perceived difficulty in coping with time pressure and musculoskeletal complaints for women. Instead, significant associations between time pressure and perceived poor health were found, with a strong risk for women. Difficulty due to age in coping with time pressure was associated with job control and poor perceived health. The results presented here point to the need to gather more detailed and objective information about time constraints, including workstyle, multiple social roles, social support, gender discrimination, and to view time constraints as dynamic and influenced by changes in production and the labor market. REFERENCES Aittomäki, A., Lahelma, E., Roos, E., Leino-Arjas, P. and Martikainen, P., (2005). Gender differences in the association of age with physical workload and functioning. Occupational Environment and Medicine, 62: pp. 95–100 Barbini, N., Squadroni, R. and Andreani, M., (2005). Gender differences regarding perceived difficulties at work with age. In: Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers, edited by Costa, G., Goedhard, WJA., Ilmarinen, J. (Amsterdam: Elsevier Press), pp. 17–22

329

Bernard, B., Sauter, S., Fine, L., Petersen, M. and Hales, T., (1994). Job task and psychosocial risk factors for work-related musculoskeletal disorders among newspaper employees. Scandinavian Journal of Work and Environmental Health, 20(6), pp. 417–426 Gaudard, C., (2000). Conditions for maintaining ageing operators at work-a case study conducted at an automobile manufacturing plant. Applied Ergonomics, 31 (5): pp. 453–462 Houtman, IL., Bongers, PM., Smulders, PG., and Kompier, MA., (1994). Psychosocial stressors at work and musculoskeletal problems. Scandinavian Journal of Work Environment and Health, 20(2): pp. 139–145 Möller, J., Theorell, T., de Faire, U., Ahlbom, A. and Hallqvist, J., (2005). Work related stressful life events and the risk of myocardial infarction. Case-control and case-crossover analyses within the Stockholm heart epidemiology programme (SHEEP). Journal of Epidemiology and Community Health, 59 (1): pp. 23–30 Nahit, ES., Pritchard, CM., Cherry, NM., Silman, AJ., and Macfarlane, GJ., (2001). The influence of work related psychosocial factors and psychologicaldistress on regional musculoskeletal pain: a study of newly employed workers. Journal of Rheumatology, 28 (6), pp. 1378–1384 Pueyo, V., and Volkoff, S., (2004). Comprendre que l’opérateur est variable: âge, horaire et activité de travail dans une tâche de contrôle qualité. Economie et Sociétés, 24: pp. 1961–1991 Schrijvers, CT., van de Mheen, HD., Stronks, K. and Mackenbach, JP., (1998). Socioeconomic inequalities in health in the working population: the contribution of working conditions. International Journal of Epidemiology, 27 (6): pp. 1011–1018 Volkoff, S. and Pueyo, V., (2005). How do elderly workers face tight time constraints? In: Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers, edited by Costa, G., Goedhard, WJA., Ilmarinen, J. (Amsterdam: Elsevier Press), pp. 17–22

330

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Age-related differences in insurance claim rates for work-related injuries and diseases for different occupations and industry sectors Karen Munk Centre for Ergonomics and Human factors, School of Human Biosciences, La Trobe University, Australia

Peter Congdon Kmetrics, Melbourne, Australia

Wendy Macdonald Centre for Ergonomics and Human factors, School of Human Biosciences, La Trobe University, Australia

ABSTRACT: In the context of our ageing populations, many governments are now aware of the need to maximise levels of workforce participation among people in their 50s and 60s. To this end, policies must be developed at both government and workplace levels to promote the maintenance of good work abilities into these age ranges and beyond, and more generally to reduce older workers’ risk of work-related injuries or health disorders. There is also a need for better information to address the widespread concern, particularly at a workplace level, that older workers represent a greater financial liability due to perceptions that they present a greater risk of suffering a severe injury or illness. Currently, there is a paucity of adequately detailed information about age-based differences in risk for different types and severities of injury or illness, and how such differences vary between occupations and industry sectors. This paper presents some results from analyses of data from approximately 127,000 individual insurance claims related to work-related injuries in the state of Victoria, Australia over a recent fouryear period. Significantly different perspectives are provided by examining age-related differences in numbers of claims per thousand employees, compared with number of claims per million hours worked, separately for males and females. These analyses are reported separately for main industry sectors. For the highest-frequency claims categories (musculoskeletal disorders and mental disorders) age-related differences in the mechanism of injury or disease are also reported. Results are discussed in terms of some apparent causes of identified increases in risk with age, and in terms of their implications for government policy development to support more effective risk reduction.

1 INTRODUCTION In Australia over the next 15 years, labour force participation rates are expected to decline due to an accelerating increase in numbers of older workers retiring, with consequent constraints on labour supply and a slowing of economic growth. However, these adverse effects of population ageing could be mitigated by measures to reduce the risk of the kind of injuries that would cause people to leave the workforce earlier than they might otherwise choose. A recent study found that many Australian workers plan to work beyond retirement age: 62% of aged-care workers, 72% of construction workers and 66% of finance workers reported that “to support a better lifestyle, they will either need or may choose to work beyond retirement age (at least part-time)” Lundberg and Marshallsay, 2007). Among people in the labour force who intend to retire, the most common factor influencing their decision about when they would retire is personal health or physical abilities (40%) (Australian Bureau of Statistics, 2005). 331

Both state and federal governments in Australia are developing policies and associated programs to encourage workforce participation of those 55 years and older. Taken together, the above survey evidence suggests that there is less need for policy to focus on persuading older workers to continue working, since the majority already are planning for this, than there is for workforce management policies to promote older workers’ health. It is therefore important to identify and implement the most effective means of promoting older workers’ health and their work-related abilities, and to develop strategies to improve OHS practices more generally (Comcare, 2003; Department of Victorian Communities, 2005). There are various reasons why older workers may be more vulnerable to some kinds of injury. While there is wide variability between individuals, and this variability increases with increasing age, it is still true that average levels of some work-related abilities and underlying capacities decline with age. In the physical domain, these include muscular strength, range of joint movement, cardiovascular capacity and aerobic power; in the cognitive domain, there is typically a decline in the maximum rate at which people can process information, so that performance under time pressure of some tasks may decline (Comcare, 2003; Spirduso, Francis and MacRae, 2005). International research evidence on age-related changes in injury risk provides a mixed picture. Numerous studies have found that the risk of accident involvement, and hence the risk of associated injury, decreases with age. On the other hand, some have found an increase with age of reported pain and some kinds of injury (Laflamme, 1996; Laflamme, 1996; Cloutier, David and Duduay, 1998; Chiang, Ko, Yu, Chen, Wu and Chang, 1993; Ringenbach and Jacobs, 1995; Gerr, Marcus, Ensor, Kleinbaum, Cohen, Edwards, gentry, Ortiz and monteilh, 2002). Clearly, broad generalisations about relationships between age and injury risk are unhelpful. This paper investigates older workers’ injury risk based on analyses of ‘workers’ compensation’ insurance claims data from the Australian state of Victoria, where there are approximately 2.5 million workers, demonstrating substantial differences in age-and gender-related variations between industry sectors and injury mechanisms.

