stress

E u r o p e a n A g e n c y f o r S a f e t y a n d H e a l t h a t W o r k

R E S E A R C H

Research onWork-related Stress

EN EN

RE

SE

AR

CH

Research on Work-rela

ted S

tres

s

In order to encourage improvements,

especially in the working environment, as

regards the protection of the safety and

health of workers as provided for in the

Treaty and successive action programmes

concerning health and safety at the

workplace, the aim of the Agency shall be

to provide the Community bodies, the

Member States and those involved in the

field with the technical, scientific and

economic information of use in the field of

safety and health at work.

Eu

ro

pe

an

A

ge

nc

y

fo

r

Sa

fe

ty

a

nd

H

ea

lt

h

at

W

or

k

ht

tp

:/

/o

sh

a.

eu

.i

nt

9 789282 892558 >

ISBN 92-828-9255-7

54

TE

-28-00-882-EN

-C

European Agencyfor Safety and Healthat Work


Gran Vía 33, E-48009 Bilbao, Spain

Tel. +34 944 794 360; Fax. +34 944 794 383

Email: [email protected]

OFFICE FOR OFFICIAL PUBLICATIONOF THE EUROPEAN COMMUNITIES

L-2985 Luxembourg

Price (excluding VAT) in Luxembourg: EUR 11

.

R E S E A R C H


Research onWork-related Stress

A great deal of additional information on the European Union is available on the Internet.It can be accessed through the Europa server (http://europa.eu.int).

Cataloguing data can be found at the end of this publication.

Luxembourg: Office for Official Publications of the European Communities, 2000

ISBN 92-828-9255-7

© European Agency for Safety and Health at Work, 2000Reproduction is authorised provided the source is acknowledged.

Printed in Belgium


3■

Report written by

Tom Cox

Amanda Griffiths

Eusebio Rial-González

Institute of Work, Health & OrganisationsUniversity of Nottingham Business SchoolJubilee Campus, Wollaton RoadNottingham NG8 1BBUnited Kingdom

R e s e a r c h o n W o r k - r e l a t e d S t r e s s

■4


5■

C o n t e n t sEXECUTIVE SUMMARY 91 TERMS OF REFERENCE 212 INTRODUCTION 23

2.1 The Nature of Work-related Stress in a Life Perspective 242.2 The Extent of the Problem 272.3 Summary 30

3 DEFINING STRESS 313.1 Conceptualisation and Frameworks 32

3.1.1 Engineering Approach 323.1.2 Physiological Approach 333.1.3 Psychological Approach 35

3.2 Interactional Theories of Stress 373.2.1 Person-Environment Fit 373.2.2 Demand-Control Model 38

3.3 Transactional Definitions 413.3.1 Theories of Appraisal and Coping 41

3.4 Summary: frameworks, theories and definitions 443.5 Coping 46

3.5.1 Coping Taxonomies 463.5.2 Coping as Problem-solving 47

3.6 Individual and Group Differences 493.6.1 Type A Behaviour 503.6.2 Vulnerable Groups 513.6.3 Selection 52

3.7 Summary: individual differences – work ability and coping 534 METHODOLOGICAL ISSUES 55

4.1 Measurement 564.2 Self-report Data and Triangulation 574.3 Summary 60

5 WORK HAZARDS AND STRESS 615.1 Physical Hazards 64

5.1.1 Noise 645.1.2 Other Physical Factors 65

5.2 Psychosocial Hazards 675.2.1 Context to Work 69


■6

5.2.2 Content of Work 755.2.3 New hazards: “The changing world of work” 80

5.3 Animal Studies 825.4 Distribution of Psychosocial Work Hazards 845.5 Summary 86

6 STRESS AND HEALTH 876.1 Effects of Stress: an overview 896.2 Psychological and Social Effects 906.3 Physiological and Physical Effects 92

6.3.1 Mechanisms of Stress-related Physiopathology 926.4 Work-related Psychoimmunology 95

6.4.1 Mechanisms 966.4.2 Other Pathologies 96

6.5 Organisational Effects 986.6 Summary 99

7 THE ASSESSMENT AND MANAGEMENT OF WORK-RELATED STRESS 1017.1 The Assessment of Work-related Stress: the control cycle 1027.2 A Risk Management Approach to Work-related Stress 1067.3 The Management of Work-related Stress 1097.4 Principles of Stress Management 110

7.4.1 Objectives 1107.4.2 Agency and Target 111

7.5 Common Interventions: Their Effectiveness 1127.6 Summary 120

8 CONCLUSIONS 1218.1 Defining Stress 1228.2 Individual Differences: work ability and coping 1238.3 Measuring Stress 1248.4 Stress Management Interventions 1258.5 Evaluation of Interventions 1268.6 Individual- and Organisational-level Interventions 1288.7 Overall Conclusions 130

9 REFERENCES 131Appendix 1. Project Organisation 167


7■

Stress at work is a priority issue of the Eu-ropean Agency for Safety and Health atWork. As part of the Agency’s 1999 pro-gramme, an information project waslaunched in order to collect, evaluate andreview research data on work-relatedstress and its causes, and on interventionstudies.

The Institute of Work, Health and Organi-sations at the University of Nottingham,United Kingdom, was appointed to carryout this project within the framework ofthe Topic Centre on Research – Work andHealth. This report on “Research on Work-related Stress” has been prepared by Pro-fessor Tom Cox CBE, Dr. Amanda Griffithsand Mr. Eusebio Rial-González from thatInstitute.

A special consultation process was conduct-ed in the autumn of 1999 by sending thedraft manuscript to the members of the

Thematic Network Group on Research -Work and Health, to the European Com-mission, to the European social partnersand to other experts on the topic. The draftReport was also presented at a Joint Con-sensus Workshop organised by the Nation-al Institute of Occupational Health(Denmark) and the Institute of Work, Healthand Organisations (United Kingdom) inCopenhagen on 25th–26th October 1999.Following the consultation process, the finalreport was prepared and published.

The European Agency wishes to thank theauthors for their comprehensive work. TheAgency wishes to thank also the partici-pants in the Copenhagen workshop andall those individuals otherwise involved inthe review process.

May 2000

European Agency for Safety and Health atWork

F O R E W O R D


■8


9■

The European Agency for Safety andHealth at Work commissioned this StatusReport on stress at work within the frame-work of the Topic Centre on Research –Work and Health (TC/WH). The Reportconsiders early and contemporary scientif-ic studies on the nature of stress at work,on its effects on health and on the way inwhich such knowledge is being applied inattempts to manage this problem. TheTopic Centre on Good Practice – Stress atWork (TC/GP-ST) collects and evaluatesgood practice information on stress atwork both within the EU and beyond.Consequently, this Report deals with theresearch evidence regarding the assess-ment and management of stress at work:it does not review stress management inpractice. However, it discusses the concep-tual frameworks implied in the practice ofstress management at work and in currenthealth and safety legislation, focusing in

particular on the utility of the ‘control cy-cle’ and problem-solving approaches tothe management of stress at work.

I n t r o d u c t i o n

Over the past three decades, there hasbeen a growing belief in all sectors of em-ployment and in government that the ex-perience of stress at work has undesirableconsequences for the health and safety ofindividuals and for the health of their or-ganisations. This belief has been reflectedboth in public and media interest and in in-creasing concern voiced by the tradesunions, and professional and scientificbodies.

There are three basic questions that needto be answered:

(1) What is the nature of stress at work?

(2) Does work stress affect health andwell-being and, if so, how? and

(3) What are the implications of existingresearch for the management of work-related stress?

This Report addresses these questions af-ter having briefly examined the difficultiesinvolved in placing work stress in the con-text of other life stressors.

T h e E x t e n t o f t h e P r o b l e m

Determining the extent of stress-relatedhealth problems at work is not an easytask. Most countries routinely collect dataon ill-health retirements, work days lostdue to sickness, injury and disability, etc.However, such data are imprecise and notreliable in terms of describing trends dueto changes in, for example, the recording

E X E C U T I V E S U M M A R Y


■10

methods used. For this reason, they canonly be used as a basis for ‘educatedguesses’ in relation to the extent or cost ofoccupational stress. It is even more difficultto obtain valid, reliable and standardiseddata across the European Union’s 15 Mem-ber States. As the 1997 European Founda-tion report on European WorkingEnvironment in Figures suggested, “al-though some information sources exist,very little comparable quantitative occupa-tional health and safety data is available atEuropean level, at present.” (EuropeanFoundation, 1997).

The European Foundation’s 1996 WorkingConditions in the European Union revealedthat 29% of the workers questioned be-lieved that their work affected their health.The work-related health problems men-tioned most frequently are musculoskele-tal complaints (30%) and stress (28%).23% of respondents said they had beenabsent from work for work-related healthreasons during the previous 12 months.The average number of days’ absence perworker was 4 days per year, which repre-sents around 600 million working days lostper year across the EU.

Although there is obviously a need formore rigorous data collection mecha-nisms, it is clear that stress-related ill-health is a major cause for concern interms of its impact on both individuals’lives and the productivity of organisationsand countries. The research summarised inthis Report shows that, even within a lifeperspective, work-related stress is a signif-icant problem and represents a major chal-lenge to occupational health in Europe.

D e f i n i t i o n o f S t r e s s

The definition of stress is not simply aquestion of semantics –playing withwords– and it is important that there isagreement, at least in broad terms, on itsnature. A lack of such agreement wouldseriously hamper research into stress andthe subsequent development of effectivestress management strategies.

The simple equating of demand with stresshas been associated with the belief that acertain amount of stress is linked to maxi-mal performance and possibly goodhealth. Belief in optimal levels of stress hasbeen used, on occasions, to justify poormanagement practices. Given this, it is anunfortunate but popular misconceptionthat there is little consensus on the defini-tion of stress as a scientific concept or,worse, that stress is in some way undefin-able and unmeasurable. This belief belies alack of knowledge of the relevant scientif-ic literature.

It has been concluded in several differentreviews of the stress literature that thereare essentially three different, but overlap-ping, approaches to the definition andstudy of stress. The first approach concep-tualises occupational stress as an aversiveor noxious characteristic of the work envi-ronment, and, in related studies, treats itas an independent variable – the environ-mental cause of ill health. This has beentermed the ‘engineering approach’. Thesecond approach, on the other hand, de-fines stress in terms of the common physi-ological effects of a wide range of aversiveor noxious stimuli. It treats stress as a de-pendent variable – as a particular physio-


11■

logical response to a threatening or dam-aging environment. This has been termedthe ‘physiological approach’. The third ap-proach conceptualises work stress in termsof the dynamic interaction between theperson and their work environment. Thisfinal approach has been termed the ‘psy-chological approach’.

Two specific criticisms have been offeredof the first two approaches: the first em-pirical and the second conceptual. First,both engineering and physiological mod-els do not adequately account for the ex-isting data. For instance, they ignore themediation of strong cognitive as well as sit-uational (context) factors in the overallstress process. The second criticism is thatthe engineering and physiological modelsof stress are conceptually dated in thatthey are set within a relatively simple stim-ulus-response paradigm, and largelyignore individual differences of a psycho-logical nature and the perceptual and cog-nitive processes that might underpin.

These two approaches, therefore, treat theperson as a passive vehicle for translatingthe stimulus characteristics of the environ-ment into psychological and physiologicalresponse parameters. They largely ignorethe interactions between the person andtheir various environments, which are anessential part of systems-based approach-es to biology, behaviour and psychology.However, the third approach to the defini-tion and study of stress pays special atten-tion to environmental factors and, inparticular, to the psychosocial and organi-sational contexts to work stress. Stress iseither inferred from the existence of prob-lematic person-environment interactions

or measured in terms of the cognitiveprocesses and emotional reactions whichunderpin those interactions. This has beentermed the ‘psychological approach’.

The development of psychological modelshas been, to some extent, an attempt toovercome the criticisms levelled at the ear-lier approaches. There is now a consensusdeveloping around this approach to thedefinition of stress. For example, psycho-logical approaches to the definition ofstress are largely consistent with the Inter-national Labour Organization’s definitionof psychosocial hazards (InternationalLabour Organization, 1986: see later) andwith the definition of well-being recom-mended by the World Health Organization(1986)1. They are also consistent with thedeveloping literature on personal risk as-sessment (see, for example, Cox & Cox,1993; Cox, 1993; Cox & Griffiths, 1994,1996). These consistencies and overlapssuggest an increasing coherence in currentthinking within occupational health andsafety.

Variants of this psychological approachdominate contemporary stress theory, andamong them two distinct types can beidentified: the interactional and the trans-actional. The former focus on the structur-al features of the person’s interaction withtheir work environment, while the latterare more concerned with the psychologi-cal mechanisms underpinning that interac-

1 Well-being is a dynamic state of mind characterisedby reasonable harmony between a person’s abilities,needs, and expectations, and environmental demandsand opportunities (World Health Organization, 1986).The individual’s subjective assessment is the only validmeasure of well-being available (Levi, 1992).


■12

tion. Transactional models are primarilyconcerned with cognitive appraisal andcoping. In a sense they represent a devel-opment of the interactional models, andare largely consistent with them.

There is a growing consensus on the defi-nition of stress as a negative psychologicalstate with cognitive and emotional com-ponents, and on its effects on the health ofboth individual employees and their or-ganisations. Furthermore, there are nowtheories of stress which can be used to re-late the experience and effects of workstress to exposure to work hazards and tothe harmful effects on health that such ex-posure might cause. Applying such theo-ries to the understanding of stress at workallows an approach to the management ofwork stress through the application of thenotion of the control cycle. Such an ap-proach has proved effective elsewhere inhealth and safety. It offers a systematicproblem-solving system for continuous im-provement in relation to work stress. Thereare several distinct areas in which more re-search is required: some relate to the indi-vidual, but others relate to the design andmanagement of work and interventions toimprove the work environment.

I n d i v i d u a l D i f f e r e n c e s : w o r k a b i l i t ya n d c o p i n g

Coping is an important part of the overallstress process. However, it is perhaps theleast well understood despite many yearsof research. It has been suggested thatcoping has three main features. First, it is aprocess: it is what the person actuallythinks and does in a stressful encounter.Second, it is context-dependent: coping is

influenced by the particular encounter orappraisal that initiates it and by the re-sources available to manage that en-counter. Finally, coping as a process is andshould be defined ‘independent of out-come’; that is, independently of whether itwas successful or not. There have beentwo approaches to the study of coping:that which attempts to classify the differ-ent types of coping strategies and producea comprehensive taxonomy, and thatwhich considers coping as a problem-solv-ing process.

Most contemporary theories of stress allowfor individual differences in the experienceof stress, and in how and how well it iscoped with. Individual difference variableshave been investigated as either: (1) com-ponents of the appraisal process, or (2)moderators of the stress-health relation-ship. Hence, researchers have asked, for ex-ample, to what extent are particularworkers vulnerable to the experience ofstress, or, for example, to what extent does,say, ‘hardiness’ moderate the relationshipbetween job characteristics and workerhealth? This Report suggests that this dis-tinction between individual differences ascomponents of the appraisal process andmoderators of the stress-outcome relation-ship can be easily understood in terms oftransactional models of stress.

The experience of stress is partly depen-dent on the individual’s ability to cope withthe demands placed on them by theirwork, and on the way in which they sub-sequently cope with those demands, andrelates issues of control and support. Moreinformation is required on the nature,structure and effectiveness of individuals’


13■

abilities to meet work demands and tocope with any subsequent stress. The needfor more information on coping is widelyrecognised (see, for example, Dewe,2000), but relatively less attention hasbeen paid to the need better to under-stand the concept of work ability or com-petence, although this is being flagged inrelation to ageing research (e.g., Griffiths,1999a; Ilmarinen & Rantanen, 1999)

M e t h o d o l o g i c a l I s s u e s

The available evidence supports a psycho-logical approach to the definition of stress,and suggests that transactional models areamong the most adequate and useful ofthose currently available. Within thisframework, stress is defined as a psycho-logical state which is both part of and re-flects a wider process of interactionbetween the person and their (work) envi-ronment.

This process is based on a sequence of re-lationships between the objective workenvironment and the worker’s percep-tions, between those perceptions and theexperience of stress, and between that ex-perience, changes in behaviour and physi-ological function, and health. Thissequence provides a basis for measure-ment, but the different measures whichcan be derived from the sequence cannotbe easily or defensibly combined into a sin-gle stress index.

Logically the measurement of the stressstate must be based primarily on self-re-port measures which focus on the ap-praisal process and on the emotionalexperience of stress. Measures relating to

appraisal need to consider the worker’sperceptions of the demands on them,their ability to cope with those demands,their needs and the extent to which theyare fulfilled by work, the control they haveover work and the support they receive inrelation to work. Therefore, eliciting andmodelling the knowledge and perceptionsof employees is central to the assessmentand measurement process. Despite theirobvious centrality and importance, self-re-port measures of appraisal and the emo-tional experience of stress are, on theirown, insufficient. While their reliabilitycan be established in terms of their inter-nal structure or performance over timewithout reference to other data, their va-lidity cannot.

The validity of self-report data has beenquestioned in particular with regard to theissue of “negative affectivity” (NA), whichcan be defined as “a general personalitytrait reflecting individual differences innegative emotionality and self-concept,i.e., concentrating on negative aspects ofeverything and experiencing considerabledistress in all situations” (Watson & Clarke,1984). NA would affect not only workers’perception of their work environment, butalso their appraisal of their own psycho-logical health status or well-being, thusbecoming a confounding variable thatcould account for a large proportion of thecorrelations between perceived hazardsand perceived outcomes.

The research literature is still divided on theextent to which NA or common methodvariance distort the assessment of thestress-strain relationship. However, thereare ways in which the design of assess-


■14

ment instruments and procedures can con-tribute to ensuring that the data obtainedare of good quality. It is clear that an as-sessment relying solely on appraisal wouldrepresent very weak evidence, and wouldneed to be supported by data from otherdomains.

Triangulation of evidence overcomes thepotential problems of NA to some extent.The principle of triangulation holds that,to be secure, a potential psychosocial ororganisational hazard must be identifiedby cross-reference to at least three differ-ent types of evidence. The degree ofagreement between those differentpoints of view provides some indicationof the reliability of the data and, depend-ing on the measures used, their concur-rent validity.

Applying this principle would require datato be collected from at least three differentdomains. This can be achieved by consid-ering evidence relating to:

1. the objective and subjective an-tecedents of the person’s experience ofstress,

2. their self-report of stress, and

3. any changes in their behaviour, physiol-ogy or health status (which might becorrelated with [1] and/or [2]).

The influence of moderating factors, suchas individual and group differences mayalso be assessed. Confidence on the valid-ity of the data thus obtained is supportedby various studies which have shown thatthere is good convergence between self-report and supervisor- and subordinate-re-port. The use of any measure must besupported by data relating to its reliability

and validity, and its appropriateness andfairness in the situation in which it is beingused. The provision of such data wouldconform to good practice in both occupa-tional psychology and psychometrics (e.g.,Cox & Ferguson, 1994), but may also berequired if any subsequent decisions arechallenged in law.

W o r k H a z a r d s a n d S t r e s s

In line with both the scientific literatureand current legislation, this Report consid-ers the evidence relating to all work haz-ards. These can be broadly divided intophysical hazards, which include the biolog-ical, biomechanical, chemical and radio-logical, and the psychosocial hazards.Psychosocial hazards may be defined as“those aspects of work design and the or-ganisation and management of work, andtheir social and environmental contexts,which have the potential for causing psy-chological, social or physical harm”.

Exposure to physical and psychosocial haz-ards may affect psychological as well asphysical health. The evidence suggeststhat such effects on health may be medi-ated by, at least, two processes: a directphysical mechanism, and a psychologicalstress-mediated mechanism. These twomechanisms do not offer alternative expla-nations of the hazard-health association;in most hazardous situations both operateand interact to varying extents and in vari-ous ways.

The psychological aspects of work havebeen the subject of research since at leastthe 1950s. Initially psychologists concen-trated primarily on the obstacles to em-


15■

ployees’ adaptation and adjustment to thework environment, rather than on the po-tentially hazardous characteristics theworkplace itself may have for workers.However, with the emergence of psy-chosocial work-environment research andoccupational psychology in the 1960s, thefocus of interest has moved away from anindividual perspective and towards consid-ering the impact of certain aspects of thework environment on health. There is nowa large body of evidence that identifies acommon set of work characteristics as po-tentially hazardous (see Table 1).

Additionally, large scale socio-economicand technological changes in recent yearshave affected workplaces considerably.They are often collectively referred to as“the changing world of work”. This termencompasses a wide range of new pat-terns of work organisation at a variety oflevels, such as:• a growing number of older workers• teleworking and increased use of infor-

mation and communication technology(ICT) in the workplace

• downsizing, outsourcing, subcontract-ing and globalisation, with the associat-ed change in employment patterns

• demands for workers’ flexibility both interms of number and function or skills

• an increasing proportion of the popula-tion working in the service sector

• self-regulated work and teamwork

The research corpus is still developing inthese areas (e.g., see Rosenstock, 1997),but there is some preliminary evidence thateven changes which may be thought toenhance the work environment can pro-duce the opposite effect. For example,

Windel (1996) studied the introduction ofself-regulating team work in the office ofan electronics manufacturer. Althoughself-regulated work may be a source of in-creased self-efficacy and offer enhancedsocial support, Windel found that after 1year work demands had increased andwell-being decreased when compared tobaseline data. The data suggested that theincrease in social support brought aboutby self-regulating teams was not sufficientto counteract increased demands causedby the combination of a reduction in thenumber of staff and increases in manager-ial duties. Meta-analytical studies have alsoshown either mixed consequences (Bet-tenhausen, 1991; Windel & Zimolong,1997) or higher rates of absenteeism andstaff turnover (Cohen and Ledford, 1994)as a result of the implementation of teamwork or self-regulated work. It is clear thatchanges which have such a profound im-pact on the way organisations operatemay carry associated potential hazardsthat need to be monitored for their impacton health and well-being.

In summary, it is possible from the avail-able literature to explore the effects of themore tangible hazards of work on the ex-perience of stress and on health, and toidentify those psychosocial hazards whichpose a threat to employees. Most literaturereviews have identified the need for fur-ther research and development to trans-late this information into a form which canbe used in the auditing and analysis ofworkplaces and organisations. Such amodel, together with practical implemen-tation strategies, has been provided byCox et al. (2000).


■16

W o r k a n d H e a l t h

Over the past two decades, there has beenan increasing belief that the experience ofstress necessarily has undesirable conse-quences for health. It has become a com-mon assumption, if not a “culturaltruism”, that it is associated with the im-pairment of health. Despite this, the evi-dence is that the experience of stress doesnot necessarily have pathological seque-lae. Many of the person’s responses to thatexperience, both psychological and physi-ological, are comfortably within the body’snormal homeostatic limits and, while tax-ing the psychophysiological mechanismsinvolved, need not cause any lasting dis-turbance or damage.

However, it is also obvious that the nega-tive emotional experiences which are asso-ciated with the experience of stress detractboth from the general quality of life andfrom the person’s sense of well-being.Thus the experience of stress, while neces-sarily reducing that sense of well-being,does not inevitably contribute to the de-velopment of physical or psychological dis-order. For some, however, the experiencemay influence pathogenesis: stress may af-fect health. At the same time, however, astate of ill health can both act as a signifi-cant source of stress, and may also sensi-tise the person to other sources of stress byreducing their ability to cope. Within theselimits, the common assumption of a rela-tionship between the experience of stressand poor health appears justified.

The Report presents a brief overview of thebroad range of health and health-relatedeffects which have been variously associat-

ed with the experience of stress. It focuseson changes in health and health-related be-haviours and physiological function, whichtogether may account for any linkage be-tween that experience and psychologicaland physical health. In summary, the experi-ence of stress can alter the way the personfeels, thinks and behaves, and can also pro-duce changes in their physiological func-tion. Many of these changes simplyrepresent, in themselves, a modest dysfunc-tion and possibly some associated discom-fort. Many are easily reversible although stilldamaging to the quality of life at the time.

However, for some workers and undersome circumstances, they might translateinto poor performance at work, into otherpsychological and social problems and intopoor physical health. Nevertheless, theoverall strength of the relationship be-tween the experience of stress, and its an-tecedents, on one hand, and health, onthe other, is consistent but moderate.There is evidence that the experience ofstress at work is associated with changes inbehaviour and physiological function, bothof which may be detrimental to employ-ees’ health. Much is known about the pos-sible mechanisms underpinning sucheffects, and particular attention has beenpaid to pathologies possibly associatedwith impaired immune activity as well asthose more traditionally linked to stress,such as ulcers, coronary heart disease andrheumatoid arthritis.

R e s e a r c h i n t o t h e A s s e s s m e n t a n dM a n a g e m e n t o f W o r k - r e l a t e d S t r e s s

There are numerous reviews of researchinto psychosocial hazards and stress and a


17■

large number of papers dealing with thestressors in almost every conceivable worksetting and occupation. However, researchinto the nature and effects of a hazard isnot the same as assessment of the associ-ated risk. Indeed, most published studieswould provide very little data that could beused for a risk assessment. Many “stresssurveys” tend to identify only hazards oronly outcomes, whereas the object of arisk assessment is to establish an associa-tion between hazards and health out-comes, and to evaluate the risk to healthfrom exposure to a hazard.

An almost unavoidable corollary of thepaucity of adequate risk assessments isthat most “stress management” interven-tions target the individual rather than theorganisation (the former is usually seen ascheaper and less cumbersome), are oftenoff-the-shelf designs, and are entirely di-vorced from the process of diagnosis ofthe problems - if diagnosis takes place atall.

A different type of approach is thereforerequired in order to carry out risk assess-ments which can then inform the designof interventions - in other words, a strate-gy that actually asks the question beforegiving the answer. Such a strategy has al-ready been suggested for the manage-ment of physical hazards at both EU andnational level: the control cycle, which hasbeen defined as “the systematic process bywhich hazards are identified, risks analysedand managed, and workers protected”. Asa systematic and comprehensive approachto assessing the risks within the work envi-ronment, the control cycle satisfies currentlegal requirements. However, it is still nec-

essary to evaluate whether it represents ascientifically valid and reliable strategy toassess psychosocial hazards. The Reportexamines the advantages and disadvan-tages of application of the control cycle(borrowed from the field of physical haz-ard control) to the assessment and man-agement of work-related stress. TheReport concludes that this model is veryhelpful as an analogy and represents a use-ful strategy for the assessment of psy-chosocial hazards at work. However, thereare a number of issues to bear in mind:

a. the operationalisation of definitions ofhazard,

b. the identification of adequate indicesof harm that can also be reliably moni-tored,

c. satisfactory proof of a causal relation-ship, and

d. problems of measurement of the workenvironment.

L i m i t a t i o n s o f C o n t e m p o r a r yR e s e a r c h i n t o t h e M a n a g e m e n t o fW o r k - r e l a t e d S t r e s s

A review of the scientific literature sug-gests that there are a number of problemswith research into the management ofwork-related stress.

1. Too narrow a view has often been tak-en of what constitutes stress manage-ment and there has been too strong afocus on ‘caring for or curing’ the indi-vidual.

2. Much of what has been offered, evenin this narrow respect, either has aweak theoretical base or has been de-veloped from theory outside occupa-tional stress research.


■18

3. There has been a tendency to treat theapplication of stress managementstrategies as a self-contained actionand to divorce that application fromany preceding process of problem diag-nosis.

4. Stress management strategies often fo-cus on single types of intervention andrarely are multiple strategies offered.

5. Such interventions are seldom offeredfor evaluation beyond participants’ im-mediate reactions or measures of facevalidity.

There are three common purposes forevaluations of stress management pro-grammes. The first is to ask whether theprogramme is effective; specifical lywhether the programme objectives are be-ing met. A second purpose is to determinethe efficiency or comparative effectivenessof two or more programmes or methodswithin a programme. The third purpose isto assess the cost-benefit or the cost-ef-fectiveness of the programme.

Evaluation data on stress managementprogrammes are relatively rare. There arerelatively fewer cost-benefit and cost-ef-fectiveness studies compared to studies onthe overall effectiveness of programmes orthe relative effectiveness of their compo-nent parts. What there is suggests thatstress management programmes may beeffective in improving the quality of work-ing life of workers and their immediatepsychological health, albeit self-reported.The evidence relating such interventions toimprovements in physical health is weaker,largely for methodological reasons. Therehave been several authoritative reviews oforganisational and personal stress man-

agement programmes in the last ten yearsreaching broadly similar conclusions.

It must be concluded that “the jury is stillout” on stress management training:whilst it seems logical that such interven-tions should promote employee health,there are not yet sufficient data to be con-fident that they do. The evidence for em-ployee assistance programmes, particularlythose broadly conceived to include healthpromotion in the workplace, may be moreencouraging, although that which relatesto counselling alone is weak. The provisionof counselling is largely designed to assistemployees who are already suffering aproblem, and is, in that sense, post hoc.

Stressor reduction / hazard control is, forseveral reasons, the most promising areafor interventions, although again, there isnot yet sufficient information to be confi-dent about the nature and extent of theireffectiveness. To date, such conclusionsare based more on moral and strategic rea-soning than on empirical data, althoughthe data that do exist are supportive. Whatcan be firmly concluded, however, is thatthere is still a need for further and moreadequate evaluation studies.

Unfortunately, there are very few well de-signed and evaluated such interventionsavailable in the literature to date. Nonethe-less, Murphy et al. (1992) conclude that“job redesign and organisational changeremain the preferred approaches to stressmanagement because they focus on re-ducing or eliminating the sources of theproblem in the work environment”. How-ever, they also point out that such ap-proaches require a detailed audit of work


19■

stressors and a knowledge of the dynam-ics of organisational change if unwelcomeoutcomes are to be minimised. Further,such interventions can be expensive andmore difficult and disruptive to design, im-plement and evaluate – factors which maymake them less popular alternatives tosecondary (reaction) and tertiary (treat-ment) interventions.

Nonetheless, Landy (1992) has sum-marised a number of possible interven-tions focused on the design of the workenvironment, and Murphy (1988) notedthat, given the varieties of work stressorsthat have been identified, many othertypes of action relating to organisationaland work development should be effectivein reducing work stress. Van der Hek &Plomp (1997) also concluded that “there issome evidence that organization-wide ap-proaches show the best results on individ-ual, individual-organizational interfaceand organizational parameters [outcomemeasures]; these comprehensive pro-grammes have a strong impact on the en-tire organization, and require the fullsupport of management”.

The emerging evidence is strong enoughfor the United States’ National Institute forOccupational Safety and Health (NIOSH) tohave identified “the organization of work”as one of the national occupational safetyand health priority areas (Rosenstock,1997). As part of their National Occupa-tional Research Agenda (NORA), NIOSH in-tend to focus research on issues such asthe impact of work organisation on overallhealth, the identification of healthy organ-isation characteristics and the develop-ment of intervention strategies.

The evaluation literature is inconclusive asto what are the exact mechanisms bywhich interventions, and particularly thosefocused on the individual, might affecthealth. Often, where different types of in-dividually focused interventions have beencompared, there is no evidence that anyone or any combination is better than anyother. This indicates that there may be ageneral, non-specific effect of intervening:the fact of an intervention may be benefi-cial, rather than its exact content. Inter-views with managers responsible forintroducing such interventions suggestthat they are aware of such effects (see,for example, Cox et al., 1988). It is there-fore possible that at least part of the ef-fects of stress management programmesis due to the way they alter workers’ per-ceptions of, and attitudes to, their organi-sations, and hence organisational culture.It was argued earlier that poor organisa-tional culture might be associated with anincreased experience of stress, while agood organisational culture might weakenor “buffer” the effects of stress on health.A defining factor for organisational cul-ture is the size of the enterprise, and thisshould be borne in mind when consider-ing intervention and evaluation issues, to-gether with the wider context in terms ofthe socio-economic environment in theMember States.

Overall, the evidence on the effectivenessof stress management interventions re-viewed in this Status Report is promising.The available data, although sparse, sug-gest that interventions, especially at theorganisational level (e.g., Ganster et al.,1982; Shinn et al., 1984; Dollard & Wine-field, 1996; Kompier et al., 1998), are ben-


■20

eficial to both individual and organisation-al health and should be investigated – andevaluated – further.

In summary, there is available scientific ev-idence to support the following:• work-related stress is a current and fu-

ture health and safety issue;• work-related stress can be dealt with in

the same logical and systematic way asother health and safety issues;

• the management of stress at work couldbe based on the adaptation and appli-cation of a control cycle approach suchas that made explicit in contemporarymodels of risk management;

• there are already practical examples ofthis approach in several countries of theEuropean Union.

The final comment concerns the maturityof stress research as an area of applied sci-ence. Two things must be apparent to the

informed reader of this Report. First, thereis a wealth of scientific data on workstress, its causes and effects, and on someof the mechanisms underpinning the rela-tionships among these. More general re-search is not needed. What is required isan answer to the outstanding method-ological questions, and to more specificquestions about particular aspects of theoverall stress process and its underpinningmechanisms. Second, although this wealthof scientific data exists, it still needs to betranslated into practice, and the effective-ness of this practice evaluated. This is an-other set of needs, and one that will onlybe settled outside the laboratory andthrough the development of consensusand eventually common practice.

While stress at work will remain a majorchallenge to occupational health, our abil-ity to understand and manage that chal-lenge is improving. The future looks bright.


21■

1.T E R M S O F R E F E R E N C E

As part of its 1999 Work Programme, theEuropean Agency for Safety and Health atWork commissioned this Report within theframework of the Topic Centre on Re-search – Work and Health (TC/WH) (seeAppendix 1). Work-related stress is treatedas an occupational health issue and cur-rent thinking and legislation in health andsafety are used to frame this Report. Itsprime objective is to provide an up-to-dateoverview of the scientific literature relating

to research into the nature and effects ofwork-related stress and of stress manage-ment interventions.

It is not possible, within the terms of refer-ence of this Report, to cite and appraise allthe published literature because of both itsvastness and its increasing specificity anddetail (see for example, Danna & Griffin,1999; Cartwright & Cooper, 1996; Cox,1993; Borg, 1990; Hiebert & Farber, 1984;Kasl, 1990). Kasl (1992) has suggestedthat many reviews in this area are an at-tempt either to “paint the big picture” orto present a detailed evaluation of a spe-cific hypothesis. The former can suffer be-cause they are too superficial or tooselective in favour of one over-archingview, while the latter can suffer simply be-cause they fail to place the hypotheses ofconcern in their wider context and, thus,fail to make an evaluation on the basis ofthe whole picture. Furthermore, much ofwhat is available for review has beendeemed to be methodologically weak. ForKasl (1992), the main methodologicalproblem is that the available evidence is, inlarge part, based on cross-sectional studiesin which the key variables are measuredand linked only in terms of self-report.While it would be unwise to reject out-of-hand all such studies, the methodologicalsophistication necessary for their properdesign, analysis and interpretation is oftenalso lacking. A second problem is thatmuch of what is published is redundant inthat it simply demonstrates well-estab-lished theories and would-be facts (Cox,1993). In many cases there is no significantgain in knowledge.

RE

SE

AR

CH


■22

impossibility of covering all the valuable re-search published in languages other thanEnglish within the constraints of time andresources. However, English has becomethe de facto lingua franca for scientificpublication in Europe and, as a result, it isunlikely that this Report has missed anyfundamental contributions.

This Report is, therefore, selective in theevidence that it draws on. At the sametime, the Report is consistent with earlierguidance on the control and monitoring ofpsychosocial and organisational hazardsprepared by the authors for the WorldHealth Organization (European Region)and published in its Occasional Series inOccupational Health no. 5 (Cox & Cox,1993), for the Health & Safety Executive ofGreat Britain (Cox, 1993; Cox et al., 2000),and for the Loss Prevention Council (UK)(Griffiths et al., 1998).

The European Agency’s Topic Centre onGood Practice – Stress at Work (TC/GP-ST)collects, evaluates and disseminates exist-ing good practice information about stressat work across the EU and beyond. Conse-quently, after reviewing the research intothe nature, causes and effects of work-re-lated stress, this Report deals briefly withthe research evidence regarding the as-sessment and management of stress atwork, but it does not examine actual stressmanagement practice in detail.

There are also some important topics thatcannot be explored in detail because ofspace constraints. For example, socio-eco-nomic and cultural factors –such as in-equalities in health and health provision,particularly in relation to ageing and socio-economic status, new working patternsand the “global economy”, cultural differ-ences in attitudes towards work andhealth, etc.– are known to have an impacton work-related stress. Stress is also relat-ed to burnout, poor occupational safetyand the reporting of work-related upperlimb disorders. Although these are signifi-cant issues that should be borne in mindwhen considering the causes and conse-quences of stress at work, this Report canonly deal with them briefly or indirectly inthe space available (for instance, see sec-tions 5.1, 5.2.1 and 5.4). Readers aretherefore encouraged to consult othersources included in the Report’s Referencessection (e.g., European Agency, 1999).

Therefore, this Report focuses on thatwhich is:• relevant to its stated objective • relevant to the treatment of work-relat-

ed stress as an occupational health issue• better known• more, rather than less, adequate both

methodologically and theoretically

By necessity, the Report also focuses pri-marily on the literature published in Eng-l ish. The authors acknowledge the


23■

2.I N T R O D U C T I O N

Over the past three decades there hasbeen a growing belief in all sectors of em-ployment and in government that the ex-perience of stress at work has undesirableconsequences for the health and safety ofindividuals and for the health of their or-ganisations. This belief has been reflectedboth in public and media interest and in in-creasing concern voiced by the tradesunions, and professional and scientificbodies.

There are three basic questions that re-quire answering: 1. What is the nature of occupational

stress?2. Does work stress affect health and

well-being and, if so, how? 3. What are the implications of existing

research for the management of work-related stress?

This Report addresses these questions af-ter having briefly examined the difficultiesinvolved in placing work stress in the con-text of other life stressors.

RE

SE

AR

CH


■24

has been made in determining the relativeimportance of different types of events.One particular example is considered here.