2 METHOD Data for workers’ compensation insurance claims in Victoria were compiled for the period July 2002 to June 2006. This dataset included all standard’ claims; it excluded: (a) claims stemming from injury during the journey to and from work (covered by the separate Transport Accident Commission); and (b) non-fatal claims for which there was compensation for less than 11 days (the first 10 days of wages are covered by the employer), and for which medical and related payments were less than the current threshold, which in 2005/06 was $517. Data from the Australian Bureau of Statistics on employment and work rates for males and females in different age groupings in different industries was then used to calculate claims per thousand workers (Cp1000) and claims per million hours worked (CpMhrs) for 115,000 claimants across 12 industries groups and nine occupations groups, broken down by age group and gender. These two different forms of claim rates are used here as proxy measures of injury risk.

3 RESULTS 3.1 Risk varies substantially with age and gender . . . and with the measure used Figures 1 and 2 allow comparison of two different forms of data: claims per 1,000 workers (Cp1000) and claims per million hours worked (CpMhrs). Looking at claims per 1,000 workers (Figure 1), rates are considerably lower for women than men in all age categories. Risk for both men and women increases with age up to 45–54 years; for men it remains steady at that level until beyond 64 years, before finally reducing (perhaps due to a ‘healthy worker’ effect), but for women the decrease commences much earlier. 332

20 18 Claims per 1,000 workers

16 14 12 10 8 6 4 2 0

F

M

F

M

F

F

M

F

M

M

F

M

F

M

F

M

F

M

15 to 19 19 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. Scheme or more average y.o. y.o. y.o. y.o. y.o. y.o. y.o.

Figure 1.

Claims per 1,000 workers (Cp1000) for age and gender for all claims (lost-time and no lost-time).

Claims per million hours worked

0.140 0.120 0.100 0.080 0.060 0.040 0.020 0.000

F

M

F

M

F

M

F

M

F

M

F

M

F

M

F

M

F

M

15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. 15 y.o. y.o. or more or more y.o. y.o. y.o. y.o. y.o. y.o.

Figure 2.

Claims per million hours worked (CpMhrs) for age and gender for all claims (lost-time and no lost-time).

Figure 2 shows that when rates are based on hours worked, gender differences are much smaller, except in the youngest age groups where risk for young males (15–19 years) is second highest, behind males aged 60–64 years. It is also worth noting that women in the two age groups from 45 to 59 years have higher rates than men of the same age. The data presented in the remainder of this paper are in terms of claims per million hours worked. 333

Claims per million hours worked

Lost time claims 13 weeks

0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000

F

M

F

M

15 to 19 20 to 24 y.o. y.o.

Figure 3.

F

M

25 to 34 y.o.

F

M

M

F

M

F

M

F

M

60 to 64 65 y.o. or more y.o.

Claims with three months lost time (65 days paid compensation) per million hours worked (CpMhrs), by age and gender. Fatality

0.0014 Claims per million hours worked

F

35 to 44 45 to 54 55 to 59 y.o. y.o. y.o.

0.0012 0.0010 0.0008 0.0006 0.0004 0.0002 0.0000 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or y.o. y.o. y.o. y.o. y.o. y.o. y.o. more Age group

Figure 4. Workplace fatalities, per million hours worked, for different ages.

3.2 Age and gender effects are different for the most severe injuries The total number of days lost from work for which compensation was paid was available as a proxy measure of injury severity, although it is acknowledged that this measure is also influenced by claims management and rehabilitation processes. Figure 3 presents data (CpMhrs) for the subset of cases where the number of compensated days exceeded three months, and Figure 4 presents data for fatalities. (In these and subsequent figures, the error bars show 95% confidence intervals; no adjustment has been made for multiple comparisons.) It can be seen in Figure 3 that the pattern of gender differences in claims per million hours worked at different ages differs from that in Figure 2. In these more severe cases, the peak for young males is absent, and the increase with age occurs earlier and more steeply for women than for men. Women’s rates are 22% higher than men for 35–44 yrs, 55% higher for 45–54 yrs and 26% higher for 55–59 yrs. In Figure 4, showing fatalities, numbers are very much smaller so male and 334

Construction

0.18 0.16

Female

Male

Claims per million hours

0.14 0.12 0.1 0.08 0.06 0.04 0.02 0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more

Figure 5.

Construction industry: all claims (lost-time and no lost-time; CpMhrs), by age and gender.

Manufacturing 0.25 Female

Male

Claims per million hrs

0.2

0.15

0.1

0.05

0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more

Figure 6.

Manufacturing sector: all claims (lost-time and no lost-time; CpMhrs), by age and gender.

female cases are combined. It can be seen that fatality rates increase nine-fold over working life: from 0.00009 CpMhrs for those aged 15–19 yrs up to 0.00083 CpMhrs for those aged 65+ yrs. 3.3 Risk profiles vary across industry sectors Relationships between age, gender and injury risk vary greatly across industry sectors. This is illustrated in Figures 5 to 8, which show claims rates (CpMhrs; note that the scale differs between these figures) for the following industry sectors: construction; manufacturing; recreation, personal and other services; and retail trade. Figure 5 shows that women working in the construction sector are at much lower risk than men; risk is highest among 20–24 year olds. For men, risk reduces from the youngest group to a low 335

Rec. Pers. & Other

0.08 Female

Male

Claims per million hours

0.07 0.06 0.05 0.04 0.03 0.02 0.01 0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more

Figure 7.

Recreation, Personal Services and Other sector: all claims (lost-time and no lost-time; CpMhrs), by age and gender. Retail trade

0.12 Female

Male

Claims per million hours

0.1 0.08 0.06 0.04 0.02 0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more

Figure 8.

Retail Trade: all claims (CpMhrs, both lost-time and no lost-time), by age and gender.

point for 25–34 year olds, and then increases to more than double for those aged 60–65 years. The huge difference between men and women seems likely to reflect major differences in the types of work performed by men versus women. In the manufacturing sector (Figure 6), risk for men increases only slightly up to around 60 years, beyond which it increases substantially; in contrast, risk for women peaks among 45–54 year olds, where it exceeds that of men, before declining again. In the recreation, personal and other services sector (Figure 7), young males have a higher risk than young females, as in the above two sectors, but the difference is smaller and women’s risk increases more rapidly with age; it exceeds that of men by age 35–44, and remains higher until 336

Body stressing

Claims per million hours worked

0.070 0.060 0.050 0.040 0.030 0.020 0.010 0.000

F

M

15 to 19 y.o.