Dohrenwend et al. (1988) have describedthe careful development (and strengthsand weaknesses) of the PERI2 Life EventsScale. A list of 102 objectively verifiable lifeevents was constructed from previousstudies in New York. These events wereclassified according to 11 life domains:school, work, love and marriage, havingchildren, family, residence, crime and legalmatters, finances, social activities, healthand miscellaneous. As in other studies (seeDohrenwend & Dohrenwend, 1974), sub-jects were asked to rate events againstmarriage, which was given an arbitrary rat-ing of 500. Subjects were grouped accord-ing to a number of criteria such as age,sex, and ethnic background, and meansub-group ratings were calculated for eachevent. This avoided giving undue weightto sub-groups over-represented in theoverall sample. However, the events werealso scored according to their mean rank-ings: this gave equal weight to all subjectsregardless of sub group. Of the 102 lifeevents, 21 related to work. The highestranked work event was suffered businessloss or failure with a mean rating of 510.Demoted and promoted at work rated 379and 374 respectively. The lowest rankedwork event was changed job for onewhich was no better or worse than lastone (251). As far as non-work events wereconcerned, the highest ranked event over-all was child died (with a rating of 1036),with divorce at 633, married at 500 (the

2.1T H E N A T U R E O F W O R K -

R E L A T E D S T R E S S I N A L I F E

P E R S P E C T I V E

There is evidence to suggest that work isonly one of a number of possible areas oraspects of life that can give rise to the ex-perience of stress and ill-health (e.g., Gold-berg & Novack, 1992; Surtees &Wainwright, 1998). Largely following onfrom the work of Selye (1956), there hasbeen an assumption that discrete, time-limited ‘life events’ requiring change oradaptation are associated with the experi-ence of stress and may contribute to awide range of disorders. Many attemptshave been made to identify and scale suchstressful life events (see, for example,Holmes & Rahe, 1967; Dohrenwend &Dohrenwend, 1974; Dohrenwend et al.,1988; Fisher, 1996). While psychometricresearch into the nature and impact ofstressful life events is not without method-ological problems (see, for example, Sara-son et al. , 1975; Perkins, 1988;Dohrenwend et al., 1988), some progress 2 PERI: Psychiatric Epidemiology Research Interview


25■

anchor event) and the lowest, acquiredpet, at 163. These data suggest that work-related life events are not trivial experi-ences, and are among those which havethe greatest perceived impact. This conclu-sion is supported by a study in the UnitedKingdom which asked a sample of maleand female employees in the East Mid-lands of England to identify that aspect (ordomain) of their lives which presentedthem with the greatest problems andstress. Work was cited as the major sourceof problems and stress for 54% of respon-dents, while another 12% cited the work-home interface (Cox et al., 1981).

However, it should be noted that wherelife event scales have included workevents, the designers have been con-cerned only with discrete, ‘acute’ work-re-lated events (such as being promoted ordemoted). As will be argued in later sec-tions of this Report, it is now widelythought that the primary stressors facingmost employees in the course of theirworking life are chronic rather than acuteand are rarely mentioned in life eventscales. Some studies have also suggestedthat rankings of life events are context-de-pendent and can vary between differentcountries (Rahe, 1969) and between urbanand rural communities (Abel et al., 1987).Thus, although at first sight life eventscales may seem to answer the question‘How important are work stressors?’, infact, they do not.

It is likely that there are interactions be-tween stressors, both acute and chronic,which do not respect the boundary be-tween work and non-work domains. In-deed, evidence does exist to suggest that

work stress can ‘spill over’ to home life(Bacharach et al., 1991; Burke, 1986), andvice versa (Quick et al., 1992b), althougheffects may vary considerably (Kanter,1977). The erroneous belief that work andnon-work activities are unrelated in theirpsychological, physiological and health ef-fects has been described as the ‘myth ofseparate worlds’ by Kanter (1977).

While it is nonsensical to attempt an exactdetermination of the relative importanceof work and non-work stressors, becausethey are not independent in their effects, itis sensible to explore that interaction andthe carry-over from one domain to theother. Although such interaction effectsexist, they are not always obvious. Whenan acute stressful life event occurs in workor outside of work (such as the death of aloved one, or a serious injury), the initialimpact of carry-over effects is often readi-ly obvious to family, friends and colleaguesor co-workers. However, when the effectsof life stressors are more subtle and longlasting, carry-over effects are less frequent-ly recognised and can be underestimated.Similarly, while the chronic experience ofwork stress may exert deleterious effectson family relationships, these may some-times go undetected (see Gutek et al.,1988; Repetti, 1987; Repetti & Crosby,1984; Voydanoff & Kelly, 1984). A surveyby the Canadian Mental Health Associa-tion (1984) found that 56% of respon-dents felt ‘some’ or ‘a great deal of’interference between their jobs and homelives. Of particular concern were the‘amount of time that the job demanded’and the ‘irregularity of working hours’ (in-cluding shift work). The interference af-


■26

fected family routines and events, childrearing and household responsibilities,made employees moody at home and con-flicted with leisure activities and social life.

The focus of this Report on work stressmay suggest that work has only a negativeeffect on health: this is not the case. Thereis evidence that, under some circum-stances, work may have positive healthbenefits, promoting psychological well-be-ing (Baruch & Barnett, 1987) and physicalhealth (Repetti et al., 1989). Unemploy-ment and retirement from work are associ-ated with excess risk of psychological illhealth (for example, Lennon, 1999; Cobb& Kasl, 1977; Feather, 1990; Jackson &Warr, 1984; Kasl, 1980b; Warr, 1982,1983, 1987). They may also be associated

with increased risk of cardiovascular dis-ease but the evidence here is, at best,equivocal (Kasl & Cobb, 1980). At thesame time, specific work characteristicsmay also be beneficial to health, in partic-ular, energy expenditure (Fletcher, 1988).Studies by Paffenbarger et al. (1977, 1984)have suggested that high-energy expendi-ture at work may be associated with re-duced risks of fatal heart attacks.

The definition of work stress and its mea-surement are central to the question of itsimportance and the determination of car-ry-over effects – positive or negative. Thefollowing sections review both early andmore contemporary theories of stress andexplore their implications for measure-ment.


27■

relation to the extent or cost of occupa-tional stress. It is even more difficult to ob-tain valid, reliable and standardised dataacross the European Union’s 15 MemberStates. As the 1997 European Foundationreport European Working Environment inFigures suggested, “although some infor-mation sources exist, very little comparablequantitative occupational health and safe-ty data is available at European level, atpresent.” (European Foundation, 1997)

The European Foundation’s 1996 WorkingConditions in the European Union revealedthat 29% of the workers questioned be-lieved that their work affected their health.The work-related health problems men-tioned most frequently are musculoskele-tal complaints (30%) and stress (28%).23% of respondents said they had beenabsent from work for work-related healthreasons during the previous 12 months.The average number of days’ absence perworker was 4 days per year, which repre-sents around 600 million working days lostper year across the EU.

Occupational diseases continue to givecause for concern across the EuropeanUnion. Figure 1, for example, shows that–despite recent decreases– the number ofoccupational diseases reported in Ger-many grew dramatically during the 1990sand remains at a very high level (Bun-desministerium für Arbeit und Sozialord-nung, 1999).

To take the United Kingdom as another ex-ample, it has been suggested that up-wards of 40 million working days are losteach year in the UK due to stress-relateddisorders (Kearns, 1986; Health & Safety

2.2T H E E X T E N T O F

T H E P R O B L E M

Determining the extent of stress-relatedhealth problems at work is not an easytask. Most countries routinely collect dataon ill-health retirements, work days lostdue to sickness, injury and disability, etc.For example, between 1981-1994 theNetherlands recorded an increase from21% to 30% in the percentage of workerswho received a disability pension becauseof stress-related disorders (ICD-9, 309, ad-justment disorder), and “the number whoreturned to work in the diagnosis group islower than in any other group” (Van derHek & Plomp, 1997).

However, such data are imprecise and notreliable in terms of describing trends dueto changes in, for example, the recordingmethods used (see, Marmot & Madge,1987; Fletcher, 1988; Jenkins, 1992, Grif-fiths, 1998). For this reason, they can onlybe used as a basis for ‘educated guesses’ in


■28

Executive, 1990b; Jones et al., 1998). In1994, the Health & Safety Executive ofGreat Britain published estimates (basedon 1990 data) of the total cost to employ-ers, the economy and society of work acci-dents and work-related ill health (Davies &Teasdale, 1994). The study attempted toquantify costs to all affected parties in-cluding employers (damage, lost output,costs of covering for sick absence), themedical services, the social security and in-surance systems, as well as the costs to thevictims of accidents and ill health, includ-ing “an amount to reflect the pain, griefand suffering involved”. The study foundthat the cost of work accidents and work-related ill health to employers in the UK in1990 was between £4.5 billion and £9 bil-lion (6.84 – 13.7 billion euro approximate-ly). Costs to victims and their families wereabout £4.5 billion. The total cost to theeconomy was between £6 billion and £12

billion (9.12 – 18.24 billion euro, about 1-2% of national output). Adding the sumfor pain, grief and suffering yields a totalcost to society of between £11 billion and£16 billion (16.72 – 24.32 billion euro).The framework can also be used to derivecost estimates for specific diseases (e.g.the Trades Union Congress (TUC) have es-timated the cost of Repetitive Strain Injuryto be £1 billion per year).

More recent figures released by the UK’sConfederation of British Industry (1999)indicate that 200 million days were lostthrough sickness absence in 1998, an av-erage of 8.5 days per employee. This rep-resents a loss of 3.7% of working time.Absence from work cost British business£10.2 billion in 1998 (approximately 15.5billion euro), an average cost of £426 perworker (approximately 647 euro). The sur-vey shows that minor illness is the biggestcause of absence for manual and non-

F i g u r e 1 : R e p o r t e d o c c u p a t i o n a l d i s e a s e s i n G e r m a n y ( 1 9 6 0 - 1 9 9 7 )

Reported occupational diseases in Germany 1960 to 1997(source: Occupational Accident Prevention Report 1997)

120000

100000

80000

60000

40000

20000

0

1960 1965 1970 1975 1980 1985 1990 1995

rep

ort

ed o

ccu

pat

ion

al d

isea

ses insured parties in 1,000

reported OD


29■

manual workers, with serious illness andhome and family responsibilities also im-portant for manual workers. For non-man-ual workers, workplace stress was felt tobe the second highest contributor to ab-sence, second only to minor illness.

The Health & Safety Executive of GreatBritain has estimated that at least half of alllost days are related to work stress (Coop-er et al., 1996). Furthermore, Kearns(1986) has suggested that up to 60% of allwork absence is caused by stress-relateddisorders, while Cooper & Davidson (1982)have reported that 71% of their sample ofmanagers in the United Kingdom felt thattheir psychological health problems wererelated to stress at work.

More adequate data exist on the relative ar-chitecture of stress-related ill health fromgeneral population surveys and from small-er scale studies of defined occupationalpopulations (see, for example, Colligan etal., 1977; Eaton et al., 1990; Jones et al.,1998). In their questionnaire-based surveyof the working population, Jones et al.(1998) found that 26.6% of the respon-dents reported suffering from work-relatedstress, depression or anxiety, or a physicalcondition which they ascribed to work-re-lated stress. The authors estimated that19.5 million working days3 were lost inGreat Britain due to work-related illness, ofwhich 11 million were due to muscu-loskeletal disorders, and 5 million to stress.However, such figures must be treated withcaution, since they rely mostly on uncorrob-orated self-report (Thomson et al., 1998).Other figures (such as the number of earlyretirements on the grounds of ill-health)which could help provide a general picture

in an oblique way have to be interpretedwith similar caution (Griffiths, 1998).

From an international perspective, it hasbeen estimated that approximately 550million working days are lost each year inthe US due to absenteeism (Harris et al.,1985), of which 54% are thought to bestress-related (Elkin & Rosch, 1990). Pro-jections from the National Health InterviewSurvey, suggested that 11 million workersin the United States could report ‘healthendangering’ levels of stress at work(Shilling & Brackbill, 1987). Only loud noisewas reported to be a more prevalent work-place hazard. Stress at work has becomeone of the main topics for the emergentdiscipline of occupational health psycholo-gy both in the United States (e.g., Quick etal., 1997) and Europe.

In Australia, the Federal Assistant Ministerfor Industrial Relations estimated the costof occupational stress to be around A$30million4 in 1994. The rising costs of work-related stress are illustrated by a recentstudy of 126 call centres (Deloitte &Touche, 1999), which revealed that impactof staff turnover and stress on call centreagents is costing organisations that usecall centres to conduct business over thetelephone a total of A$90 million a year.They found that stress-related absen-teeism costs $150 per agent per year - atotal cost of A$7.5 million per annum (ap-proximately 4.54 million euro).

3 Days lost per worker were defined as “number ofdays lost per person who has worked in the last 12months, including people without a work-related ill-ness”.

4 The Australian, June 17, 1994


■30

2.3S U M M A R Y

In a survey of the statutory bodies in all theEU Member States carried out by the Euro-pean Agency for Safety and Health atWork, most of them identified stress andrelated psychosocial issues as a current andfuture priority (European Agency, 1998).Although there is a need for more rigorousdata collection mechanisms, as identifiedby several bodies (e.g., European Founda-tion, 1997), it is clear that stress-related ill-health is a major cause for concern interms of its impact on both individuals’lives and the productivity of organisationsand countries. The research summarised inthis section shows that, even within a lifeperspective, work-related stress is a signif-icant problem and represents a major chal-lenge to occupational health in Europe.


31■

3.D E F I N I N G S T R E S S

The definition of stress is not simply aquestion of semantics –playing withwords– and it is important that there isagreement, at least in broad terms, on itsnature. A lack of such agreement wouldseriously hamper research into stress andthe subsequent development of effectivestress management strategies. Given this,it is an unfortunate but popular miscon-ception that there is little consensus on thedefinition of stress as a scientific concept

or, worse, that stress is in some way unde-finable and unmeasurable. This belief be-lies a lack of knowledge of the relevantscientific literature.

RE

SE

AR

CH


■32

third approach conceptualises work stressin terms of the dynamic interaction be-tween the person and their work environ-ment. When studied, stress is eitherinferred from the existence of problematicperson-environment interactions or mea-sured in terms of the cognitive processesand emotional reactions which underpinthose interactions. This final approach hasbeen termed the ‘psychological approach’.The engineering and physiological ap-proaches are obvious among the earliertheories of stress, while the more psycho-logical approaches characterise contempo-rary stress theory.

3 . 1 . 1 E n g i n e e r i n g A p p r o a c h

The engineering approach has treatedstress as a stimulus characteristic of theperson’s environment, usually conceived interms of the load or level of demandplaced on the individual, or some aversive(threatening) or noxious element of thatenvironment (Cox, 1978, 1990; Cox &Mackay, 1981; Fletcher, 1988). Occupa-tional stress is treated as a property of thework environment, and usually as an ob-jectively measurable aspect of that envi-ronment. In 1947, Symonds wrote, inrelation to psychological disorders in theRoyal Air Force flying personnel, that“stress is that which happens to the man,not that which happens in him; it is a set ofcauses not a set of symptoms.” Somewhatlater, Spielberger (1976) argued, in thesame vein, that the term stress should re-fer to the objective characteristics of situa-tions. According to this approach, stresswas said to produce a strain reactionwhich although often reversible could, on

3.1C O N C E P T U A L I S A T I O N A N D

F R A M E W O R K S

It has been concluded in several differentreviews of the scientific literature on stressthat there are essentially three different,but overlapping, approaches to the defini-tion and study of stress (Lazarus, 1966;Appley & Trumbull, 1967; Cox, 1978,1990; Cox & Mackay, 1981; Fletcher,1988, Cox, 1993). The first approach con-ceptualises occupational stress as an aver-sive or noxious characteristic of the workenvironment, and, in related studies, treatsit as an independent variable –the environ-mental cause of ill health. This has beentermed the ‘engineering approach’. Thesecond approach, on the other hand, de-fines stress in terms of the common physi-ological effects of a wide range of aversiveor noxious stimuli. It treats stress as a de-pendent variable –as a particular physio-logical response to a threatening ordamaging environment. This has beentermed the ‘physiological approach’. The


33■

occasions, prove to be irreversible anddamaging (Cox & Mackay, 1981; Suther-land & Cooper, 1990). The concept of astress threshold grew out of this way ofthinking and individual differences in thisthreshold have been used to account fordifferences in stress resistance and vulner-ability.

3 . 1 . 2 P h y s i o l o g i c a l A p p r o a c h

The physiological approach to the defini-tion and study of stress received its initialimpetus from the work of Selye (1950,1956). He defined stress as “a state mani-fested by a specific syndrome which con-sists of all the non-specific changes withinthe biologic system” that occur when chal-lenged by aversive or noxious stimuli.Stress is treated as a generalised and non-specific physiological response syndrome.For many years, the stress response waslargely conceived of in terms of the activa-tion of two neuroendocrine systems, theanterior pituitary-adrenal cortical systemand the sympathetic-adrenal medullarysystem (Cox & Cox, 1985; Cox et al.,1983). The psychophysiology of stress isdiscussed in more detail in section 6.3.1.Selye (1950, 1956) argued that the physi-ological response was triphasic in natureinvolving an initial alarm stage (sympathet-ic-adrenal medullary activation) followedby a stage of resistance (adrenal corticalactivation) giving way, under some circum-stances, to a final stage of exhaustion(terminal reactivation of the sympathetic-adrenal medullary system). Repeated, in-tense or prolonged elicitation of thisphysiological response, it has been sug-gested, increases the wear and tear on the

body, and contributes to what Selye (1956)has called the ‘diseases of adaptation’.This apparently paradoxical term arisesfrom the contrast between the immediateand short-term advantages bestowed byphysiological response to stress (energymobilisation for an active behavioural re-sponse) to the long-term disadvantages(increased risk of certain ‘stress related’diseases).

Scheuch (1996) considers stress as one ofthe psychophysiological activities of hu-man beings as they attempt to adapt tochanges in the internal and external mi-lieux. This activity relates to the quantityand quality of the relationship betweendemands and individual somatic, psycho-logical and social capacities or resources ina specific material and social environment.Stress is understood by Scheuch as a reac-tive activity to a disturbed homeostaticstate of organic functions, psychologicalfunctions and/or in the interaction be-tween the human being and his or her so-cial environment. The adaptation followsthe principles of economisation of func-tion, the principle of minimisation of ef-fort, and the principle of well-being. Stressitself is the expression of a disorder ofthese principles (Scheuch, 1990, 1996).

C r i t i c i s m s o f E n g i n e e r i n g &P h y s i o l o g i c a l A p p r o a c h e s

Two specific criticisms have been offeredof these two approaches: the first empiri-cal and the second conceptual.

First, engineering and physiological modelsdo not adequately account for the existingdata. In relation to the engineering model,


■34

consider the effects of noise on perfor-mance and comfort. The effects of noiseon task performance are not a simple func-tion of its loudness or frequency but aresubject both to its nature and to individualdifferences and context effects (see, for ex-ample, Cox, 1978; Flanagan et al. 1998;Ahasan et al. 1999). Noise levels which arenormally disruptive may help maintain taskperformance when subjects are tired or fa-tigued (Broadbent, 1971), while evenhigher levels of music may be freely chosenin social and leisure situations.

Scott & Howard (1970) wrote: “certainstimuli, by virtue of their unique meaningto particular individuals, may prove prob-lems only to them; other stimuli, by virtueof their commonly shared meaning, arelikely to prove problems to a larger numberof persons.” This statement implies themediation of strong cognitive as well as sit-uational (context) factors in the overallstress process (see below). This point hasbeen forcefully made by Douglas (1992)with respect to the perception of risks (andhazards). Such perceptions and related be-haviours, she maintains, are not adequate-ly explained by the natural science ofobjective risk and are strongly determinedby group and cultural biases.

The simple equating of demand with stresshas been associated with the belief that acertain amount of stress is linked to maxi-mal performance (Welford, 1973) and pos-sibly good health. Belief in optimal levelsof stress has been used, on occasions, tojustify poor management practices.

The physiological model is equally open tocriticism. Both the non-specificity and the

time course of the physiological response toaversive and noxious stimuli have beenshown to be different from that describedby Selye (1950, 1956) and required by themodel (see Mason, 1968, 1971). Mason(1971), for example, has shown that somenoxious physical stimuli do not produce thestress response in its entirety. In particular,he has cited the effects of heat. Further-more, Lacey (1967) has argued that the lowcorrelations observed among differentphysiological components of the stress re-sponse are not consistent with the notion ofan identifiable response syndrome. There isalso a difficulty in distinguishing betweenthose physiological changes which repre-sent stress and those which do not, particu-larly as the former may be dissociated intime from the stressor (Fisher, 1986).

There is now much research that suggeststhat if the stress response syndrome existsit is not non-specific. There are subtle butimportant differences in the overall patternof response. There is evidence, for exam-ple, of differentiation in the response ofthe catecholamines (reflecting sympathet-ic-adrenal medullary activation) to stressfulsituations (Cox & Cox, 1985). Several di-mensions have been suggested as a basisof this differentiation but most relate to theexpenditure of effort of different types, forexample, physical versus psychological(Dimsdale & Moss, 1980a, 1980b; S. Coxet al., 1985). Dimsdale & Moss (1980b)studied plasma catecholamine levels usinga non-obtrusive blood withdraw pump andradioenzymatic assay. They examined 10young physicians engaged in public speak-ing, and found that although levels of bothadrenaline and noradrenaline increased


35■

under this set of demands, the levels ofadrenaline were far more sensitive. Thissensitivity was associated with feelings ofemotional arousal which accompanied thepublic speaking. S. Cox and her colleagues(1985) examined the physiological re-sponse to three different types of task as-sociated with short cycle repetitive work:urinary catecholamine excretion rates weremeasured using an adaptation of Diament& Byers (1975) assay technique. She foundthat both adrenaline and noradrenalinewere sensitive to work characteristics, suchas pay scheme and pacing, but differential-ly so. It was suggested that noradrenalineactivation was related to the physical activ-ity inherent in the various tasks, and to theconstraints and frustrations present, whileadrenaline activation was more related tofeelings of effort and stress.

The second criticism is that the engineer-ing and physiological models of stress areconceptually dated in that they are setwithin a relatively simple stimulus-re-sponse paradigm, and largely ignore indi-vidual differences of a psychologicalnature and the perceptual and cognitiveprocesses that might underpin them (Cox,1990; Sutherland & Cooper, 1990; Cox,1993). These models treat the person as apassive vehicle for translating the stimuluscharacteristics of the environment intopsychological and physiological responseparameters. They also ignore the interac-tions between the person and their variousenvironments which are an essential partof systems-based approaches to biology,behaviour and psychology. In particular,they ignore the psychosocial and organisa-tional contexts to work stress.

3 . 1 . 3 P s y c h o l o g i c a l A p p r o a c h

The third approach to the definition andstudy of stress conceptualises it in terms ofthe dynamic interaction between the per-son and their work environment. Whenstudied, it is either inferred from the exis-tence of problematic person-environmentinteractions or measured in terms of thecognitive processes and emotional reactionswhich underpin those interactions. This hasbeen termed the ‘psychological approach’.

The development of psychological modelshas been, to some extent, an attempt toovercome the criticisms levelled at the ear-lier approaches. There is now a consensusdeveloping around this approach to thedefinition of stress. For example, psycho-logical approaches to the definition ofstress are largely consistent with the Inter-national Labour Office’s definition of psy-chosocial hazards (International LabourOrganization, 1986) and with the defini-tion of well-being recommended by theWorld Health Organization (1986)5. Theyare also consistent with the developing lit-erature on personal risk assessment (see,for example, Cox & Cox, 1993; Cox, 1993;Cox & Griffiths, 1995, 1996). These con-sistencies and overlaps suggest an increas-ing coherence in current thinking withinoccupational health and safety.

Variants of this psychological approachdominate contemporary stress theory, and

5 Well-being is a dynamic state of mind characterisedby reasonable harmony between a person’s abilities,needs, and expectations, and environmental demandsand opportunities (WHO, 1986). The individual’s sub-jective assessment is the only valid measure of well-be-ing available (Levi, 1992).


■36

–among them– two distinct types can beidentified: the interactional and the trans-actional. The former focus on the structur-al features of the person’s interaction withtheir work environment, while the latterare more concerned with the psychologi-cal mechanisms underpinning that interac-tion. Transactional models are primarilyconcerned with cognitive appraisal andcoping. In a sense they represent a devel-opment of the interactional models, andare essentially consistent with them.


37■

ory of work stress based on the explicit con-cept of the Person-Environment Fit (see, forexample, French et al., 1982). Two basic as-pects of fit were identified:• The degree to which an employee’s atti-

tudes and abilities meet the demands ofthe job.

• The extent to which the job environ-ment meets the workers’ needs, and inparticular the extent to which the indi-vidual is permitted and encouraged touse their knowledge and skills in the jobsetting.

It has been argued that stress is likely tooccur, and well-being is likely to be affect-ed, when there is a lack of fit in either orboth respects (French et al., 1974). Twoclear distinctions are made in this theory:first, between objective reality and subjec-tive perceptions, and, second, betweenenvironmental variables (E) and personvariables (P). Given this simple 2 x 2 con-figuration of P x E interaction, lack of fitcan actually occur in four different ways,and each appear to challenge the worker’shealth. There can be both a lack of subjec-tive and objective P-E fit: these are themain foci of attention with particular inter-est being expressed in the lack of subjec-tive fit: how the worker sees their worksituation. This provides a strong link withother psychological theories of stress.There can also be a lack of fit between theobjective environment (reality) and thesubjective environment (hence, lack ofcontact with reality), and also a lack of fitbetween the objective and subjective per-sons (hence, poor self-assessment).

French et al. (1982) have reported on alarge survey of work stress and health in 23

3.2I N T E R A C T I O N A L T H E O R I E S O F

S T R E S S

Interactional theories of stress focus on thestructural characteristics of the person’s in-teraction with their work environment.Two particular interactional theories standout as seminal among the various whichhave been offered: the Person-Environ-ment Fit theory of French et al. (1982) andthe Demand–Control theory of Karasek(1979). Neither is, however, without criti-cism: see, for example, Edwards & Cooper(1990) and Warr (1990).

3 . 2 . 1 P e r s o n - E n v i r o n m e n t F i t

Several researchers have suggested that thegoodness of fit between the person andtheir (work) environment frequently offersa better explanation of behaviour than indi-vidual or situational differences (see, for ex-ample, Bowers, 1973; Ekehammer, 1974).Largely as a result of such observations,French and his colleagues formulated a the-


■38

different occupations in the United Statesand a sample of 2010 working men. Thesurvey was framed by the P-E Fit theory,and, in their summary, the authors com-mented on a number of questions of theo-retical and practical importance. Inparticular, they argued that their subjectivemeasures mediated the effects of objectivework on health. Their data showed thatthere was a good correspondence betweenthe objective and subjective measures andthat the effects of those objective measureson self-reported health could be very large-ly accounted for by the subjective mea-sures. This has been reflected more recentlyin the work of various researchers (see, forexample, Bosma & Marmot, 1997; Jex &Spector, 1996; Chen & Spector, 1991;Spector, 1987b). In French et al.’s study, ob-jective occupation only accounted for some2 to 6 percent of the variance in self-re-ported health beyond that accounted forby the subjective measures.

3 . 2 . 2 D e m a n d - C o n t r o l M o d e l

Karasek (1979) drew attention to the pos-sibility that work characteristics may notbe linearly associated with worker health,and that they may combine interactively inrelation to health. He initially demonstrat-ed this theory through secondary analysesof data from United States and Sweden,finding that employees in jobs perceived tohave both low decision latitude and highjob demands6 were particularly likely to re-

port poor health and low satisfaction. Lat-er studies appeared to confirm the theory.For example, a representative sample ofSwedish working men was examined fordepression, excessive fatigue, cardiovascu-lar disease and mortality. Those workerswhose jobs were characterised by heavyworkloads combined with little latitude fordecision making were represented dispro-portionately on all these outcome vari-ables. The lowest probabilities for illnessand death were found among workgroups with moderate workloads com-bined with high control over work condi-tions (Ahlbom et al., 1977; Karasek, 1981;Karasek et al., 1981). The combined effectof these two work characteristics is oftendescribed as a true interaction, but despitethe strong popular appeal of this sugges-tion there is only weak evidence in its sup-port (Kasl, 1989; Warr, 1990). Karasek’s(1979) own analyses suggest an additiverather than a synergistic effect, and he hasadmitted that “there is only moderate evi-dence for an interaction effect, under-stood as a departure from a linear additivemodel”. Simple additive combinationshave been reported by a number of re-searchers, for example, Hurrell & McLaney(1989), Payne & Fletcher (1983), Perrewe& Ganster (1989), and Spector (1987a).

Other criticisms have been levelled againstKarasek’s model. For instance, it wasclaimed that the model was too simpleand ignores the moderating effect of socialsupport on the main variables. Johnson(1989) and Johnson et al. (1991) expand-ed Karasek’s model by adding a thirddimension, resulting in the “Demand-Con-trol-Support” model. The dimension “so-

6 Karasek (1979) defined ‘decision latitude’ as ‘theworking individual’s potential control over his tasks andhis conduct during the working day’. He defined ‘jobdemands’ as ‘the psychological stressors involved in ac-complishing the workload’.


39■

cial support” refers to overall levels ofhelpful social interaction available on thejob from both co-workers and supervisors.“Social support” seems to play an essen-tial role in the management of stress atwork. It serves as a buffer against possibleadverse health affects of excessive psycho-logical demands (Theorell, 1997). Johnsonet al. (1991) distinguish between fourtypes of low social support work situationsand four of high social support. Winnubst& Schabracq (1996) found that high de-mands, low control and low support (highsocial isolation) were associated with an el-evated cardiovascular risk. Most studiesbased on this model focus on jobs, i.e.,broad occupational categories. Junghannset al. (1999) applied the “Demand-Con-trol-Support” model to specific conditionsof work and confirmed that job character-istics such as decision latitude, psychologi-cal demands and social support affecthealth. They found that white-collar work-ers in “high-strain” work situations hadthe highest level of health complaints.Working situations characterised as highlydemanding with low decision latitude andlow social support predispose workers toexperience health problems, especiallymusculoskeletal (shoulder and neck pain)and psychosomatic complaints (exhaus-tion, inner restlessness) (Ertel et al., 1997;Junghanns et al., 1999).

The expanded “Demand-Control-Support”model has also been criticised for its failureto consider individual differences in suscep-tibility and coping potential: The relation-ship between the dimensions of the modeland the outcome measures may dependupon workers’ individual characteristics (de

Rijk et al., 1998). For instance, “disturbedrelaxation ability” (also known as “inabilityto relax/work obsession”) was found to bea valid predictor of increased sympatheticactivation and delayed recovery of cardio-vascular parameters. It reflects experiencedintensity of work and job-related exhaus-tion (Richter et al., 1988, Richter et al.,1995). “Disturbed relaxation ability” re-lates to excessive work involvement, char-acterised by an extreme degree of workeffort and by work “carry-over” into do-mestic life (to the extent of affecting sleep,relaxation and leisure, and neglecting per-sonal needs). While a certain degree ofwork involvement can be considered“healthy” and stimulating, in its extremeform involvement can become ‘work ob-session’ and lead to the inability to relax af-ter work, with the risk of negative healtheffects (Rotheiler et al., 1997). “Disturbedrelaxation ability” can moderate the healtheffects of the work-situations generated bythe “Demand-Control-Support” model.Junghanns et al., (1998) found that highpsychological demands and a high level ofdisturbed relaxation ability predisposeworkers to ill-health.

Finally, Carayon (1993) has offered fourpossible explanations for the inconsistencyin the evidence concerning Karasek’s mod-el. First, the model seems to be supportedin large, heterogeneous samples, but notin homogeneous samples: this may be dueto the confounding effects of socio-eco-nomic status in heterogeneous samples orthe lack of sensitivity of measures used inhomogeneous samples. Second, inconsis-tencies may stem partly from the way jobdemands and decision latitude are concep-


■40

tualised and measured. Karasek conceptu-alised decision latitude as a combination ofdecision authority (similar to control or au-tonomy) and skill discretion (similar to skillutilisation). Subsequent studies have in-cluded a wide variety of measures for deci-sion latitude, and it is therefore possiblethat those that have used more focusedmeasures are testing the effects of ‘con-trol’ as opposed to the effects of ‘decisionlatitude’, which is a mixture of control andjob complexity. Similarly, as far as ‘de-mands’ are concerned, the original mea-sures tapped one main construct,‘workload’, but subsequent studies havetended to employ a wider range of mea-sures. Measures have varied considerablyand are often far removed from Karasek’soriginal formulation. Third, much of the re-search into this model relies on self-reportmeasures of both dependent and indepen-dent variables; ‘job satisfaction’ is an ex-ample where there is content overlapbetween the measures. A related issueconcerns the predominance of cross-sec-tional rather than longitudinal data, limit-ing interpretations as to cause and effect.Fourth, Carayon suggests there may bemethodological and statistical reasons forthe failure to find interactive effects. How-ever, whether perceived job demands anddecision latitude combine additively orthrough a true interaction, it is clear fromKarasek’s work that they are importantfactors determining the effects of work onemployees’ health.


41■

are most prevalent in occupations wheresituational constraints prevent workersfrom reducing “high cost - low gain” con-ditions.

3 . 3 . 1 T h e o r i e s o f A p p r a i s a l a n dC o p i n g

Most transactional models appear to buildon the conceptual structures suggested inthe interactional models of the Michiganschool and Karasek and colleagues. Theyfocus on the possible imbalance betweendemands and ability or competence. This ismost obvious in the models advanced byLazarus and Folkman in the United States(for example, Lazarus & Folkman, 1984)and Cox and Mackay in the United King-dom (for example, Cox, 1978; Cox, 1990;Cox & Mackay, 1981). According to trans-actional models, stress is a negative psy-chological state 7 involving aspects of bothcognition and emotion. They treat thestress state as the internal representationof particular and problematic transactionsbetween the person and their environ-ment.

Appraisal is the evaluative process thatgives these person-environment transac-tions their meaning (Holroyd & Lazarus,1982). Later refinements of the theorysuggest both primary and secondary com-ponents to the appraisal process (Lazarus,1966; Folkman & Lazarus, 1986). Primaryappraisal involves a continual monitoringof the person’s transactions with their en-

3.3T R A N S A C T I O N A L

D E F I N I T I O N S

Most transactional theories of stress focuson the cognitive processes and emotionalreactions underpinning the person’s inter-action with their environment. For exam-ple, Siegrist’s transactional model of“effort-reward imbalance” (Siegrist, 1990)argues that the experience of chronicstress can be best defined in terms of amismatch between high costs spent andlow gains received. In other words, accord-ing to the model, stress at work resultsfrom high effort spent in combination withlow reward obtained. Two sources of ef-fort are distinguished: an extrinsic source,the demands of the job, and an intrinsicsource, the motivation of the individualworker in a demanding situation. Three di-mensions of reward are important: finan-cial gratifications, socio-emotional rewardand status control ( i .e., promotionprospects and job insecurity). Adversehealth effects, such as cardiovascular risk,

7 The term psychological stress is ambiguous. Whilethe experience of stress is psychological in nature, itsantecedents and outcomes are not restricted to anyparticular domain, psychological or otherwise.


■42

vironment (in terms of demands, abilities,competence, constraints and support), fo-cusing on the question ‘Do I have a prob-lem?’ The recognition of a problemsituation is usually accompanied by un-pleasant emotions or general discomfort.Secondary appraisal is contingent uponthe recognition that a problem exists andinvolves a more detailed analysis and thegeneration of possible coping strategies:‘What am I going to do about it?’.

Stress arises when the person perceivesthat he or she cannot adequately copewith the demands being made on them orwith threats to their well-being (Lazarus,1966, 1976; Cox, 1990), when coping isof importance to them (Sells, 1970; Cox,1978) and when they are anxious or de-pressed about it (Cox & Ferguson, 1991).The experience of stress is therefore de-fined by, first, the person’s realisation thatthey are having difficulty coping with de-mands and threats to their well-being,and, second, that coping is important andthe difficulty in coping worries or depress-es them. This approach allows a clear dis-tinction between, say, the effects of lack ofability on performance and those of stress.If a person does not have the necessaryability or competence –the knowledge orlevel of skill– to complete a task, then theirperformance will be poor. They may notrealise this or if they do it might not be feltto be of importance or give rise to con-cern. These are not stress scenarios. How-ever, if the person (a) does realise that theyare failing to cope with the demands of atask, and (b) experiences concern aboutthat failure because it is important, thenthis is a ‘stress’ scenario. The effects of

such stress might then cause a further im-pairment of performance over and abovethat caused by lack of ability.

The question of ‘consciousness’ has beenraised in relation to stress and the appraisalprocess (Cox & Mackay, 1981). Appraisal isa conscious process. However, in its earli-est stages, changes characteristic of thestress state may be demonstrated, yet theexistence of a problem may not be recog-nised or recognition may only be ‘hazy’. Ithas been suggested that different levels ofawareness may exist during the appraisalprocess. These may be described by thefollowing sequence:1. Growing awareness of problem mark-

ers, both individual and situational, in-cluding feeling uncomfortable, notsleeping, making mistakes, etc.

2. Recognising the existence of a ‘prob-lem’ in a general or ‘hazy’ way.

3. Identifying the general problem areaand assessing its importance.

4. Analysing in detail the nature of theproblem and its effects.

It is useful to think of the stress state asembedded in an on-going process that in-volves the person interacting with their en-vironment, making appraisals of thatinteraction and attempting to cope with,and sometimes failing to cope with, theproblems that arise. Cox (1978) describedthis process in terms of a five-stage model.The first stage, it was argued, representsthe sources of demand faced by the per-son and is part of their environment. Theperson’s perception of these demands inrelation to their ability to cope representsthe second stage: effectively primary ap-praisal. Consistent with Lazarus & Folkman


43■

(Lazarus, 1966; Folkman & Lazarus, 1986)and French et al. (1982), stress was de-scribed as the psychological state thatarose when there was a personally signifi-cant imbalance or lack of fit between theperson’s perceptions of the demands onthem and their perceived ability to copewith those demands. The psychologicaland physiological changes which are asso-ciated with the recognition of such a stressstate, and which include coping, representthe third stage of the model. Emotionalchanges are an important part of the stressstate. These tend to be negative in natureand often define the experience of stressfor the person. The fourth stage is con-cerned with the consequences of coping.The fifth stage is the general feedback(and feed forward) that occurs in relationto all other stages of the model. This mod-el has been further developed in severalways. The importance of perceptions ofcontrol and of social support have beenemphasised as factors in the appraisalprocess, and there has been some discus-sion of the problem of measuring stressbased on this approach (Cox, 1985a,1990) with the development of possiblesubjective measures of the experiential(mood) correlates of the stress state (seeMackay et al., 1978; Cox & Mackay, 1985).

The experience of stress through work istherefore associated with exposure to par-ticular conditions of work, both physicaland psychosocial, and the worker’s realisa-tion that they are having difficulty in cop-ing with important aspects of their worksituation. The experience of stress is usual-ly accompanied by attempts to deal withthe underlying problem (coping) and by

changes in cognition, behaviour and phys-iological function (e.g., Aspinwall & Taylor,1997; Guppy & Weatherstone, 1997). Al-though probably adaptive in the shortterm, such changes may threaten health inthe long term. The experience of stress andits behavioural and psychophysiologicalcorrelates mediate8, in part, the effects ofmany different types of work demand onhealth. This point has been made by manyauthors over the last three decades (for ex-ample, Levi, 1984; Szabo et al., 1983;Scheck et al., 1997).