F

M

20 to 24 y.o.

F

M

F

M

F

M

25 to 34 35 to 44 45 to 54 y.o. y.o. y.o.

F

M

F

M

F

M

55 to 59 60 to 64 65 y.o. or more y.o. y.o.

Figure 9. All CpMhrs (lost-time and no lost-time) attributed to body stressing mechanisms, by age and gender.

beyond 60 years. For men, risk decreases with age until the late fifties, before rising to match that of women in the 60–64 year group. In retail trade (Figure 8), patterns for men and women are very similar. Risk rises steadily with age through to the 45–54 year old group and then declines again, with risk for men being higher at all ages.

3.4 Risk profiles vary between different ‘mechanisms of injury’ Injury compensation claims in this data set are coded according to the main reported ‘mechanism of injury’. Data are reported here for the following five mechanisms: • Mental stress: exposure to traumatic event; exposure to workplace violence; harassment; work pressure; other mental stress factors; • Body stressing: muscular stress while lifting, carrying or putting down objects; muscular stress while handling objects other than lifting, carrying or putting down; muscular stress with no objects being handled; repetitive movement, low muscle loading (in other words, these are musculoskeletal disorders); • Sound and pressure: exposure to single sudden sound; long term exposure to sound; other variations in pressure; • Heat radiation and electricity: contact with hot or cold objects; exposure to environmental heat or cold; exposure to ionising and non-ionising radiation; contact with electricity; • Falls, trips and slips: fall from height; fall on same level; stepping, kneeling or sitting on objects. Claims for musculoskeletal injuries and disorders (‘body stressing’) are the largest single category. As shown in Figure 9, risk increases with age and peaks higher and earlier for women than for men. Risk for men is higher than for women, except in the age groups from 45 to 59 years. Figure 10 shows a much higher risk among women for injuries due to ‘mental stress’, across all age groups; as with ‘body stressing’ injuries, risk peaks in the 45–54 years age group. There is a very different pattern for noise-related injuries, as shown in Figure 11. Claims are largely confined to older workers, and men are at much greater risk. The relationship between risk and age is reversed for injuries associated with heat, radiation and electricity (Figure 12), where risk decreases steeply with age. Claim rates for injuries stemming from ‘chemicals and other 337

Mental stress

Claims per million hours worked

0.0180 0.0160 0.0140 0.0120 0.0100 0.0080 0.0060 0.0040 0.0020 0.0000

F

M

15 to 19 y.o.

F

M

20 to 24 y.o.

M F M F 25 to 34 35 to 44 y.o. y.o.

F M F M F M 45 to 54 55 to 59 60 to 64 y.o. y.o. y.o.

F

M

65 y.o. or more

Figure 10. All CpMhrs (lost-time and no lost-time) attributed to mental stress mechanisms, by age and gender. Sound and pressure Claims per million hours worked

0.018 0.016 0.014 0.012 0.010 0.008 0.006 0.004 0.002 0.000

F M F M 15 to 19 20 to 24 y.o. y.o.

F M 25 to 34 y.o.

F M M F M F F M 35 to 44 45 to 54 55 to 59 60 to 64 y.o. y.o. y.o. y.o.

F M 65 y.o. or more

Figure 11. All CpMhrs (lost-time and no lost-time) attributed to sound and pressure mechanisms, by age and gender.

substances’, ‘hitting objects’, ‘hit by moving object’ are not presented here, but all show patterns similar to that in Figure 12. The final mechanism of injury category presented is ‘slip, trip, fall’ (Figure 13). Here, risk for women increases dramatically with age, more than tripling from 0.009 (25–34 years) to 0.028 (55–59 years). Male rates vary much less with age, peaking at 60–64 years. Further, Figure 14 shows that following such an injury, older women are more than twice as likely to sustain a fracture, compared with older men.

4 IMPLICATIONS FOR PRACTICE Many managers see ‘ageing workforce’issues as intractable problems that are not worth addressing. They believe that most of the injuries sustained (and claims lodged) are the result of a lifetime of 338

Heat, radiation and electricity

Claims per million hours worked

0.0035 0.0030 0.0025 0.0020 0.0015 0.0010 0.0005 0.0000

F

M

15 to 19 y.o.

F

M

F

M

F

M

F

M

F

M

F

M

F

M

20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more

Figure 12. All CpMhrs (lost-time and no lost-time) attributed to heat, radiation and electricity mechanisms, by age and gender. Slip, trip, fall Claims per million hours worked

0.035 0.030 0.025 0.020 0.015 0.010 0.005 0.000

F

M

15 to 19 y.o.

F

M

F

M

F

M

F

M

F

M

20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 y.o. y.o. y.o. y.o. y.o.

F

M

F

M

60 to 64 65 y.o. y.o. or more

Figure 13. All CpMhrs (lost-time and no lost-time) attributed to slip, trip, fall mechanisms, by age and gender.

accumulated wear and tear, and that nothing we do today will have sufficient impact to warrant the cost. The data analysed here tell a different story. Results highlight the importance of analysing such data in sufficient depth to identify ‘hot spots’ that will enable the development of appropriately targeted intervention strategies, with potential for immediate cost-effectiveness. The relationship between age and injury risk was found to vary substantially between industry sectors and between different ‘mechanisms of injury’ or injury types, as well as between genders. Differences between occupations within industry sector are also expected to be very substantial. The large difference in risk profiles shown by Figures 1 and 2 demonstrates the importance of using an appropriate denominator when calculating risk. Australian OHS authorities routinely publish the results of such analyses only in terms of claims per 1000 workers. Consequently, 339

Fractures within slip, trip, fall Claims per million hours worked

0.016 0.014 0.012 0.010 0.008 0.006 0.004 0.002 0.000

F

M

15 to 19 y.o.

F

M

F

M

F

M

F

M

F

M

F

M

F

M

20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or more y.o. y.o. y.o. y.o. y.o. y.o.