8 The mediator of a particular relationship, for examplebetween stress and health, is a variable which effec-tively supplies the link between the two variables in-volved: it transmits the effects of one variable to theother.


■44

Stress can be defined as a psychologicalstate which is part of and reflects a widerprocess of interaction between the personand their work environment. It is conclud-ed that there is a growing consensusaround the adequacy and utility of the psy-chological approach to stress. Severaloverview models have been offered assummaries of the stress process. The mostnotable is that of Cooper (see, for exam-ple, Cooper & Marshall, 1976), as present-ed in Figure 2 below. Cooper’s modelusefully focuses on the nature and detail ofwork stresses and their individual and or-ganisational outcomes.

3.4S U M M A R Y : F R A M E W O R K S ,

T H E O R I E S A N D D E F I N I T I O N S

F i g u r e 2 : C o o p e r ’ s m o d e l o f t h e d y n a m i c s o f w o r k s t r e s s ( a d a p t e d f r o m C o o p e r& M a r s h a l l , 1 9 7 6 )

Home-workinterface

Organisationalstructure and

climate

Careerdevelopment

Relationshipsat work

Role in theorganisation

Intrinsic to thejob

Sources of stress

Individual symptoms

■ Raised blood pressure■ Depressed mood■ Excessive drinking■ Irritability■ Chest pains

Symptoms of stress

INDIVIDUAL

Organisationalsymptoms

■ High absenteeism■ High labour turnover■ Industrial relations difficulties■ Poor quality control

Disease

Coronary heart

Mental illness

Prolonged strikes

Frequent andsevere accidents

Apathy

?

?


45■

The stress state is a conscious state but thelevel of awareness of the problem varieswith the development of that state. Part ofthe stress process are the relationships be-tween the objective work environmentand the employee’s perceptions of work,between those perceptions and the expe-rience of stress, and between that experi-ence and changes in behaviour andphysiological function, and in health. Cop-ing is an important component of thestress process but one which is relativelypoorly understood. Stress may be experi-enced as a result of exposure to a widerange of work demands and, in turn, con-tribute to an equally wide range of healthoutcomes: it is one link between hazardsand health.


■46

produce a comprehensive taxonomy, andthat which considers coping as a problem-solving process (Dewe, 2000).

3 . 5 . 1 C o p i n g T a x o n o m i e s

Lazarus (1966) has argued that the personusually employs both task and emotion fo-cused coping strategies. The former at-tempt some form of action directlytargeted at dealing with the source ofstress (adaptation of the environment),while the latter attempt to attenuate theemotional experience associated with thatstress (adaptation to the environment).The perceived success, or otherwise, ofsuch strategies feeds back into the ap-praisal process to alter the person’s per-ception of the situation. Lazarus and hiscolleagues (Lazarus, 1966; Lazarus & Folk-man, 1984) also emphasise that the im-portance of the situation to the individualis critical in determining the intensity oftheir response.

Dewe (1987), in a typical study, examinedsources of stress and strategies used tocope with them in ministers of religion inNew Zealand. Using factor analytical tech-niques, he identified five clusters of copingstrategies: seeking social support, post-poning action by relaxation and distractingattention, developing greater ability todeal with the problem, rationalising theproblem, and drawing on support throughspiritual commitment. It was possible toclassify 33% of the strategies which madeup these clusters as task focused and 67%as emotion focused. The most frequentsource of stress experienced by the minis-ters related to the emotional and time dif-ficulties associated with crisis work, and

3.5C O P I N G

Coping is an important part of the overallstress process. However, it is perhaps theleast well understood despite many yearsof research. This point is widely acknowl-edged in the literature (see, for example,Dewe et al., 1993, 2000) Lazarus (1966)has suggested that it has three main fea-tures. First, it is a process: it is what theperson actually thinks and does in a stress-ful encounter. Second, it is context-depen-dent: coping is influenced by the particularencounter or appraisal that initiates it andby the resources available to manage thatencounter. Finally, coping as a process isand should be defined ‘independent ofoutcome’; that is, independently ofwhether it was successful or not (see Folk-man, 1984; Folkman et al., 1986a, 1986b;Lazarus & Folkman, 1984). There havebeen two approaches to the study of cop-ing: that which attempts to classify the dif-ferent types of coping strategies and


47■

the experience of such problems appearedto be associated with coping by seekingsocial support and rationalising the prob-lem.

Pearlin and associates (Pearlin & Schooler,1978; Pearlin et al., 1981) have further de-veloped this general approach and distin-guished between responses concernedwith changing the situation, those con-cerned with changing its meaning (re-ap-praisal) and those relating to themanagement of the symptoms of stress. Ina different vein, Miller (1979; Miller et al.,1988) has distinguished between two in-formational styles which she terms‘blunters’ and ‘monitors’: the former tendto use denial strategies and the latter in-formation seeking strategies in relation tostressful situations.

These and the many other classificationsavailable in the literature are, generally,neither inconsistent nor meant to be mu-tually exclusive. Most authors emphasisethat no one type of coping strategy is nec-essarily better than any other in solving aproblem. People use a mixture of strate-gies in most situations, although certainsituations may tend to be associated withparticular types of strategy. Some studieshave tried to explore the existence of sys-tematic links between stressors and copingstyles, but found little empirical supportfor their hypotheses. Salo (1995) founddifferences in teachers’ ways of coping,but those differences related to theamount, not the source, of stress experi-enced, and the timing (changed through-out the autumn term). Wykes &Whittington (1991) studied the differentways in which psychiatric nursing staff

dealt with incidents of violent physical as-sault. They found that each respondent re-ported an average three distinct copingstrategies. These studies seem to supportthe existence of complex, dynamic andcontext-dependent coping behaviours,rather than causally driven schemata ofcoping. Furthermore, although in theoryLazarus’ model allows for environmentalfeedback to alter the perceptions –and,hence, perhaps to determine future cop-ing– in practice his taxonomy is rather sta-tic and emphasises coping styles, whilsttending to ignore coping behaviours(Dewe et al., 1993).

3 . 5 . 2 C o p i n g a s P r o b l e m - s o l v i n g

Coping can also be viewed as a problem-solving strategy (Cox, 1987; Fisher, 1986;Dewe, 1993; Aspinwall & Taylor, 1997).Cox (1987), for example, has described acycle of activities, beginning with recogni-tion and diagnosis (analysis) followed byactions and evaluation through to re-analysis, which possibly represents the ide-al problem-solving process. However,Schonpflug & Battmann (1988) have em-phasised that by adopting the wrong ac-tions, or by failing, a person may createfurther problems and stress. At the sametime, Meichenbaum (1977) argues that‘catastrophizing’ or reacting too stronglyto such failure serves no adaptive purposeand it is often said that one of the few pos-itive aspects to coping with stress is thatthe person learns from such experience.However, Einhorn & Hogarth (1981) sug-gest that there are at least three problemswith this proposition: first, one does notnecessarily know that there is something


■48

to be learned, second, what is to belearned is not clear, and third, there is am-biguity in judging whether one haslearned. Furthermore, the problem solvermay be fully occupied and not have anyspare cognitive capacity for learning, andthe emotion associated with stress may in-terfere with the learning process (Mandler,1982).

Coping may be seen as functional in its at-tempts to manage demands, by eitherchanging them, redefining them (re-ap-praisal) or adapting to them. The stylesand strategies used need to be relevantand applicable to the situation at hand.The choice and successful use of these re-sponses will be determined by both thenature of the situation, by the personaland social resources available and also bythe type of causal reasoning adopted inthe appraisal process.


49■

or (2) moderators9 of the stress-health re-lationship (see Cox & Ferguson, 1991).Hence, researchers have asked, for exam-ple, to what extent are particular workersvulnerable to the experience of stress, or towhat extent does, say, ‘hardiness’ (Kobasa,1979; Kobasa & Pucetti, 1983; Kobasa etal., 1981, 1982) moderate the relationshipbetween job characteristics and workerhealth?

This distinction between individual differ-ences as components of the appraisalprocess and moderators of the stress-out-come relationship can be easily under-stood in terms of transactional models ofstress (e.g., Cox & Griffiths, 1996).

Primary appraisal is, by its very nature, sub-ject to individual differences. First, individ-ual differences may exist in relation to theperson’s perception of job demands andpressures. Kahn (1974), for example,found a modest relationship between ob-jective and subjective measures of roleconflict. The objective measure was basedon the sum of pressures to change behav-iour as reported by those who had formalinfluence on the person in the role in ques-tion. Further analyses revealed that this re-lationship largely resulted from those inthe sample who were high on anxietyproneness. Anxiety proneness appeared tomoderate the person’s perception of roleconflict. In the same vein, Payne & Hartley

3.6I N D I V I D U A L A N D G R O U P

D I F F E R E N C E S

Most contemporary theories of stress al-low for individual differences in the experi-ence of stress, and in how and how well itis coped with. In 1988, Payne presented aseries of questions, including:

• How do individual differences relate toperceptions of stress in the work envi-ronment?

• Do they affect the way people cope withstress?

• Do they act as moderators of the stress-health relationship?

• How do individual differences, such ascompetence and work ability, relate tothe development of ill health?

There would appear to be two differentapproaches to research on individual dif-ferences based on Payne’s (1988) ques-tions. Effectively individual differencevariables have been investigated as either:(1) components of the appraisal process,

9 A moderator of a particular relationship, say be-tween stress and health, is a variable which may alterthe strength or direction of that relationship. The tech-nical concept of moderation implies no particular di-rection of effect although in every day usage it tends toimply a weakening of effect.


■50

(1987) found a positive correlation be-tween perceptions of the severity of prob-lems facing unemployed men and ameasure of locus of control. The more theybelieved that important life events werenot under their personal control, the moresevere they perceived their problems to be.Second, people vary in their ability to copewith demands, and in their perceptions ofthose abilities. Such variation may be afunction of their intelligence, their experi-ence and education, or their beliefs in theirability to cope (self efficacy: Bandura,1977; job self-efficacy: Schaubroeck &Merritt, 1997). Third, people may vary inthe amount of control that they can exer-cise over any situation, not only as a func-tion of that situation but also as a functionof their beliefs about control. Fourth, peo-ple may vary in their need for social sup-port and the skills that they have forexploiting such support, and in their per-ceptions of support. Finally, the stress-health relationship is obviously moderatedby individual differences not only in sec-ondary appraisal but also in coping behav-iour and emotional and physiologicalresponse tendencies, latencies and pat-terns.

3 . 6 . 1 T y p e A B e h a v i o u r

Over the last 30 years, much attention hasfocused on individual vulnerability in rela-tion to coronary heart disease and on therole of psychological and behavioural fac-tors in reacting to and coping with stress-ful situations. The concept of type Abehaviour was originally developed as adescription of overt behaviour by Friedman& Rosenman (1974) but has since been

considerably broadened and, some haveargued, weakened as a result (Arthur etal., 1999; Powell, 1987). Friedman &Rosenman (1974) described type A behav-iour as a major behavioural risk factor forcardiovascular ill health. There are at leastthree characteristics that mark out the typeA individual whose risk of coronary heartdisease appears, from studies in the Unit-ed States, to be at least twice that of thenon type A:• A strong commitment to work and

much involvement in their job• A well developed sense of time urgency

(always aware of time pressures andworking against deadlines)

• A strong sense of competition and amarked tendency to be aggressive

Such behaviour is probably learnt, and isoften valued by and maintained throughparticular organisational cultures.

There is some confusion in the literature asto the status of the behaviours referred toabove and their relative importance andthat of related constructs. Some refer totype A behaviour as a learnt style of be-haviour, others as a coping pattern, andstill others as a personality trait (Powell,1987). At the same time, there have beenvarious suggestions as to its most impor-tant dimension. For example, Glass (1977)has argued that control is the determiningfactor, while Williams et al. (1985) andothers have argued in favour of hostility oraggression (see for example, Dembroski etal., 1985; MacDougal et al., 1985), andothers for low self esteem (Friedman & Ul-mer, 1984). Various different measureshave been developed, not all of whichstrongly inter-relate (e.g.: Arthur et al.,


51■

1999; Powell, 1987), thus questioningtheir operational precision and constructvalidity.

Perhaps of the three, the two that have at-tracted most attention are (i) control and(ii) anger and hostility.

C o n t r o l

The issue of control is important in under-standing the nature of type A behaviour.The type A individual feels that they are al-ways fighting to maintain control overevents, which are often seen to be just be-yond their grasp. Faced with these situa-tions, they simply expend more time andeffort trying to “get events under control”–and never really feel as if they have suc-ceeded (Glass & Singer, 1972). The issue ofcontrol, and of being in control, is an im-portant one and may distinguish betweenthe vulnerability of type As and the resis-tance of hardy types (Kobasa, 1979; Wein-berg et al., 1999; Kobasa & Pucetti, 1983;Kobasa et al., 1981, 1982). Kobasa’s hardytypes report feeling in control of their workand their lives. Type A behaviour predictscardiovascular ill health, while hardinesspredicts general good health.

A n g e r & H o s t i l i t y

Indices of anger and hostility have beenvalidated in prospective research as predic-tors of cardiovascular ill health. For exam-ple, Matthews et al. (1977) scored 10responses to the Structured Interview fortype A behaviour (see Jenkins et al., 1968)of 186 cases and controls in the WesternGroup Collaborative Study (see, for exam-ple, Rosenman et al., 1964a and 1964b).

Seven of the 10 responses discriminatedbetween the cases and controls and themajority of these related to anger and hos-tility. Others have also found evidence sug-gesting that measures of hosti l ity,repressed hostility or potential for hostilitycan strongly predict cardiovascular health(Dembroski et al., 1985; Arthur et al.,1999; MacDougall et al., 1985; Williams etal., 1980; Barefoot et al., 1983; Shekelle etal., 1983). Perez et al. (1999) have recent-ly found that expression of anger discrimi-nated between coronary patients on theone hand, and non-coronary patients andhealthy people on the other.

The relationship between type A behav-iour and cardiovascular health is potential-ly moderated by a host of factors such asage, sex, socio-economic and educationalstatus, employment status, medicationand the cardiovascular outcome chosenfor study (Powell, 1987). Interestingly, Kit-tel and his colleagues (1983) have con-cluded that there are also markeddifferences between studies in the UnitedStates and those in Europe. The resultsfrom Europe do not appear to have ful-filled the early promise of those conductedin the United States. There may be socio-linguistic and cultural differences which af-fect either the validity of the measuringinstruments or the validity or role of theconcept (e.g., Lu et al., 1999; Martinez &Martos, 1999; Mudrack, 1999; Kawakami& Haratani, 1999).

3 . 6 . 2 V u l n e r a b l e G r o u p s

Individual differences are obvious in thestress process affecting appraisal and cop-ing mechanisms, and the stress-health re-


■52

lationship. Group differences –and the cre-ation of vulnerable groups– may representthe effects of individual differences whichare common to, and characteristic of, par-ticular groups, and/or the effects of com-mon patterns of exposure to hazardouswork conditions (or some combination ofthe two; see, for example, Weinberg etal.’s (1999) study of British Members ofParliament). Several different reviews haveidentified possible vulnerable groups (see,for example, Levi, 1984; Davidson & Earn-shaw, 1991) including: young workers,older workers, migrant workers, disabledworkers and women workers. Kasl (1992)has attempted to summarise the differentcriteria and factors that define vulnerabili-ty as: socio-demography (for example, ageand educational status), social status (forexample, living alone), behavioural style(type A behaviour), skills and abilities,health status and medical history, and on-going non-work problems. Such vulnera-bil ity factors are moderators of thehazard-stress-harm relationship and prob-ably interact in defining the high risk orvulnerable groups mentioned above.

The recognition of the vulnerability of suchgroups is not new and, in the United King-dom, its origins can be traced back to theearliest health and safety legislation as, forexample, in the Health & Morals of Ap-prentices Act of 1802.

3 . 6 . 3 S e l e c t i o n

The individual and group differences havebeen highlighted in relation to the experi-ence and health effects of stress. Such dif-ferences may be treated in a number ofways depending as much on moral and le-

gal as on scientific considerations. Exclud-ing particular workers or types of workerfrom work, which is judged to be stressful,may appear, at first sight, to be scientifi-cally justified, but may not be legally sanc-tionable under the Equal Opportunitieslegislation in the EU Member States, ormorally acceptable if other approaches arepossible.

Furthermore, although individual differ-ences can be shown to moderate the haz-ard-stress-health process, the evidence isnot strong enough to support the designof defensible selection procedures. Thereappears to be little evidence of trait-likevulnerability to stress beyond that impliedfor psychological health by a personal orfamily history of related psychological dis-orders. Evidence for the apparent exis-tence of any such traits may simply reflectcommonly occurring patterns of person xenvironment interactions. Alternativestrategies, focused on the design of jobsand organisation of work are available andmore defensible given current knowledgeof the relationship between work hazardsand stress. Equally, approaches based onworker education and training and on en-hanced support for workers in order to in-crease their work ability are also possible,and have been tried and evaluated.


53■

3.7S U M M A R Y : I N D I V I D U A L

D I F F E R E N C E S – W O R K

A B I L I T Y A N D C O P I N G

The experience of stress is partly depen-dent on the individual’s ability to cope withthe demands placed on them by theirwork, and on the way in which they sub-sequently cope with those demands, andrelated issues of control and support. Moreinformation is required on the nature,structure and effectiveness of individuals’abilities to meet work demands and tocope with any subsequent stress. The needfor more information on coping is widelyrecognised (see, for example, Dewe,2000), but relatively less attention hasbeen paid to the need better to under-stand the concept of work ability or com-petence, although this is being flagged inrelation to ageing research (e.g., Griffiths,1999a; Ilmarinen & Rantanen, 1999)


■54


55■

4.M E T H O D O L O G I C A L I S S U E S

RE

SE

AR

CH


■56

Logically the measurement of the stressstate must be based primarily on self-re-port measures which focus on the ap-praisal process and on the emotionalexperience of stress (Cox, 1985a; Cox &Ferguson, 1994). Measures relating to ap-praisal need to consider the worker’s per-ceptions of the demands on them, theirability to cope with those demands, theirneeds and the extent to which they are ful-filled by work, the control they have overwork and the support they receive in rela-tion to work. Dewe (1991) has argued thatit is necessary to go beyond simply askingworkers whether particular demands, etc.are present (or absent) in their work envi-ronments and measure various dimensionsof demand such as frequency, durationand level. Furthermore, such measuresneed to be used in a way which allows forthe possibility of interactions between per-ceptions, such as demand with control(Karasek, 1979; Warr, 1990) or demandand control with support (Payne & Fletch-er, 1983; Cox, 1985a; Karasek & Theorell,1990). The importance to the worker ofcoping with particular combinations andexpressions of these work characteristicsneeds also to be taken into account (Sells,1970; Cox, 1978).

4.1M E A S U R E M E N T

It has been suggested that the availableevidence supports a psychological ap-proach to the definition of stress, and thattransactional models are among the mostadequate and useful of those currentlyavailable. Within this framework, stress isdefined as a psychological state (see sec-tion 3.1.3) which is both part of and re-flects a wider process of interactionbetween the person and their (work) envi-ronment. This process is based on a se-quence of relationships between theobjective work environment and the work-er’s perceptions, between those percep-tions and the experience of stress, andbetween that experience, changes in be-haviour and physiological function, andhealth. This sequence provides a basis formeasurement, but the different measureswhich can be derived from the sequencecannot be easily or defensibly combinedinto a single stress index (see below).