Figure 14. All CpMhrs (lost-time and no lost-time) attributed to slip, trip, fall mechanisms and resulting in a fracture, by age and gender.

risk for teenagers and for women tends to be underestimated, since these groups typically work shorter hours (Breslin, Pole and Koehoorn, 2002). It is far more useful to calculate risk based on a denominator that takes account of hours worked, as is the case in jurisdictions such as British Columbia, following the practice of NIOSH and WHO researchers (Department of Health and Human Services, 2003; World Health Organisation, 2007; Hebert, Gervais, Duguay, Champoux and Massicoote, 2005; WorkSafe BC, 2006). These results confirm that claims for cumulative injuries such as hearing loss and many musculoskeletal disorders increase with age, which highlights the need for effective interventions among younger workers. However, they also highlight an age-related increase in claims stemming from slips, trips or falls, from which the injuries themselves are clearly traumatic rather than cumulative. In this case it is workers’ vulnerability that is cumulative with age, probably related both to poorer balance and/or to greater fragility (Comcare, 2003; Spirduso et al., 2005). Occupational health and safety authorities could ensure that in industry sectors where there are high proportions of older women – which in Australia include health and community services, and education – there should be a major focus on strategies to reduce the risk of slips, trips and falls. Such authorities could also look more broadly to develop partnerships with other authorities or groups with mutual interests (e.g. health departments, health promotion practitioners, etc) to target older women through a range of appropriate channels. More active involvement of primary care physicians in diagnosing and treating osteoporosis among those at risk, including referral for appropriate exercise training where appropriate, might also be important (Zuhosky, Irwin, Sable, Sullivan, Pangos and Foye, 2007). Partnership might also include opticians and footwear retail outlets, in educational programs aimed at reducing slip/trip/fall injuries in the older population. Consistent with Ilmarinen’s workability model (Illmarinen, 2001), wherein both the workplace environment and the health and functional capacities of individual workers must be addressed in order to promote workability, WorkSafe Victoria (the OHS regulator for Victoria) is increasing its focus on body stressing hazards. Using both regulatory and educational strategies, it is currently working with government and other parties on a joint health promotion programme for construction workers, addressing issues such as work-life balance, mental health, cardiovascular health, physical fitness and diabetes. Finally, it should be noted that the results reported here represent only a small part of the picture that is emerging from ongoing analyses of this data set. Current work is investigating agerelated differences between occupations nested within industry sectors. Results will be used both in formulating projects to collect additional data on some issues, and to develop targeted interventions for immediate implementation. 340

ACKNOWLEDGEMENTS The authors would like to thank WorkSafe Victoria for providing access to the data used in this study. REFERENCES Australian Bureau of Statistics (2005). 6238.0 – Retirement and Retirement Intentions, Australia, August 2004 to June 2005. www.abs.gov.au (accessed Oct 2007) Breslin, C., Pole, J., and Koehoorn, M. (2002). Working Paper #184. Work injuries among young workers: How much do different methods of calculating denominators matter? Institute for Work and Health. www.iwh.on.ca (sourced Oct 2007) Chiang, H.-C., Ko, Y.-C., Yu, H.-S., Chen, S.-S, Wu, T.-N., and Chang, P.-Y. (1993). Prevalence of shoulder and upper-limb disorders among workers in the fish processing industry. Scandinavian Journal of Work, Environment and Health, 19: pp. 126–31 Cloutier, E., David, H., and Duduay, P. (1998). Accident indicators and profiles as a function of the age of female nurses and food services workers in the Quebec health and social services sector. Safety Science, 28 (2) Comcare (2003). Productive and safe workplaces for an ageing workforce. www.apsc.gov.au/publications03/ maturecomcare.htm (sourced Oct 2007) Department of Health and Human Services (NIOSH) (2003). NIOSH Alert. Deaths, Injuries, and Illnesses of Young Workers Publication No. 2003–128. www.cdc.gov/niosh Publication No. 2003–128 Department of Victorian Communities (2005). Victoria: Working Futures. Report of Victoria’s Workforce participation taskforce 2005. www.dvc.vic.gov.au (sourced Oct 2007) Gerr, F., Marcus, M., Ensor, C., Kleinbaum, D., Cohen, S., Edwards, A., Gentry, E., Ortiz, D., and Monteilh, C. (2002). A prospective study of computer users: I. Study design and incidence of musculoskeletal symptoms and disorders. American Journal of Industrial Medicine, 41: pp. 221–235 Hébert, F., Gervais, M., Duguay, P., Champoux, D., and Massicotte, D. (2005). Young People: Work Constraints And Risks Institut De Recherche Robert-Sauvé En Santé Et En Sécurité Du Travail IRSST. www.irsst.qc.ca (sourced Oct 2007) Illmarinen, J. (2001). Ageing workers. Occupational and Environmental Medicine, 58: pp. 546–552 Laflamme, L. (1996). Age-related accident ratios in assembly work: a study of female assembly workers in the Swedish automobile industry. Safety Science, 23(1): pp. 27–37 Laflamme, L. (1996). Age-related accident risks among assembly workers: a longitudinal study of male workers employed in the Swedish automobile industry. Journal of Safety Research, 27(4): pp. 259–268 Lundberg, D., and Marshallsay, Z. (2007). Older workers’ perspectives on training and retention of older workers. A national vocational education and training, research and evaluation program report. www.ncver.edu.au (sourced Oct 2007) Ringenbach, K., and Jacobs, R. (1995). Injuries and aging workers. Journal of Safety Research, 26(3): pp. 169–176 Spirduso, W., Francis, K., and MacRae, P. (2005). Physical dimensions of ageing. 2nd Ed. Human Kinetics, Champaign WorkSafe, BC. (2006). 2006 Statistics.www.worksafebc.com (accessed Oct 2007) World Health Organisation (2007). Work injuries in children and young people. Fact Sheet No. 4.7, May 2007, www.euro.who.int/ Document/EHI/ENHIS_Factsheet_4_7.pdf (sourced Oct 2007) Zuhosky, J., Irwin, R., Sable, A., Sullivan, W., Panagos, A., and Foye, P. (2007). Industrial Medicine and Acute Musculoskeletal Rehabilitation. 7. Acute Industrial Musculoskeletal Injuries in the Aging Workforce. Arch Phys Med Rehabil Vol 88, Suppl: pp. 134–39

341

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Age-related differences in patterns of return to work and compensation costs following work-related injury or illness Karen Munk Centre for Ergonomics and Human Factors, School of Human Biosciences, La Trobe University, Australia

Peter Congdon Kmetrics, Melbourne, Australia

Wendy Macdonald Centre for Ergonomics and Human Factors, School of Human Biosciences, La Trobe University, Australia

ABSTRACT: There is some evidence that older workers who suffer work-related injuries or illness are likely to be absent from work for longer periods before returning to work – if they return at all. However, there is very little publicly available information concerning the magnitude of the effect, how it might vary between different types of injury/illness. This paper presents information on these issues, based on analyses of approximately 127,000 individual insurance claims related to work-related injuries in the state of Victoria, Australia over a recent four-year period. Age-related differences in numbers of days compensated and medical costs are reported for different injury/illness categories, separately by gender. Results are discussed in terms of their implications for the development of more effective policies to facilitate the rehabilitation and early return to work of older workers.