57■

be defined as “a general personality traitreflecting individual differences in negativeemotionality and self-concept, i.e., con-centrating on negative aspects of every-thing and experiencing considerabledistress in all situations” (Watson & Clark,1984). NA would affect not only workers’perception of their work environment, butalso their appraisal of their own psycho-logical health status or well-being, thusbecoming a confounding variable thatcould account for a large proportion of thecorrelations between perceived hazardsand perceived outcomes. Kasl (1987) re-ferred to this methodological weaknesswhen he wrote about the triviality trap(i.e., some researchers’ reliance on trivialcorrelations that can be explained away bycommon method variance): “The sheervolume of studies which has been gener-ated by cross-sectional retrospective de-signs, in which only self-reports ofindependent, intervening, and outcomevariables are correlated to each other, is soenormous that they have created theirown standard of “acceptable methodolo-gy” Kasl, 1987).

The research literature is still divided on theextent to which NA or common methodvariance distort the assessment of thestress-strain relationship (e.g., Jex & Spec-tor, 1996; Stansfeld et al., 1995; Heinisch &Jex, 1998). However, there are ways inwhich the design of assessment instru-ments and procedures can contribute toensuring that the data obtained are ofgood quality. It is clear that an assessmentrelying solely on appraisal would representvery weak evidence, and would need to besupported by data from other domains. Tri-

4.2S E L F - R E P O R T D A T A A N D

T R I A N G U L A T I O N

Since the most readily available data onpsychosocial and organisational hazards ofwork are usually the self-reports of thoseinvolved in the work under consideration,eliciting and modelling the knowledge andperceptions of employees is central to theassessment process. Despite their obviouscentrality and importance, self-reportmeasures of appraisal and the emotionalexperience of stress are, on their own, in-sufficient. While their reliability can be es-tablished in terms of their internalstructure or performance over time with-out reference to other data, their validitycannot.

The validity of self-report data has beenquestioned in particular with regard to theissue of “negative affectivity” (e.g.Heinisch & Jex, 1998; Kristensen, 1996;Beehr, 1995; Sheffield et al. 1994; Frese &Zapf, 1988). Negative affectivity (NA) can


■58

angulation10 of evidence overcomes the po-tential problems of NA to some extent (Jick,1979; Cox & Ferguson, 1994). The princi-ple of triangulation holds that, to be se-cure, a potential psychosocial ororganisational hazard must be identified bycross-reference to at least three differenttypes of evidence. The degree of agree-ment between those different points ofview provides some indication of the relia-bility of the data and, depending on themeasures used, their concurrent validity.Applying this principle would require datato be collected from at least three differentdomains (Cox, 1990). This can be achievedby considering evidence relating to [1] theobjective and subjective antecedents of theperson’s experience of stress, [2] their self-report of stress, and [3] any changes in theirbehaviour, physiology or health status11

which might be correlated with [1] and/or[2]. The influence of moderating factors,such as individual and group differences(see section 3.6), may also be assessed.

Several authors have recommended mea-surement strategies that are consistentwith the concept of triangulation. For ex-ample, Kristensen (1996) proposes a “3-Smatrix” which would apply the principlesof triangulation to the three main ele-ments of the ‘stress equation’ (stressors,stress and sickness). Bailey & Bhagat(1987) have recommended a multi-

method approach to the measurement ofstress. They have argued in favour of bal-ancing the evidence from self-report, phys-iological and unobtrusive measures. Theirunobtrusive measures relate to what Fol-ger & Belew (1985) and Webb et al. (1966)have called non-reactive measures, and in-clude: physical traces (such as poor housekeeping), archival data (such as that on ab-senteeism), private records (such as di-aries), and non intrusive observation andrecordings. Bailey & Bhagat (1987) alsopoint up the problem that obtrusive mea-sures often change the very nature of thebehaviour or other response being as-sessed. It is also necessary to devise stan-dardised procedures for the corroboration,or otherwise, of qualitative data withquantitative measures, and between setsof qualitative data from different sources.

Confidence on the validity of the data thusobtained is supported by various studieswhich have shown that there is good con-vergence between self-report and supervi-sor- and subordinate-report (e.g. Bosma &Marmot, 1997; Jex & Spector, 1996; Spec-tor et al., 1988).

Triangulation would require evidencedrawn from an audit of the work environ-ment (including both its physical and itspsychosocial aspects: see sections 5.1 and5.2), from a survey of workers’ perceptionsof and reactions to work, from the mea-surement of workers’ behaviour in respectto work, and their physiological and healthstatus (see section 6). It is not possible hereto offer a comprehensive review of theplethora of measures which might be usedin such audits and surveys. However, sec-tion 5 suggests the various physical and

10 The concept of triangulation in measurement relatesto the strategy of fixing a particular position or findingby examining it from at least three different points ofview.

11 The changes in behaviour, physiology and healthstatus which may be correlated with the antecedentsand/or experience of stress are discussed in section 6.


59■

psychosocial antecedents of stress thatmight be measured in the workplace,while the measurement of the stress statehas been outlined above. Measures of thethird domain (behaviour, physiology andhealth status) are well established in thegeneral literature on occupational psychol-ogy and psychophysiology. The use of anymeasure must be supported by data relat-ing to its reliability and validity, and its ap-propriateness and fairness in the situationin which it is being used. The provision ofsuch data would conform to good practicein both occupational psychology and psy-chometrics (e.g., Cox & Ferguson, 1994),but may also be required if any subsequentdecisions are challenged in law. Preferablysuch data collection would take the formof continuous monitoring and thus be ca-pable of mapping work-related changes inall three domains.

Ideally, the principle of triangulationshould be applied both within and be-tween domains. This should help over-come the problem of missing data andhelp resolve inconsistencies in the data giv-en that these are not extreme. Its use be-tween domains has been briefly discussedabove. Within domains, several differentmeasures should be taken and preferablyacross different measurement modalitiesto avoid problems of common methodvariance. This may be most relevant andeasiest to achieve in relation to the mea-surement of changes in the third domain:behaviour, physiology and health status.There are no available studies to suggestthat the various measures from the differ-ent domains can be statistically combinedinto a single and defensible ‘stress index’.

It needs to be emphasised that what is be-ing measured is a process: antecedents –perceptions and experience (and moderat-ing factors) – immediate outcomes –health status. This approach underlinesboth the complexity of measurement,when approached scientifically, and the in-adequacy of asking for or using single one-off measures of stress (however defined).This process can be simplified conceptual-ly to ‘[work] hazards – stress – harm’, andthis is the framework used to structure theevidence relating to work stress and healthin the following sections of this Status Re-port. This has the practical advantage ofplacing the issue of occupational stresswithin a framework familiar to thoseworking with health and safety problemsand consistent with recent European legis-lation. The following sections of this Re-port consider work hazards and stress(section 5), and stress and health (section6).


■60

viewed a number of “triangulation”strategies that researchers have adoptedto that end, and has highlighted the needto develop standardised procedures for thecorroboration of qualitative data withquantitative measures, and between setsof qualitative data from different sources.Finally, it must be noted that the conceptsof process and interaction have importantimplications for the operationalisation ofstress theory: the measurement of the“stress process” is, when approached sci-entifically, unavoidably complex and notadequately addressed by single one-offmeasures. The following sections of thisReport describe a framework for the as-sessment and management of work-relat-ed stress that aims to reflect the dynamicnature of the process.

4.3S U M M A R Y

Previous sections have examined the evi-dence that supports the transactionalmodels of stress as the most adequate anduseful of those currently available. Withinthis framework, work-related stress is de-fined as a psychological state that is bothpart of and reflects a wider process of in-teraction between the person and theirwork environment. It follows that the mea-surement of stress must be based primari-ly on self-report measures which focus onthe appraisal process, the emotional expe-rience and the person-environment inter-action. However, such self-reportmeasures are, on their own, insufficient,and there has been much debate on themethodological problems posed by “neg-ative affectivity” and common methodvariance. The existing literature has identi-fied the need to establish the validity ofself-report data with reference to addition-al, external evidence. This section has re-


61■

W O R K H A Z A R D S A N D S T R E S S

In line with both the scientific literature andcurrent legislation, this Report considers theevidence relating to all work hazards. Thesecan be broadly divided into physical hazards(section 5.1), which include the biological,biomechanical, chemical and radiological,and psychosocial hazards (section 5.2).

The International Labour Organization(1986) has defined psychosocial hazards interms of the interactions among job con-tent, work organisation and management,environmental and organisational condi-tions, on the one hand, and the employees’

competencies and needs on the other.Those interactions which prove hazardousinfluence employees’ health through theirperceptions and experience (InternationalLabour Organization, 1986). While this de-finition is consistent with transactionalmodels of stress, it associates exposure topsychosocial hazards too strongly with theexperience of stress. It is argued here thatpsychosocial hazards may also have directeffects on the person, effects which are notmediated by the experience of stress. As aresult, a more satisfactory definition of psy-chosocial hazards might be “those aspectsof work design and the organisation andmanagement of work, and their social andenvironmental contexts, which have thepotential for causing psychological, socialor physical harm” (Cox & Griffiths, 1995).

Exposure to physical and psychosocial haz-ards may affect psychological as well asphysical health. The evidence suggeststhat such effects on health may be medi-ated by, at least, two processes (see Figure3 below): first, a direct physical pathway,and second, a psychological stress-mediat-ed pathway. These two mechanisms donot offer alternative explanations of thehazard-health association; in most haz-ardous situations both operate and inter-act to varying extents and in various ways(Levi, 1984; Cox & Cox, 1993). Levi (1984)has noted that both additive and synergis-tic interactions12 are possible.

RE

SE

AR

CH

5.12 The outcome of effects that interact additively issimply the sum of the separate effects; however, theoutcome of effects that interact synergistically is otherthan the sum of the separate effects. It may be greater,where one set of effects facilitates or enhances anoth-er, or it may be smaller, where one set attenuates orweakens another.


■62

Many of the existing discussions of thehazard-stress-health relationship have fo-cused on psychosocial hazards and havetended to omit reference to physical workhazards (Levi, 1984). The psychological ef-fects of physical hazards reflect not onlytheir direct action on the brain and theirunpleasantness but also the worker’sawareness, suspicion or fear that they are

being exposed to harm. It is the latterwhich can give rise to the experience ofstress. For example, exposure to organicsolvents may have a psychological effecton the worker through their direct effectson the brain, through the unpleasantnessof their smell and through the worker’sfear that such exposure might be harmful(Levi, 1981; Kasl, 1992). Such fear may

F i g u r e 3 : T h e d u a l p a t h w a y h a z a r d – h a r m

SOCIAL & ORGANISATIONAL CONTEXT

DESIGN & MANAGEMENT OF WORK

PHYSICAL WORKENVIRONMENT

PSYCHOSOCIALWORK

ENVIRONMENTS

HARM TO EMPLOYEES’ PHYSICALPSYCHOLOGICAL & SOCIAL HEALTH

EMPLOYEES’ AVAILABILITY FOR &PERFORMANCE AT WORK

HEALTHINESS & PERFORMANCE OF THEORGANISATION

EXPERIENCE OFSTRESS

Directphysicalpathway

Indirectstress

pathway


63■

have consequences for task performanceas well as for health 13. The psychologicaleffects of similar agents, carcinogens andtoxic materials, appear dependent on theinformation available to and the aware-ness of workers (Houts & MacDougall,1988). The prevalence of stressful physicalenvironments cannot be ignored, and ithas been reported to be on the increaseacross the EU between 1991 and 1996(European Foundation, 1996).

Concern for occupational stress thereforefocuses on two scenarios: first, the stressassociated with exposure to the physicalhazards of work (section 5.1), and, sec-ond, the stress which arises from exposureto psychosocial hazards (section 5.2).

13 While low levels of anxiety and fear may have a mo-tivating quality, higher levels can impair task perfor-mance (see, for example, M. Eysenck, 1983; Idzikowski& Baddeley, 1983) as well as impairing the quality oflife. Deterioration in performance can be expected intasks involving manual dexterity and sensory-motor co-ordination, such as tracking, in complex cognitive tasksand in secondary tasks. Some of these effects may bemediated by impairments of memory processes, andsome by an increase in task-irrelevant and intrusivethoughts. The performance effects of anxiety and fearmay increase with task difficulty. Deterioration in sec-ondary task performance is likely to occur before per-formance of the primary task is affected. Baddeley(1972) has suggested that dangerous situations whichare emotionally arousing may affect performance by anarrowing of attention which may cause peripheralstimuli to be missed. Together these different effectsmay interfere with the safeness of working practices.


■64

1998; Kryter, 1972; Kasl, 1992). Smith(1991) suggests that “the (non auditory)health effects of noise may often reflectpsychological reactions to the noise -stress-as well as objective exposure levels”. Highlevels of noise directly damage the middleand inner ears with consequent impair-ment of hearing (Jones, 1983). Less severenoise may interfere with speech perceptionand communication (Jones, 1999) and,particularly if it is prolonged, may give riseto the experience of stress, and to anxiety,irritability and tension, increase fatigue andimpair performance efficiency (see, for ex-ample, Cohen, 1969, 1974; Barreto et al.,1997; Glass & Singer, 1972; Miller, 1974;Cohen, 1980; Ahasan et al., 1999). How-ever, Jones (1983) has concluded that evi-dence of the relationship between noiseand psychological and physical health (be-yond damage to the ear and hearing im-pairment) is equivocal: while health effectshave been found in a number of studies,they cannot be unequivocally linked to ex-posure to high levels of noise. He arguesthat in most studies the effects of noise areconfounded with those of other hazards:noisy work is often hazardous in other re-spects. While such arguments are valid,they do need to be placed in perspectivegiven the complexity of all work design andthe availability of other data (e.g., Land-strom et al., 1995). Smith (1991) has con-cluded that there is considerable evidencethat acute noise exposure produces physi-ological responses which, if prolonged,could have harmful effects on health. Hehas also argued that the available epidemi-ological data suggest that noise is a riskfactor for health. Furthermore, interven-tion and epidemiological studies suggest

5.1P H Y S I C A L H A Z A R D S

A wide variety of physical hazards havebeen extensively studied for their effectson the psychological experience of stressand on health (see, for example, Gobel etal., 1998; United States Department ofHealth, 1980; Holt, 1982; Neale et al.,1983). Most can be measured objectively,and with some degree of reliability and va-lidity, and are therefore relatively easilymonitored in the workplace. In some cas-es, standards exist which can be used inthe regulation of exposure to these poten-tial sources of harm. Particular attentionhas been paid to noise as a source of stressand threat to health (Holt, 1982), and thisis taken here as an exemplar of physicalhazards.

5 . 1 . 1 N o i s e

Noise can act as a physical and a psycho-logical stimulus (Akerstedt & Landstrom,


65■

that noise can have harmful effects onhealth (see, for example, Cohen, 1976;Wallhagen et al., 1997). As with most oc-cupational health issues, it is a case of inte-grating different types of evidence inreaching a balanced conclusion.

Cohen (1974) examined the effects ofnoise on absence from work due to illness,on accidents and on diagnosed medicalproblems over a five-year period in twomajor plants. Data were collected fromplant records. One plant manufacturedlarge boilers and the other manufacturedelectronic missile and weapon compo-nents. Workers drawn from high noise ar-eas (95 dBA or more) were compared toworkers drawn from low noise areas (80dBA or less). Those from the high noise ar-eas exhibited a higher incidence of prob-lems on all measures. Especially prevalentin those exposed to high noise were aller-gies, respiratory and gastrointestinal disor-ders and complaints associated withmusculo-skeletal and cardiovascular condi-tions. However, larger differences in the in-cidence of these problems appeared whenthey were compared by job type (ratherthan noise), and although attempts weremade to control for job type in the analysisof noise effects these were not entirelysuccessful (Jones, 1983). If noise was ofaetiological significance, then its effectsappeared to be less than –or secondary to–those of job design and work organisation.However, the noise effects were not in-significant and a follow up study by Cohen(1976) found evidence of a reduction inaccident rate and incidence of medicalproblems as a result of introducing ear de-fenders.

5 . 1 . 2 O t h e r P h y s i c a l F a c t o r s

Overall, there is much evidence to suggestthat poor physical working conditions, ingeneral, can affect both workers’ experi-ence of stress and their psychological andphysical health (Warr, 1992). However,there are few studies which directly estab-lish the hazard-stress-harm pathway. Somestudies have suggested that the effects ofphysical hazards on the experience ofstress and on health are not related. Alt-house & Hurrell (1977), for example, com-pared 486 coal miners in the United Stateswith 452 workers in jobs of similar status.Despite a difference in the levels of physi-cal dangerousness of the two types ofwork (exposure of workers to possible in-jury and death), there were no differencesin experience of stress although miners didreport significantly more symptoms of illhealth such as irritation and somatic com-plaints.

In the case of some hazards, such as tem-perature and humidity (Biersner et al.,1971), it is the extremes of physical workconditions which are associated with theexperience of stress and with effects onhealth: workers are often able to adapt tomid-range conditions without effort or at-tention (Holt, 1982; Szabo et al., 1983). Inthe case of others it is more simply thepresence of the hazard or even the per-ceived threat of its presence which is asso-ciated with the experience of stress. Anexample is provided by doctors’ and nurs-es’ reports of anxiety in relation to dealingwith patients who might be infected withthe human immunodeficiency virus (Kege-les et al., 1989; Cox et al., 1993). Physicalhazards not only interact with one and an-


■66

other in producing their effects, but mayalso interact with psychosocial hazards(e.g., Melamed et al., 1999; Schrijvers etal., 1998). Broadbent (1971) has describedhow noise and sleep loss might interact inrelation to task performance, while there isother evidence that exposure to poorequipment and work station design, inconjunction with poor task design andwork organisation give rise to work-relat-ed upper limb disorders (Chatterjee, 1987,1992: Health & Safety Executive, 1990a).


67■

chological or physical harm” (Cox & Grif-fiths, 1995). There is now a large body ofevidence (e.g., Cox, 1993; Landy et al.,1994; Kasl, 1987 & 1990) that identifies acommon set of work characteristics as po-tentially hazardous (see Table 1).

Psychosocial hazards may affect both psy-chological and physical health directly orindirectly through the experience of stress(see Figure 3). Most attention has beenpaid to their possible indirect, stress-medi-ated effects. It is this literature which is re-viewed below.

Work situations are experienced as stress-ful when they are perceived as involvingimportant work demands which are notwell matched to the knowledge and skills(competencies) of workers or their needs,especially when those workers have littlecontrol over work and receive little supportat work (see section 5.2.1). Levi (1984) hasgrouped the various psychosocial charac-teristics of work under four headingswhich can be derived from this model:quantitative overload, qualitative under-load, lack of control over work and lack ofsocial support. Each aspect of such worksituations carries a potential for harm andthus represents a hazard. These are thefundamental dimensions of psychosocialhazards in that they underpin the person’sperception of the stressfulness of any worksituation. They may, however, find ‘sur-face’ expression and combine in differentways for different hazards depending onthe type of work and work environment.

There is a reasonable consensus amongthe various attempts to review the litera-ture on those psychosocial hazards of

5.2P S Y C H O S O C I A L H A Z A R D S

The psychological aspects of work havebeen the subject of research since at leastthe 1950s (Johnson, 1996; Sauter et al.,1998). Initially psychologists concentratedmostly on the obstacles to employees’adaptation and adjustment to the workenvironment, rather than on the potential-ly hazardous characteristics the workplaceitself may have for workers (Gardell,1982). However, with the emergence ofpsychosocial work environment researchand occupational psychology in the 1960s(Johnson & Hall, 1996) the focus of inter-est has moved away from an individualperspective and towards considering theimpact of certain aspects of the work envi-ronment on health. As suggested earlier,“psychosocial hazards” can be defined as“those aspects of work design and the or-ganisation and management of work, andtheir social and environmental contexts,which have the potential for causing psy-


■68

work which are experienced as stressfuland/or otherwise carry the potential forharm (Baker, 1985; Blohmke & Reimer,1980; Cooper & Marshall, 1976; Cox,1978, 1985b; Cox & Cox, 1993; Franken-hauser & Gardell, 1976; Karasek & Theo-rell, 1990; Kasl, 1992; Levi, 1972, 1984;Levi et al., 1986; Loher et al., 1985; Mar-mot & Madge, 1987; National Institute,1988; Sauter et al., 1992; Sharit & Sal-vendy, 1982; Szabo et al., 1983; Warr,1987, 1992). This consensus is sum-marised in , which outlines ten different

categories of job characteristics, work en-vironments and organisations which maybe hazardous. It has been suggested(Hacker, 1991; Hacker et al., 1983) thatsuch characteristics of work might be use-fully conceived as relating to the context towork or the content of work. Under cer-tain conditions each of these ten aspectsof work has proved stressful and harmfulto health: these conditions are also notedin . Much of the evidence relates to psy-chological health and to the risk of cardio-vascular disease (see section 6).

T a b l e 1 : S t r e s s f u l C h a r a c t e r i s t i c s o f W o r k

Category Conditions defining hazard

Context to work

Organisational culture and function Poor communication, low levels of support for problem-solvingand personal development, lack of definition of organisationalobjectives.

Role in organisation Role ambiguity and role conflict, responsibility for people.

Career development Career stagnation and uncertainty, underpromotion or overpro-motion, poor pay, job insecurity, low social value to work.

Decision latitude / Control Low participation in decision making, lack of control over work(control, particularly in the form of participation, is also a contextand wider organisational issue)

Interpersonal relationships at work Social or physical isolation, poor relationships with superiors, in-terpersonal conflict, lack of social support.

Home-work interface Conflicting demands of work and home, low support at home,dual career problems.

Content of work

Work environment and work equipment Problems regarding the reliability, availability, suitability andmaintenance or repair of both equipment and facilities.

Task design Lack of variety or short work cycles, fragmented or meaninglesswork, underuse of skills, high uncertainty.

Workload / workpace Work overload or underload, lack of control over pacing, highlevels of time pressure.

Work schedule Shift working, inflexible work schedules, unpredictable hours,long or unsocial hours.


69■

5 . 2 . 1 C o n t e x t t o W o r k

The following section describes those psy-chosocial hazards which are related to the“context to work” and which are experi-enced as stressful and/or otherwise carrythe potential for harm.

O r g a n i s a t i o n a l C u l t u r e a n dF u n c t i o n

The very fact of working within an organi-sation, as do most workers in Europe (Coxet al., 1990), can be perceived as a threatto individual freedom, autonomy andidentity (Hingley & Cooper, 1986). Studieson employees’ perceptions and descrip-tions of their organisations suggest thatthese revolve around three distinct aspectsof organisational function and culture: theorganisation as a task environment, as aproblem-solving environment and as a de-velopment environment (Cox & Howarth,1990; Cox & Leiter, 1992). The availableevidence suggests that if the organisationis perceived to be poor in respect to theseenvironments, then this is likely to be asso-ciated with increased levels of stress. Onthe other hand, if the organisation is per-ceived to be good in these respects thenthe relationship between the experience ofstress and the report of symptoms of illhealth is attenuated (Cox & Kuk, 1991).

Kasl (1992) has listed various aspects of or-ganisation which he believes may be haz-ardous; for example, organisational sizeand structure (having a flat structure withrelatively few levels), cumbersome and ar-bitrary procedures, and role-related issues.The latter are dealt with below. Much ofthe effect of organisation and function

and culture on workers will be transmittedthrough the behaviour of managers andsupervisors. There is evidence, for exam-ple, that management behaviour and su-pervisory styles have a substantial impacton the emotional well-being of workers(Landy, 1992; Corey & Wolf, 1992). Suchan influence may be partly a reflection oftheir handling of the job context and jobcontent issues listed in . Following this ar-gument, any effect of style might be large-ly a reflection of more general issues ofinterpersonal relationships.

R o l e i n O r g a n i s a t i o n

The evidence that ‘role in organisation’ is apotential psychosocial hazard relates largelyto issues of role ambiguity and role conflict(Kahn et al., 1964; Kahn, 1973; Ingersoll etal., 1999; Jackson & Schuler, 1985). How-ever, other potentially hazardous aspects ofrole have been identified including roleoverload, role insufficiency and responsibili-ty for other people (see below). French et al.(1982) have concluded that such variablesare among the most powerful predictors ofpsychological health. Measures of all five as-pects of role were used in a study of white-collar workers by Bhalla et al. (1991). Theywere related to workers’ reports of strain,job satisfaction and organisational commit-ment. The data suggested that overall roleambiguity, role conflict and role insufficien-cy were more strongly related to the out-come variables than were role overload orresponsibility for other people.

R o l e A m b i g u i t y

Role ambiguity occurs when a worker hasinadequate information about his or her


■70

work role. As Warshaw (1979) has stated,“the individual just doesn’t know how heor she fits into the organisation and is un-sure of any rewards no matter how well heor she may perform.” A wide range ofevents can create role ambiguity, many ofthem relating to novel situations andchange (see Ivancevich & Matteson, 1980).

Role ambiguity manifests itself in a generalconfusion about appropriate objectives, alack of clarity regarding expectations, and ageneral uncertainty about the scope andresponsibilities of the job. Kahn et al.(1964) found that workers who sufferedfrom role ambiguity were more likely to ex-perience lower job satisfaction, a greaterincidence of job-related tension, greaterfeelings of futility and lower levels of self-confidence. French & Caplan (1970) foundthat role ambiguity was related to a similarcluster of symptoms. They also showedthat role ambiguity was related to in-creased blood pressure and higher pulserates. Later research by Margolis et al.(1974) found a number of significant rela-tionships between role ambiguity andsymptoms of depression and low job moti-vation and intention to leave the job.Cooper and Marshall (1976) have pointedout that although the correlations reportedin all these studies were significant and to-gether paint a consistent picture, they werenot particularly strong (only accounting forabout 2-5% of the data variance). Further-more, many of the measures of ill healthwere based on self-report (see section 4.2).

R o l e C o n f l i c t

Role conflict occurs when the individual isrequired to play a role which conflicts with

their values, or when the various roles thatthey play are incompatible with one an-other. Kahn and his colleagues (1964) haveshown that the greater role conflict inmen, the lower job satisfaction and thegreater job-related tension. French & Ca-plan (1970) found that mean heart ratewas strongly related to perceived level ofrole conflict. It may also be related to in-creased risk of cardiovascular ill health(Ivancevich & Matteson, 1980). For exam-ple, Shirom et al. (1973), in a large study ofIsraeli men drawn from a range of occupa-tions, found that there was a significant re-lationship between role conflict andincidence of coronary heart disease butonly for white-collar workers. Cooper &Smith (1986) concluded that white-collarworkers are more prone to role conflictthan are manual workers.

Kahn et al. (1964) have suggested thatthose in ‘boundary roles’ (links betweenorganisational levels or departments), suchas foremen, are particularly prone to expe-rience stress. Such roles have a high po-tential for conflict, and Margolis & Kroes(1974) found that foremen were seventimes more likely to develop ulcers thanshop floor workers.

R o l e I n s u f f i c i e n c y

Role insufficiency refers to a failure of theorganisation to make full use of the indi-vidual’s abilities and training (for example,O’Brien, 1982). Such insufficiency hasbeen reported to lead to feelings of stress(Brook, 1973) and is associated with psy-chological strain and low job satisfactionand organisational commitment (Bhalla etal., 1991).


71■

R e s p o n s i b i l i t y f o r P e o p l e

Responsibility for people has been identi-fied as a potential source of stress associ-ated with role issues. Wardell et al. (1964)showed that responsibility for people,compared to responsibility for things, waslikely to lead to greater risk of coronaryheart disease. French & Caplan (1970)found that responsibility for people wassignificantly related to heavy smoking,raised diastolic blood pressure and elevat-ed serum cholesterol levels. The literatureon burn out (e.g., Leiter, 1991) also sug-gests that, in the caring professions atleast, responsibility for people is associatedwith emotional exhaustion and the deper-sonalisation of relationships with patients.There is also evidence from the study ofmental health referrals, by occupation,that those occupations involving continualcontact with and responsibility for peopleare high risk (Colligan et al., 1977).

C a r e e r D e v e l o p m e n t

The lack of expected career developmentmay be a source of stress, particularly in or-ganisations which emphasise the relation-ship between career development andcompetence or worth. Marshall (1977)identified two major clusters of potentialsources of stress in this area: first, lack ofjob security and obsolescence (fear of re-dundancy and forced early retirement);and, second, status incongruity (under orover promotion, and frustration at havingreached the career ceiling). These havebeen related to adverse psychological ef-fects as well as poor physical health (Kasl &Cobb, 1982; Margolis et al., 1974) and arediscussed below. These two sources of

stress probably interact. Cooper (1978) hassuggested that fear of obsolescence andfailure resulting in demotion is likely to bestrongest in those who believe they havereached their career ceiling, and that mostwill experience some erosion of status be-fore they retire. Roberston & Cooper(1983) believe that these fears may giverise to stress if workers are unable to adapttheir expectations to the reality of their sit-uation. Not surprisingly, older workers areparticularly vulnerable as they tend toplace a high value on stability (Sleeper,1975).

J o b I n s e c u r i t y a n d P o o r P a y

Job insecurity and fear of redundancy canbe major sources of anxiety, particularly iforganisations expect, at the same time,commitment from their employees. Thesense of inequity may exacerbate the ex-perience of stress (Porter, 1990). Poor paymay be hazardous to health. While mostworkers will complain about levels of pay,the extremes of poor pay clearly have aneffect on the worker’s ability to remainhealthy (Warr, 1992). Method or scheduleof payment may also be a source of stress(for example, piece work) and may interactin its effects with the rate of working (Kasl,1992).

S t a t u s I n c o n g r u i t y

The cost of status incongruity has beenwell researched in the United States. Forexample, Arthur & Gunderson (1965), in astudy of naval personnel, claimed that pro-motional lag was significantly related topsychiatric illness. Interestingly, the litera-ture on status incongruity also suggests a


■72

strong effect of non-work factors. For ex-ample, Kasl & Cobb (1967) concluded thatstress related to parental status had‘strong long term effects on physical andmental health of adult offspring’. Shekelleet al. (1969) found that their sample ofmen in the United States whose presentsocial class was substantially different fromthat of their childhood ran a significantlyhigher risk of coronary heart disease thanmen whose present social class was not.

D e c i s i o n L a t i t u d e a n d C o n t r o l

Decision latitude and control are importantissues in job design and work organisation.They are often reflected in the extent towhich employees can participate in deci-sion-making affecting their work. Howev-er, there are other aspects to participationsuch as status which may also affect healthand behaviour.

The experience of low control at work orof loss of control –low decision latitude–has been repeatedly associated with theexperience of stress, and with anxiety, de-pression, apathy and exhaustion, low selfesteem and increased incidence of cardio-vascular symptoms (Terry & Jimmieson,1999; Ganster & Fusilier, 1989; Sauter etal., 1989; Karasek & Theorell, 1990). Inter-estingly, in a study of 244 occupations inSweden, men consistently reported higherlevels of control than women, even withinfemale stereotyped jobs (Hall, 1991).

Following on from the work of Karasek,among others, it is often implied that in-creasing workers’ control is universallybeneficial. For example, Cox (1990) andWarr (1992) have argued that workers

should, ideally, be empowered to plantheir work, and control their workloads,make decisions about how that workshould be completed and how problemsshould be tackled. However, it has been ar-gued by Neufeld & Paterson, (1989) thatcontrol can also be a double-edged sword:the demands implied by the choices in-volved in controlling situations can them-selves be a source of stress.

P a r t i c i p a t i o n

Research suggests that where there aregreater opportunities for participating indecision-making, greater satisfaction andhigher feelings of self-esteem are reported(French & Caplan, 1970, 1972; Buck,1972; Margolis et al. 1974; Spector, 1986).Non-participation appears related to work-related stress and overall poor physicalhealth (Margolis & Kroes, 1974). French etal. (1982) have reported that lack of par-ticipation shows a strong relationship tojob dissatisfaction but that this effect maybe mediated by other variables relating tothe overall person-environment fit.

I n t e r p e r s o n a l R e l a t i o n s h i p s a tW o r k

It has been argued strongly that good rela-tionships amongst workers and membersof work groups are essential for both indi-vidual and organisational health (Cooper,1981). A survey by the Ministry of Labourin Japan (1987) revealed that 52% of thewomen interviewed had experienced anxi-ety and stress, the main cause being un-satisfactory interpersonal relations at work(61%). Similarly, Jones et al. (1998) foundthat workers reporting high levels of stress


73■

and stress-related illnesses were 61/2 timesmore likely to report “lack of support frompeople in charge at work” than the gener-al working population.

Three important sets of relationships havebeen identified: relationships with superi-ors, relationships with subordinates andrelationships with colleagues (Sauter et al.,1992). Low interpersonal support at workhas been found to be associated with highanxiety, emotional exhaustion, job tensionand low job satisfaction and increased riskof cardiovascular disease (for example,Beehr & Newman, 1978; Davidson &Cooper, 1981; Pearse, 1977; Warr, 1992).

Social relationships both at work and out-side the workplace are most commonlyviewed as playing a moderating role, andadverse effects of exposure to other psy-chosocial hazards are more likely or morepronounced when relationships provide lit-tle support (Cobb & Kasl, 1977; Cohen &Willis, 1985; House & Wells, 1978).Karasek and colleagues (1982) in a studyof over 1,000 male workers in Sweden,showed that support from supervisors andco-workers buffered the effects of job de-mands on depression and job satisfaction.

However, other research suggests a moredirect effect of social support in offsettingthe adverse effects of working conditions(Ganster et al., 1986). In a recent meta-an-alytical study of 68 previous papers,Viswesvaran et al. (1999) confirmed thepresence of three general constructs (stres-sors, strains and social support). Their re-sults indicated that social support had athreefold effect on work stressor-strain re-lations: it reduced the strains experienced,

mitigated perceived stressors, and moder-ated the stressor-strain relationship. Lob-ban et al. (1998) found that supervisorystyles (in terms of providing direction andcommunicating with employees) may playa more dominant role in the stress processthan is currently appreciated. They alsosuggest that supervisory relationships, ei-ther directly or mediated by other job char-acteristics, have significant additionalinfluence on occupational stress that can-not be explained by the role ordemand/latitude variables. Fielden &Peckar (1999) found that, although thereis a direct link between the number ofhours worked and stress levels, the num-ber of hours worked was positively relatedto the perceived availability of social sup-port.

Buck (1972) has reported that the ‘consid-erate’ behaviour of superiors appears tocontribute inversely to workers’ feelings ofjob pressure. Workers’ participation in de-cision making results in them reportinggreater job satisfaction and stronger feel-ings of self-esteem (Buck, 1972; French &Caplan, 1970, 1972; Margolis et al.,1974). However, Donaldson & Gowler(1975) consider that pressure on managersto ‘manage by participation’ actuallyplaces them under increased pressure, andmay cause feelings of resentment and anx-iety. Robertson & Cooper (1983) discusshow competition at work, particularlyamong managers, may inhibit problemsharing and increase stress.

V i o l e n c e a t W o r k

There is growing literature on violence inthe workplace (Cox & Leather, 1994; Beale


■74

et al., 1998, Beale et al., 1999; Leather etal., 1998; Chappell & Di Martino, 1998;see also Standing & Nicolini, 1997, for arecent review) and on the related issue ofpost traumatic stress disorder (see Figley,1985; Simon, 1999, for a review). There isstrong evidence that exposure to violencein the workplace can cause damage topsychological as well as physical health14

(Leather et al., 1999). This is an area of in-creasing concern within the EU because,despite problems of definition across thedifferent EU cultures, violence at work is agrowing problem: 3 million workers re-ported being subjected to sexual harass-ment, 6 million to physical violence, and12 million to intimidation and psychologi-cal violence (European Foundation, 1996).As a result of this concern, the EuropeanCommission (DG V) has published guid-ance on the prevention of violence at work(Wynne et al., 1997).

H o m e - W o r k I n t e r f a c e

The concept of the work-home interface(or “work-home interference”, WHI) re-lates not only to domestic life and the fam-ily but also to the broader domain of lifeoutside of work. Most research has fo-

cused either on the relationship betweenmanagers and their spouses (Cooper,1981) or on the use of leisure time(Gardell, 1973; Cox, 1980).

W o r k a n d F a m i l y

Hingley & Cooper (1986) have argued thatproblems relating to the interface betweenwork and the family either involve resolv-ing conflicts of demands on time and com-mitment, or revolve around issues ofsupport. Much of the former literature fo-cuses on women workers (see, for exam-ple, Davidson & Cooper, 1983) althoughcommentary has been offered on men andparticularly young managers (Weinberg etal., 1999; Beattie et al., 1974; Geurts et al.,1999). The difficulties faced in resolvingconflicts between work and family appearenhanced if the family has young children;again this may be particularly so forwomen workers (Larwood & Wood, 1979;Bhagat & Chassie, 1981). Early researchsuggested that most middle class wivesappear to see their role, in relation to theirhusbands’ job, as primarily ‘supportive anddomestic’ (Pahl & Pahl, 1971). Some yearslater Cooper & Hingley (1985) found asimilar pattern in the wives of their sampleof executive men in the United Kingdom,although attitudes appeared to be chang-ing. Failure to resolve adequately the con-flicting demands between work and familymay damage the support available fromspouses, in particular, and the family ingeneral.

Handy (1975) has explored the nature of anumber of possible ‘marriage-role’ combi-nations in a study of executive managers.Consistent with other research, the most

14 There are three main issues here: first, the accuraterecording of data on violence at work and its aftermathso that an informed judgement can be made on thesize of the problem; second, understanding the natureof such violence both from an individual and an or-ganisational perspective; and third the developmentand evaluation of interventions designed to reduce thelikelihood of violence occurring or reduce its impact onstaff (Leather et al., 1999). Recommendations on themanagement of violence at work have been publishedby a variety of bodies, including the British Health andSafety Executive (Mackay, 1987), the British Psycholog-ical Society (1992) and the Tavistock Institute of HumanRelations (1986).


75■

frequent combination was the “thrustingmale–caring female”, which was mostbeneficial to the working husband. Anoth-er increasingly common combination waswhat was effectively the dual career cou-ple. In this combination, traditional role ex-pectations appear to be challenged withthe possibility of either or both partnersexperiencing feelings of threat and anxiety(Hingley & Cooper, 1986).