1 INTRODUCTION Work-related injuries result in major costs both to the individuals involved and the wider community. In the Australian state of Victoria, where there are approximately 2.5 million workers, injury compensation costs in 2005/06 were more than $AUS 1.2 billion (Victorian WorkCoverAuthority, 2006). Establishing effective return to work (RTW) programs that result in injured workers returning to work as quickly as possible, either to pre-injury or alternative duties, is clearly important to minimise this economic burden. Injured workers themselves are interested in minimising personal costs and re-integrating within the workforce as soon as possible, and RTW duration is also an indicator of the effectiveness of rehabilitation and other healthcare professionals. To facilitate more effective management of the RTW process, identification of early prognostic factors would be useful, enabling discrimination of those individuals who are likely to recover quickly with minimal medical or other attention, from those who are likely to require intensive medical and/or vocational interventions (Hogg-Johnson and Cole, 2003). On this basis, RTW programs could be tailored to meet individuals’ differing needs. In particular, high-risk patients could be offered specific treatment and/or rehabilitation interventions at an early enough stage to minimise their risk of progressing to chronic disability. Older age has been identified as one predictor of prolonged work disability (Okurowski, Pranski, Webster, Shaw and Varma, 2003; Steenstra, Verbeek, Heymans and Bongers, 2005; Seland, Cherry and Beach, 2006), and in Victoria, RTW outcomes are worse for workers aged 55+ years (Victorian WorkCover Authority, 2006). This paper presents results from some analyses of Victorian workers’ compensation insurance claims data. Age-related patterns in number of days compensation paid 343

(wage replacement) and in medical costs for all claims together are reported for five different injury types. Such information may be used in identifying the types of injuries for which early interventions to facilitate recovery will be worthwhile, particularly for older workers.

2 METHODS Data for ‘workers’ compensation’ insurance claims in Victoria were compiled for the period July 2002 to June 2006. This dataset included all ‘standard’claims, which excluded: (a) claims stemming from injury during the journey to and from work (covered by the separate Transport Accident Commission); and (b) non-fatal claims for which there was compensation for less than 11 days (the first 10 days of wages are covered by the employer), and for which medical and related payments were less than the current threshold, which in 2005/06 was $517. Data from the Australian Bureau of Statistics on employment and work rates for males and females in different age groupings in different industries was used to calculate claims per thousand workers (Cp1000) and claims per million hours worked (CpMhrs) for 115,000 claimants across 12 industries groups, broken down by age group and gender. These two different forms of claim rates are used here as proxy measures of injury risk. For each claim, data were also available on number of days lost from work for which compensation was paid by the scheme, and on the associated dollar costs of wage replacement and medical expenses. For analyses based on number of compensated days lost, all claims for which zero days were compensated were removed from the sample, leaving 64,809 cases.

3 RESULTS 3.1 Total compensation payments increase with increasing age The total compensation cost reported here is the sum of wage replacement costs and medical costs; common law payouts are not included. Figure 1 shows total compensation costs (sum of wage replacement and medical costs), broken down by age and gender. Median rather than mean values are presented, since the distribution was quite skewed by relatively small numbers of very high values. Error bars in this and some subsequent figures show 95% confidence intervals; no adjustment has been made for multiple comparisons. It can be seen that cost increases with age, and that there is an increasing gender gap in the older age groups. One possible reason for the large variation in cost with age, and to a lesser degree with gender, is variation in wage level, which in turn would be influenced by variations across age and gender in proportions of part-time versus full-time employees. 3.2 Both number of days compensated and medical costs increase with age Figure 2 shows variation in the median number of days for which compensation was paid, separately by age and gender (which serves here as a proxy for total amount of lost time). This increases from 17 days for those aged 15–19 years up to 51 days for those aged 60–64 years (Figure 2). For all age groups, women lost considerably more time than men; this gender difference was greatest for those aged 35–44 years. Median values of medical costs. by age and gender, are shown in Figure 3. It can be seen that cost rises with age, and are higher for males than for females. Women’s medical costs asymptote for those aged around 50 years, whereas for men they continue to rise up to around 60 years of age. Since neither days nor medical costs would be expected to vary substantially with wage level, these data suggest that differences in wage levels are not the only factor responsible for age- and gender-related differences in overall costs. 344

Med total payment amount

Gender F M

$6,000

$4,000

$2,000

$0 45 to 55 to 60 to 65 to y.o. 15 to 20 to 25 to 35 to 19 y.o. 24 y.o. 34 y.o. 44 y.o. 54 y.o. 59 y.o. 64 y.o. or more Age group

Figure 1.

Median cost of claim (wage replacement costs plus medical costs) in Australian dollars, separately by age and gender.

Median days paid by the scheme

70 60 50 40 30 20 10 0 F

M

15 to 19 y.o.

Figure 2.

F

M

20 to 24 y.o.

F

M

25 to 34 y.o.

F

M

35 to 44 y.o.

F

M

45 to 54 y.o.

F

M

55 to 59 y.o.

F

M

60 to 64 y.o.

F

M

65 y.o. or more

Median number of days for which compensation was paid, separately by age and gender (excluding claims for which zero compensation was paid for days lost).

3.3 Claim rates, medical costs and compensated days differ for key injury types Claim rates (CpMhrs) and median values for medical costs and number of days compensated are presented for five different injuries types, selected as representing some of the most costly and frequent types of injury in Victoria (Figures 4 to 13). Key findings are summarised below for each injury type separately. 345

Med medical amount

$2,000

Gender F M

$1,500

$1,000

$500

$0 45 to 55 to 60 to 65 y.o. 15 to 20 to 25 to 35 to 19 y.o. 24 y.o. 34 y.o. 44 y.o. 54 y.o. 59 y.o. 64 y.o. or more Age group

Figure 3.

Median value of medical costs paid (Australian dollars), separately by age and gender. Fractures

Claims per million hours worked

0.012 0.010 0.008 0.006 0.004 0.002 0.000 15 to 19 y.o.

Figure 4.

20 to 24 y.o.

25 to 34 35 to 44 45 to 54 y.o. y.o. y.o. Age group

55 to 59 y.o.

60 to 64 65 y.o. or y.o. more

Claims rates (per million hours worked, CpMhrs) for fractures.

3.3.1 Fractures (Figures 4 and 5) • Claim rates increased with age from a low for those aged 25–34 years (CpMhrs = 0.0059) to a high for those aged 60–64 year (CpMhrs = 0.0088). • Number of compensated days increased with age from 19 days to 35 days. • Medical costs increased with age from $1,258 (15–19 years) to $2,360 (60–64 years).

3.3.2 Carpal tunnel (Figures 6 and 7) • Claim rates were highest for the 45–54 year group (CpMhrs = 0.0019). • Number of days compensated increased with age, starting from a low of 13 days up to 21 days for the 55–59 years age group; there was a sharp increase to 47 days for those aged 65 years or older. • Medical costs more than doubled with age from $978 (15–19 years) to $2,148 (55–59 years), with a sharp increase to $3,190 for those aged 65 years or older. 346

Fractures

$2,500

Scheme paid days

35

Medical costs

$2,000

30 25

$1,500

20 $1,000

15 10

$500

Median medical costs

Median scheme paid days

40

5 $0

0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or y.o. y.o. y.o. y.o. y.o. y.o. y.o. more Age group

Figure 5.

Median medical cost ($AUS) and number of compensated days for fractures. Carpal tunnel

Claims per million hours worked

0.003 0.002 0.002 0.001 0.001 0.000 15 to 19 y.o.

20 to 24 y.o.

25 to 34 y.o.

35 to 44 45 to 54 y.o. y.o.

55 to 59 y.o.

60 to 64 65 y.o. or y.o. more

Age group

Figure 6.

Claims rates (claims per million hours worked, CpMhrs) for carpal tunnel injury.

3.3.3 Open wounds (Figures 8 and 9) • Claim rates were highest for the youngest workers (15–19 years; CpMhrs = 0.02), after which they fell steeply down to the mid-30s age range. • Number of compensated days was low and varied little with age, except for a sharp increase for those aged 65 years or older. • Medical costs more than doubled with age, from $1,284 for the youngest group to $2,605 for the oldest. 3.3.4 Back pain (Figures 10 and 11) • Claim rates were highest for those aged 45–54 years (CpMhrs = 0.16). • Number of days compensated was steady at 5 days from 20–24 years through 55–59 years, before peaking at 60–64 years (12 days) and then falling for those aged 65 years or older (seven days). 347

Carpal tunnel $3,500

45

Scheme paid days

40

Medical costs

$3,000

35

$2,500

30

$2,000

25 20

$1,500

15

$1,000

10

Median medical costs

Median scheme paid days

50

$500

5

$0

0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or y.o. y.o. y.o. y.o. y.o. y.o. y.o. more Age group

Figure 7.

Median medical cost ($AUS) and compensated days for carpal tunnel injuries.

Claims per million hours worked

0.025

Open wound

0.020

0.015

0.010

0.005

0.000 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or y.o. y.o. y.o. y.o. y.o. y.o. y.o. more Age group

Figure 8.

Claims rates (claims per million hours worked, CpMhrs) for open wounds.

• In contrast, medical costs rose rapidly up to 25–30 years, after which the rate of increase was lower, peaking at 55–59 years and decreasing thereafter. These costs more than tripled with age from $385 to $1,211.

3.3.5 Stress (Figures 12 and 13) • Claim rates rise steadily to peak at the 45–54 years age group (CpMhrs = 0.0108), and then decline steeply so that the levels are similar for the youngest and oldest groups. • Number of days compensated peaks at the same age (21 days); the peak continues through 55–59 years before declining less steeply, remaining considerably higher than the youngest age groups. • Medical costs follow the same pattern as the claims rate, peaking at 45–54 yrs ($479). 348

Open wound

$3,000

7 Scheme paid days

$2,500

Medical costs

5

$2,000

4 $1,500 3 $1,000

2

$500

1

$0

0 15 to 19 y.o.

Figure 9.

20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. or y.o. y.o. y.o. y.o. y.o. y.o. more Age group

Median medical cost ($AUS) and compensated days for – open wounds.

Back pain

0.018 Claims per million hours worked

Median medical costs

Median scheme paid days

6

0.016 0.014 0.012 0.010 0.008 0.006 0.004 0.002 0.000 15 to 19 y.o.

Figure 10.

20 to 24 25 to 34 y.o. y.o.

35 to 44 45 to 54 55 to 59 60 to 64 y.o. y.o. y.o. y.o. Age group

65 y.o. or more

Claims rates (claims per million hours worked, CpMhrs) for back pain.

4 IMPLICATIONS FOR PRACTICE The results reported above demonstrated that the total cost of compensation claims was higher for older workers, consistent with some previous research (Wood, Morrison and Macdonald, 1993). Some others have found no significant relationship between cost and age (Peela, Xua and Colombib, 2005), although in the latter case the claims related only to one type of injury – musculoskeletal disorders. Considering all of the above injury types together, a general trend was evident for older workers to take longer to return to work following an injury. This finding is consistent with other research evidence on lost time following injury (Laflamme, 1995; Cloutier, David and Duduay, 1998; Mital, Pennathur and kansal, 1999; Lydell, Baigi, Marklund and Mansson, 2005). 349

Back pain

14

$1,400

12

$1,200

Medical costs

10

$1,000

8

$800

6

$600

4

$400

2

$200

0

Median medical costs

Median scheme paid days

Scheme paid days

$0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. or more y.o. y.o. y.o. Age group

Figure 11.

Median medical cost ($AUS) and compensated days for back pain. Stress

Claims per million hours worked

0.012 0.010 0.008 0.006 0.004 0.002 0.000 15 to 19 y.o.

20 to 24 y.o.

25 to 34 y.o.

35 to 44 y.o.

45 to 54 y.o.

55 to 59 y.o.

60 to 64 65 y.o. or y.o. more

Age group

Figure 12.

Claims rates (claims per million hours worked, CpMhrs) for stress claims.

One probable cause of the age-related increase in days compensated is the high prevalence of co-morbidities among older workers, since it has been shown that those with co-morbid conditions are at risk of a more prolonged healing process (Benjamin and Pransky, 2001). There is evidence that for the Victorian population from which the present sample was drawn show that: • of those aged 35–44 years, 13 percent reported a long term cardiovascular condition, increasing to 23 percent for those aged 45–54 years and 63% for those aged 75 years and over (Australian Bureau of Statistics, 2006); • prevalence of self-reported diabetes ranged from 3.0 percent of people aged 16–24 years, to 16.8 percent of those aged 65–74 years (State Government of Victoria, 2006); • almost half of the adult population (48%) are categorised as being either overweight or obese, and the prevalence of obesity increases steadily up to the age group 55–64 years, after which it falls (Australian Centre for the Study of Obseity, 2006); 350

Stress 25

$600

20

$500

Medical costs

$400 15 $300 10 $200 5

Median medical costs

Median scheme paid days

Scheme paid days

$100 $0

0 15 to 19 20 to 24 25 to 34 35 to 44 45 to 54 55 to 59 60 to 64 65 y.o. y.o. y.o. y.o. y.o. y.o. y.o. y.o. or more Age group

Figure 13.

Median medical cost ($AUS) and compensated days for stress claims.

• and prevalence of depression tends to increase with age (Zuhosky, Irwin, Sable, Sullivan, Pangos and Foye, 2007) and 17.9 percent of Victorians over 18 years report depression or anxiety conditions State Government of Victoria, 2006). Depression may be a particularly significant condition, since it can be associated with a wide range of factors, including the injury that was the initial cause of the claim. The presence of depressive symptoms in workers with musculoskeletal disorders is associated with more days on total compensation (Lotters, Franche, Hogg-Johnson, Burdof and Pole, 2006). It appears that treating health professionals should be alert to the possible presence of depression during the injury recovery process, especially in those patients who are not responding to treatment as would normally be expected. Consistent gender differences have been found in this study, as previously (Seland et al., 2006), with women generally taking longer to return to work. This contrasts with the situation for claims rates where typically there is little disparity between men and women, which can lead to the importance of gender differences being overlooked. Neither gender nor age can themselves be modified, so their effects might be seen as not warranting particular analysis. However, the data reported here demonstrate the potential value of considering both these variables when developing claims management and RTW programs, particularly for women aged over 35 years. WorkCover South Australia (another Australian state workers compensation jurisdiction) has identified gender differences as important, and argued that current rehabilitation and RTW programs do not adequately take these into account. They particularly highlight the pressures of balancing work, domestic and community life, arguing that there is inadequate recognition at a systems level of the greater challenges confronted by women in combining paid and unpaid work (WorkCover Corporation SA, 2005). This paper extends previously published research by addressing differences in RTW patterns associated with different types of injury. It is beyond the scope of this paper to discuss all of these differences; two that are particularly noteworthy are fractures and back pain. For fractures, it was found that number of days compensated increased with age, consistent with results from similar analyses of Canadian data from the province of Alberta (Seland et al., 2006). Further research is needed to determine the reasons for this pattern. Munk and colleagues (these proceedings) reported that falls in the workplace are more likely to result in fractures for women aged over 55 years, which might possibly be related to decreasing bone density. Further, they reported that the risk of a fall increases very significantly with age; potential causes might include deteriorating vision, and deteriorating balance (Comcare, 2006). The importance of identifying and 351

treating cases of osteoporosis in this population has been identified by other researchers as very important, whether within or outside the workers’ compensation system (Zuhosky et al., 2007). In this kind of way, medical and rehabilitation professionals could play a major role in the prevention of further injuries, as well as in treating injuries. The state authorities responsible for occupational health and safety and workers’ compensation have yet to realise fully the potential of partnerships with such professionals. In the case of back pain, it was found here that number of days compensated remained stable until 59 years, before doubling for workers aged 60–64 years and then falling again for the oldest age group. Steenstra and colleagues (2005) concluded from a literature review that there is strong evidence for age as a prognostic factor for longer duration of sick leave, especially in those aged over 51 years. However, their review included two high quality studies and three lower quality studies that found no association with age. Although number of days compensated for back injuries was less in the present study than might have been expected, it was found that associated medical costs for this type of injury more than doubled with age. Increases in medical costs with age were also found for the other types of injury, which clearly presents major challenges to the medical community. An example of the kind of strategy that appears helpful is provided by Zuhosky and colleagues (2007), who recently published a learning module for practitioners and trainees in physical medicine and rehabilitation which highlights the unique challenges faced by physicians when treating members of the aging workforce. Finally, it is acknowledged that numerous factors other than those considered here have been identified by previous researchers as predictors of poor RTW outcomes (Okuroski et al., 2003). However, the evidence presented here suggests that taking account of age, gender and the type of injury has the potential to promote more effective treatment and rehabilitation processes, resulting in substantial benefits both to the individuals concerned and to the wider community.

ACKNOWLEDGEMENTS The authors thank WorkSafe Victoria for providing access to the data used in this study. REFERENCES Australian Bureau of Statistics (2006). 4821.0.55.001 – Cardiovascular Disease in Australia: A Snapshot, 2004–05. www.abs.gov.au (sourced 25 Sept 2007) Australian Centre for the Study of Obesity (2006). Obesity in Australian adults: Prevalence data. www.asso.org.au/freestyler/gui/files//factsheet_adults_prevalence.pdf (sourced 25 Sept 2007) Benjamin, K., and Pransky, G., (2001). Occupational injuries and the older worker: Challenges in research, policy and practice. Southwest J Aging, 16: pp. 14–17 in G. Pransky, K. Benjamin, and J. Savageau, (2005). Early retirement due to occupational injury: who is at risk. American Journal of Industrial Medicine, 47: pp. 285–295 Cloutier, E., David, H., and Duduay, P., (1998). Accident indicators and profiles as a function of the age of female nurses and food services workers in the Quebec health and social services sector. Safety Science, 28(2) Comcare, (2003). Productive and safe workplaces for an ageing workforce. www.apsc.gov.au/publications03/ maturecomcare.htm (sourced Oct 2007) Hogg-Johnson, S., and Cole, D., (2003). Early prognostic factors for duration on temporary total benefits in the first year among workers with compensated occupational soft tissue injuries. Occup Environ Med, 60: pp. 244–253 Laflamme, L., (1995). Aging and occupational accidents. A review of the literature of the last three decades. Safety Science, 21: pp. 145–161 Lotters, F., Franche, RL., Hogg-Johnson, S., Burdorf, A., and Pole, J., (2006). The prognostic value of depressive symptoms, fear-avoidance, and self-efficacy for duration of lst-time benefits in workers with musculoskeletal disorders. Occup Environ Med, 63: pp. 794–801

352

Lydell, M., Baig, A., Marklund, B., and Mansson, J., (2005). Predictive Factors For Work Capacity In Patients With Musculoskeletal Disorders. J Rehabil Med, 37: pp. 281–285 Mital, A., Pennathur, A., and Kansal, A., (1999). Nonfatal occupational injuries in the United States Part I – overall trends and data summaries. International Journal of Industrial Ergonomics, 25: pp. 109–129 Okurowski, L., Pransky, G., Webster, B., Shaw, W., and Verma, S., (2003). Prediction of prolonged work disbility in occupational low-back pain based on nurse case management data. JOEM 45 (7); pp. 763–770 Peelea, P., Xua, Y., and Colombib, A., (2005). Medical care and lost work day costs in musculoskeletal disorders: Older versus younger workers. International Congress Series Assessment and Promotion of Work Ability, Health and Well-being of Ageing Workers Volume 1280, June: pp. 214–218 Seland, K., Cherry, N., and Beach, J., (2006). A Study of Factors Influencing Return to Work After Wrist or Ankle Fractures. American Journal Of Industrial Medicine 49: pp. 197–203 State Government of Victoria, Australia, Department of Human Services (2006). Victorian Population Health Survey 2005 www.health.vic.gov.au/healthstatus/vphs_current.htm (sourced 12 Sept 2007) State Government of Victoria, Australia, Department of Human Services, Victorian Government Health Information, (June 2006) www.health.vic.gov.au/nhpa/diabetes.htm (sourced 13 Sept 2007). Steenstra, I., Verbeek, J., Heyman, SW., Bongers, P., (2005). Prognostic factors of sick leave in patients sick listed with acute low back pain: a systematic review of the literature. Occ Environ med 62: pp. 851–860 Victorian WorkCover Authority, Statistical Summary 2005–2006. www.worksafe.vic.gov.au (sourced 10 Sept 2007) Victorian WorkCover Authority (2006). Injured Worker Survey 2006, www.workcover.vic.gov.au (sourced 10 Sept 2007) Wood, G., Morrison, D., and Macdonald, S., (1993). Factors influencing the cost of workers’ compensation claims: The effects of settlement method, injury characteristics, and demographics. Journal of Occupational Rehabilitation 3(4): pp. 201–211 WorkCover Corporation SA (2005). Gender, workplace injury and return to work: A South Australian perspective. www.workcover.com (sourced June 21 2007) Zuhosky, J., Irwin, R., Sable, A., Sullivan, W., Panagos, A., and Foye, P., (2007). Industrial Medicine and Acute Musculoskeletal Rehabilitation. 7. Acute Industrial Musculoskeletal Injuries in the Aging Workforce. Arch Phys Med Rehabil Vol 88, Suppl: pp. 134–39

353

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Keyword Index

A Aging, 219 B Balance, 271 Base of support, 271 C Clinical assessment, 253 Combined system, 263 D Disabled workers, 117 E Elderly adult, 271 Elderly stroke patient, 253 F Female workers’ superior peculiarity, 219 Force plate, 263 H Head mounted display system, 253 Health, 41, 83

I Intergenerational relations, 155 J Job satisfaction, 83 K KAIZEN, 219 O Occupation, 83 Older subordinates, 155 R Reference values, 71 Rehabilitation engineering, 263 Retirement, 83 S SMEs, 37 Social inclusion, 117 Standing balance, 263

355

Super-advanced age and fewer children, 219 T Teachers, 41 Toes and soles, 263 U Unilateral spatial neglect, 253 V Vibratory stimuli, 263 Virtual reality, 253 W Workability, 83 Work Ability Index, 37, 41, 71, 83, 117 Work conditions, 37 Y Young employees, 71 Younger managers, 155

Promotion of Work Ability towards Productive Aging – Kumashiro (ed) © 2009 Taylor & Francis Group, London, ISBN 978-0-415-48590-6

Author Index

Ahonen, G. 183 Alavinia, S.M. 89 Ando, M. 293

Järvelin, S. 75 Jiang, J. 27 Johansson, I. 155

Noda, N. 293 Norma, B. 325 Näsman, O. 183

Bredt, F.J. 89 Brooke, L. 27 Burdorf, A. 15, 89

Kawakami, M. 275 Kawata, Y. 303, 319 Khor, A. 45 Kimura, T. 247 Kiss, P. 239 Kobayashi, D. 227 Kristjuhan, Ü. 209 Kumashiro, M. 9, 59, 219

Oakman, J. 133 Okui, Y. 293 Oyama, Y. 253, 263, 271

Camerino, D. 283 Cocco, P. 113 Congdon, P. 331, 343 Conway, P.M. 283 Corrêa-Filho, H.R. 37 Costa, G. 93, 113, 283 De Meester, M. 239 Duarte, L.R. 161 Elders, L.A.M. 15, 89 Faber, A. 105 Fernandes, A.P. 71 Fichera, G.P. 93 Francesco, S. 325 Freude, G. 33 Furunes, T. 169, 177 Gauggel, B. 33 Goedhard, R.G. 139 Goedhard, W.J. 3 Goedhard, W.J.A. 139 Goes, E.P. 71 Hashimoto, M. 293 Hasselhorn, H.M. 33 Healy, P. 27 Hirosawa, M. 299, 303, 307, 311, 315, 319 Hodge, E.P. 71, 117 Ifukube, T. 253, 263, 271 Iida, K. 219 Ilmarinen, J. 37, 71 Ino, S. 253, 263, 271 Izumi, H. 59 Izumi, T. 263

Lindeboom, D. 89 Louhevaaral, V. 19, 75 Macdonald, W. 331, 343 Maeda, Y. 253, 263, 271 Majery, N. 121 Malo, J.-L. 59 Marquié, J.-C. 161 Martikainen, R. 101 Matsuda, F. 191 McCarthy, P. 127 Melon, M. 113 Mikami, K. 219 Misawa, T. 191 Miura, T. 247 Miyamoto, C. 293 Mizuno, M. 299, 303, 307, 311, 315, 319 Mizuno, Y. 191 Monteiro, I. 37, 41, 71, 117 Moore, S. 127 Motegi, N. 191 Müller, B.H. 33 Munk, K. 331, 343 Mykletun, R.J. 169, 177 Nakanishi, Y. 275 Neri, L. 93 Nga, N.N. 65 Nishi, Y. 299, 303, 307, 311, 315, 319 357

Palermo, J. 45 Rees, A. 83 Reilly, T. 83 Rentzsch, M. 213 Ropponen, A. 19 Rosa, S. 325 Sakai, K. 191 Sartori, S. 93, 283 Schramm, J. 195 Seitsamo, J. 37, 71, 101 Seliger, D. 213 Sell, L. 105 Shinohara, K. 247 Shirogane, S. 253, 263, 271 Smith, K. 45 Smolander, J. 19 Søgaard, K. 105 Stattin, M. 143 Suenaga, Y. 293 Sugihara, S. 253, 263 Sugiura, M. 191, 299, 303, 307, 311, 315, 319 Takahara, M. 247 Tanaka, S. 299, 303, 307, 311, 315, 319 Tanaka, T. 253, 263, 271 Taylor, P. 27 Tempel, J. 33, 195 Tielsch, R. 33 Tipton, M. 83 Tokuhiro, M. 59 Tuomi, K. 37, 71 Tuominen, E. 37

Uehara, N. 59 van den Berg, T.I.J. 15, 89 Van, D.K. 65 Vedovato, T.G. 41

Webber, L. 45 Wells, Y. 133 Yamada, Y. 299, 303, 307, 311, 315, 319

358

Yamamoto, S. 227 Yamanaka, K. 275 Yoshikawa, T. 191