W a s t e d L e i s u r e T i m e S y n d r o m e

Spill-over effects from work might ac-count for the possible wasting of con-structive leisure time among some groupsof employees (Gardell, 1973; Cox, 1980).The ‘wasted leisure time syndrome’ hasbeen described by Gardell (1973) in termsof employees not finding time out of workto do more than potter about the home,skim through newspapers, watch televi-sion, and eat and sleep. Lundahl (1971)had observed in her Swedish study thatthose employed on heavy fatiguing jobsshowed less involvement in leisure thanthose who were not. Both Gardell (1973)and Cox (1980) have suggested that moreis involved than physical exertion, and thelatter author has linked wasted leisuretime to employees’ psychological and be-havioural adaptation to the demands ofshort cycle repetitive work. Wilensky(1960) has offered an explanation of thework-leisure relationship in terms of com-pensation, and this concept has also beenused to account for the effects of repeti-tive work on the use of leisure time. Con-sistent with Wilensky’s (1960) hypothesis,Strauss (1974) has suggested that em-ployees can adjust to non-challengingwork by lowering their expectations,

changing their need structure and makingthe most of social opportunities on andoff the job. However, Kornhauser (1965)offered a similar explanation but with amore negative emphasis consistent withthe hypotheses of Gardell (1973) and Cox(1980). He suggested that “the unsatis-factory mental health of working peopleconsists in no small measure of theirdwarfed desires and deadened initiative,reduction of their goals and restriction oftheir efforts to a point where life is rela-tively empty and only half meaningful”.

C h a n g e

Change is often cited as a psychosocialhazard. However, it is not clear from the lit-erature whether change per se is stressfulor hazardous, or whether its possiblystressful nature is due to the uncertaintyand lack of control which it often repre-sents.

5 . 2 . 2 C o n t e n t o f W o r k

This section describes those psychosocialhazards which are related to the contentof work and which are experienced asstressful and/or otherwise carry the poten-tial for harm.

T a s k D e s i g n

There are several different aspects of jobcontent which are hazardous: these in-clude low value of work, the low use ofskills, lack of task variety and repetitivenessin work, uncertainty, lack of opportunity tolearn, high attentional demands, conflict-ing demands and insufficient resources(Kasl, 1992).


■76

S e m i - s k i l l e d a n d U n s k i l l e d W o r k

Such work is often characteristic of semi-skilled and unskilled jobs (Kornhauser,1965; Caplan et al., 1975; French et al.,1982; Smith, 1981; Salvendy & Smith,1981; Cox, 1985b). Cox (1985b) has re-viewed the physical and psychologicalhealth effects of such work. Exposure torepetitive and monotonous work is oftenassociated with the experience of bore-dom, and, in turn, with anxiety and de-pression, resentment, and generally poorpsychological health (see: Kornhauser,1965; Gardell, 1971; Laville & Teiger, 1976;Caplan et al., 1975; Broadbent & Gath,1981; O’Hanlon, 1981; Smith, 1981). Forexample, Kornhauser (1965) showed thatamong production workers in a car manu-facturing plant in the United States, under-utilisation of skill was a particularly strongpredictor of poor psychological health.There may also be an increased incidenceof postural and musculo-skeletal problems,including work-related upper limb disor-ders (see, for example, Kuorinka, 1979;Chatterjee, 1987, 1992; Health & SafetyExecutive, 1990a), disorders of the diges-tive system (Laville & Teiger, 1976; Nerell,1975) –although these disorders may beassociated with shift working in such jobs(Rutenfranz, 1982)– and various changes inhealth-related behaviours, such as smokingand drinking (Ferguson, 1973). Exposure tonoisy heavy repetitive work may also giverise to ‘wasted leisure time syndrome’(Gardell, 1973; Cox, 1980) (see above).

U n c e r t a i n t y

Uncertainty in work, in the form of lack offeedback on performance, is also a sourceof stress particularly when it extends across

a long period of time (Warr, 1992). Suchuncertainty may be expressed in ways oth-er than lack of performance feedback, andmay partly underpin the effects of otherhazardous job characteristics; for example,uncertainty about desirable behaviours(role ambiguity) and uncertainty about fu-ture (job insecurity and redundancy).

W o r k l o a d a n d W o r k P a c e

Kornhauser (1965), from his study of De-troit car workers, suggested that “poormental health was directly related to un-pleasant working conditions, the necessityto work fast and to expend a lot of physi-cal effort and to excessive and inconve-nient hours”. These various points, andothers, are dealt with below.

W o r k l o a d

Workload was one of the first aspects ofwork to receive attention (Stewart, 1976),and it has long been clear that both workoverload and work underload can be prob-lematic (Frankenhauser, 1975; Franken-hauser & Gardell, 1975; Lundberg &Forsman, 1979; Szabo et al., 1983; Joneset al., 1998). French and his colleagues,among others, have made a further dis-tinction between quantitative and qualita-tive workload (French & Caplan, 1970;French et al., 1974). Both have been asso-ciated with the experience of stress. Quan-titative workload refers to the amount ofwork to be done while qualitative work-load refers to the difficulty of that work.The two dimensions of workload are inde-pendent and it is possible to have workwhich involves quantitative overload andqualitative underload. Much short cycle


77■

repetitive assembly work is of this nature,and there is strong evidence that it offers athreat to both physical and psychologicalhealth (see above). Kahn & Byosiere (1990)have extended this line of argument bysuggesting that workload is a function ofquality, quantity and time.

Jones et al. (1998) found that workers re-porting high levels of stress and stress-re-lated illnesses were 41/2 times more likely toreport problems with “working to dead-lines” and “having too much work” thanthe general working population. Man-agers often cope with work overload byworking longer hours (Uris, 1972), and al-though this may offer a short term solutionto the immediate problem, long workinghours –if sustained– may in themselves be-come problematic (see below).

W o r k p a c e a n d T i m e U r g e n c y

Workload has to be considered in relationto work pace; that is the speed at whichwork has to be completed and the natureand control of the pacing requirement:self-, systems- or machine-paced. Withinlimits, control may be the decisive factorin determining health (Sauter et al.,1989). There is strong evidence that ma-chine- and systems-paced work, particu-larly if of high rate, is detrimental to bothpsychological and physical health (Bradley,1989; Cox, 1985a, 1985b; Smith et al.,1981; Smith, 1985). There is also evidencethat electronic performance monitoring,for computer-based work, can produce asimilar pattern of effects (see special edi-tion of Applied Ergonomics, February,1992).

Schriber & Gutek (1987) have identified anumber of temporal dimensions that canbe measured in organisational settings.Time urgency is usually treated as a prop-erty of the person (for example, in relationto type A behaviour) but it may well alsobe a property of the job. Johansson &Aronsson (1984) have suggested that VDTworkers experience more time urgency intheir work than do other occupationalgroups. Furthermore, Gael (1988) andLandy (1989), using task analysis, havedemonstrated that differences in time de-mands of tasks can be readily identifiedwith large and homogeneous samples ofindustrial workers.

W o r k S c h e d u l e

There are two main issues that relate to theeffects of work scheduling on health: shiftworking and long work hours (see, for ex-ample, Canadian Mental Health Associa-tion, 1984). Work often involves both thesefactors (see, for example, Folkard & Monk,1985; Work & Stress, 1989, special issue: 3).

S h i f t W o r k

Much of the literature relates to shift (andnight) working and has been adequatelyreviewed elsewhere (see, for example,Boggild & Knutsson, 1999; Harrington,1978; Johnson, 1981; Rutenfranz et al.,1977, 1985; Monk & Tepas, 1985; Water-house et al., 1992). Harrington (1978)concluded that “whereas good evidenceexists to show that shift work, particularlynight work, causes disruption of circadianrhythms and sleep patterns, the evidencefor there being any major effect on healthis slim.” He did, however, also conclude


■78

that there may be a link between nightwork and digestive disorders, and be-tween shift work in general and fatigue.He also commented that whatever effectsdid exist, they were likely to be greater inthose who had difficulty in adapting tosuch forms of working or who had existingdigestive or sleep related problems. Monk& Tepas (1985) reached broadly similarconclusions. In their recent study of night-shift nurses, Kobayashi et al. (1999) foundthat the cortisol and NK cell activity levelswere low during the night shift, suggest-ing that night shift work is highly stressfuland may prejudicial to biodefence.

Boggild & Knutsson (1999) reviewed 17studies dealing with shift work and cardio-vascular disease risk. They suggest thatmethodological problems are present inmost of these studies: selection bias, expo-sure classification, outcome classification,and the appropriateness of comparisongroups. Boggild & Knutsson found that,on balance, shift workers were found tohave a 40% increase in risk. Possible causalmechanisms of this risk via known cardio-vascular risk factors relate to circadianrhythms, disturbed sociotemporal pat-terns, social support, stress, health behav-iours (smoking, diet, alcohol, exercise),and biochemical changes (cholesterol,triglycerides, etc). They conclude that therisk is probably multifactorial, and that theliterature has focused on the behaviour ofshift workers, thus neglecting other possi-ble causal connections.

L o n g W o r k H o u r s

The European Community Directive onWorking Time, which should have been

implemented in Member States of the Eu-ropean Community by November 1996,contains several requirements related toworking hours, including the right of em-ployees to refuse to work more than 48hours a week. Much of the research in thisarea has focused on the problems of shift-working, emphasising this aspect of work-ing hours. However, there is much lessinformation about the effects of overtimework, which is a central element of theterms of the Directive. Research to datehas been restricted to a limited range ofhealth outcomes –namely, mental healthand cardiovascular disorders (Spurgeon etal., 1997). Other potential effects whichare normally associated with stress (for ex-ample, gastrointestinal disorders, muscu-loskeletal disorders, and problemsassociated with depression of the immunesystem) have received little attention.There have also been few systematic inves-tigations of performance effects, and littleconsideration of the implications for occu-pational exposure limits of extensions tothe working day. Existing data relate large-ly to situations where working hours ex-ceed 50 a week and there is a lack ofinformation on hours below this level,which is of direct relevance to EuropeanUnion legislation.

In their review, Spurgeon et al. (1997) con-clude that the attitudes and motivation ofthe people concerned, the job require-ments, and other aspects of the organisa-tional and cultural climate are likely toinfluence the level and nature of healthand performance outcomes. However,they also suggest that there is currentlysufficient evidence to raise concerns about


79■

the risks to health and safety of long work-ing hours. Long hours of work, from ex-tended work days of 12 hours (see, forexample, Rosa et al., 1989) to sustainedworking over several days with sleep loss(see, for example, Stampi, 1989; Patton etal., 1989; von Restorff et al., 1989), hasbeen shown to increase fatigue. Much ofthe evidence, especially in the later area,has come from studies on military workand performance.

The European Foundation’s (1996) Work-ing Conditions Report indicated that ahigh proportion of workers across the EUwork long hours15 (49% work more than40 hours per week, and 23% more than45 hours). The data also revealed thathealth problems (stress and back ache) in-creased with the hours worked. Com-pressed work weeks, with 12-hourworking days, have been associated withfeelings of increased fatigue (Rosa & Colli-gan, 1986). Rosa et al. (1989) have shownthat after seven months adaptation to a 3-4 day /12 hour rotating shift schedulethere were reductions in sleep and decre-ments in subjective alertness compared toprevious work on a 5-7 day / 8 hour sched-ule. The increases in self-reported stresswhich also occurred were attenuated bythe shortened work week.

Sustained working can cause or be other-wise associated with sleep loss and per-ceived exertion or fatigue (for example,Ryman et al., 1989). Performance can beseverely compromised by accumulation ofsleep debt (Stampi, 1989). The upper limitof human performance for working inten-

sively and continuously is 2-3 days(Haslam, 1982; Naitoh et al., 1983). Per-formance effects can be detected in vigi-lance tasks and those involving cognitiveand verbal performance (Angus & Hesle-grave, 1983; Haslam, 1982). Physical per-formance, particularly if of moderateintensity appears more resistant to impair-ment (for example, Patton et al., 1989).

Some occupational groups, such as juniordoctors, are cause for special concern. Forexample, Spurgeon & Harrington (1989)have reviewed the effects of long workinghours on the performance and health ofjunior hospital doctors. In the United King-dom, particular work rotas meant that un-til recently junior doctors were workingspells of around 102 hours. Spurgeon &Harrington (1989) concluded that a num-ber of studies have shown that a signifi-cant proportion of newly qualified doctorsdevelop some degree of psychological illhealth. They argue that this may be relatedto sleep loss which probably increases doc-tors’ vulnerability to other work hazards.The establishment of a Task Force hasbrought about significant reductions in thenumbers of hours worked by junior doc-tors, but Fielden & Peckar (1999) stillfound that direct link between the numberof hours worked and stress levels (al-though the number of hours worked waspositively related to the perceived availabil-ity of social support). Junior hospital doc-tors used social support as a copingstrategy significantly more often than se-nior hospital doctors, with both perceivingthe hospital environment as a more effec-tive source of social support than thehome environment. Despite having access15 Defined as more than 40 hours per week.


■80

to higher levels of effective social support,junior hospital doctors faced significantlygreater sources of stress and poorer men-tal health than their senior counterparts.

There is an association between longhours of work and death from coronaryheart disease Breslow & Buell (1960) foundthat individuals under 45 years of age whoworked more than 48 hours a week hadtwice the risk of death from coronary heartdisease than similar individuals whoworked 40 or fewer hours per week. An-other study of young coronary patients re-vealed that one in four had been workingat two jobs and an additional two in fivehad been working more than 60 hours aweek (Russek & Zohman, 1958).

Control over work schedules is an impor-tant factor in job design and work organi-sation. Such control may be offered byflexitime arrangements (Landy, 1989). It isinteresting to note that although the intro-duction of flexitime arrangements may beassociated with little change in behaviour(Ronen, 1981), they nonetheless can havea positive effect on workers (Narayanan &Nath, 1982; Orpen, 1981). In this case it islikely that it is the perceived control of-fered by such arrangements rather thanthe actual exercise of control that is impor-tant (Landy, 1992). Lack of control overwork schedules may represent a source ofstress to workers.

5 . 2 . 3 N e w h a z a r d s : “ T h e c h a n g i n gw o r l d o f w o r k ”

Large scale socio-economic and techno-logical changes in recent years have affect-ed workplaces considerably. They are often

collectively referred to as “the changingworld of work”. This term encompasses awide range of new patterns of work or-ganisation at a variety of levels: telework-ing and increased use of information andcommunication technology (ICT) in theworkplace; downsizing, outsourcing, sub-contracting and globalisation, with the as-sociated change in employment patterns;demands for workers’ flexibility both interms of number and function or skills; anincreasing proportion of the populationworking in the service sector, and a grow-ing number of older workers; self-regulat-ed work and teamwork, etc. The researchcorpus is still developing in these areas(e.g., see Rosenstock, 1997, on NIOSH’songoing research project on downsizing),but there is some preliminary evidence thateven changes which may be thought toenhance the work environment can pro-duce the opposite effect. For example,Windel (1996) studied the introduction ofself-regulating team work in the office ofan electronics manufacturer. Althoughself-regulated work may be a source of in-creased self-efficacy and offer enhancedsocial support, Windel found that after 1year work demands had increased andwell-being decreased when compared tobaseline data. The data suggested that theincrease in social support brought aboutby self-regulating teams was not sufficientto counteract increased demands causedby the combination of a reduction in thenumber of staff and increases in manager-ial duties. Meta-analytical studies have alsoshown either mixed consequences (Bet-tenhausen, 1991; Windel & Zimolong,1997) or higher rates of absenteeism andstaff turnover (Cohen and Ledford, 1994)


81■

as a result of the implementation of teamwork or self-regulated work.

It is clear that changes which have such aprofound impact on the way organisationsoperate may carry associated potentialhazards that need to be monitored fortheir impact on health and well-being.


■82

tasks largely defining the interval betweenaversive events, and such intervals are notalways the shortest possible. They varywith task and outcome (see, for example,Brady (1958) and Rice (1963) for the ef-fects of avoidance schedule timing on ul-ceration in laboratory animals). Van Raaijet al. (1996) studied the effects of a low-intensity chronic intermittent unpre-dictable noise regimen on variousparameters of immune function. Male wis-tar rats were exposed to a randomisednoise protocol (white noise, 85 dB, 2-20kHz) for 10 hours per day, 15 minutes perhour over a total period of 3 weeks. Con-trol animals were exposed to ambientsound only. Immune function was moni-tored after 24 hours, 7 days, and 21 daysof noise exposure. Noise induced severalsignificant changes in immune function ina time-dependent differential pattern in-volving both immunosuppression and im-munoenhancement. Their results showthat various parameters of immune func-tion are affected differentially over time ina period of chronic mild noise stress, possi-bly due to sequential activation of differentphysiological mechanisms.

The availability of a warning signal appearsto attenuate the physiological response toan aversive event (for example, Weiss,1972; Miller et al., 1978) as do the avail-ability of avoidance or escape contingen-cies (for example, Anisman et al., 1980;Sklar & Anisman, 1981). Changing estab-lished or learnt procedures produces ex-tensive endocrinological changes (forexample, Brady, 1975). Short exposures toaversive stimuli may not have cumulativeeffects, and animals appear to adapt to

5.3A N I M A L S T U D I E S

Generally, the literature on animal behav-iour has not been incorporated into thisReport. However, such studies have alsosuggested the characteristics which definestressful situations for many differentspecies (Turkkan et al., 1982). Most relateto acute and well-defined stressors in theworkplace. These include: the interval be-tween aversive events, the availability ofwarning signals, the availability of avoid-ance or escape contingencies, changes inestablished procedures, and the durationof exposure to the aversive event and itsseverity. While the importance of thesecharacteristics has been establishedthrough studies on animal behaviour,mostly within a conditioning paradigm,they do have face validity in relation to theworkplace, and some map easily onto thecharacteristics listed in Table 1.

There appear to be critical temporal di-mensions involved with most aversive


83■

long exposures. Medium range durationsof exposure may therefore be most effec-tive in producing physiological responsesto aversive stimulation (for example,Forsyth & Harris, 1970). Generally, thegreater the intensity of the aversive event,the stronger the physiological and patho-logical responses (Turkkan et al., 1982), al-though this is not always the case (see, forexample, Ulrich & Azrin, 1962).


■84

tional problem areas were ‘lack of influ-ence over one’s work’ (35-40%), ‘involve-ment in short cycle repetitive work’ (about25%) and ‘long working hours’. 16% ofmen and 7% of women reported workingover 50 hours per week. The findings frommore recent studies are broadly similar (Eu-ropean Foundation, 1996, 1997).

Broad comparisons can be drawn be-tween, say, manual and managerial work.Warr (1992) has suggested that muchmanual work tends to be associated withextremes of workload (overload or under-load), low levels of decision-making andparticipation, and low task variety. Wherethe work is deemed to be semi-skilled orunskilled, there is also the problem of lowuse of skill or skill potential. Managerialwork, in stark contrast, is more often asso-ciated with work overload, role relatedproblems and uncertainty. French et al.(1982) have provided some support forthis suggestion. In their survey in the Unit-ed States, manual workers reported hav-ing low job complexity and lowrequirement for concentration (and an un-derutilisation of their skills), low participa-tion and low support. Professionalworkers, by comparison, reported havinghigh job complexity and no under utilisa-tion of their skills, and good participationand support.

The ongoing series of Whitehall studies(e.g., Marmot & Madge, 1987; Stansfeldet al., 1995; Bosma & Marmot, 1997;Stansfeld et al., 1999), offer data whichcompare the work characteristics of menof different grades in the Civil Service inthe United Kingdom. The work of the low-er grades has been characterised by under

5.4D I S T R I B U T I O N O F

P S Y C H O S O C I A L W O R K

H A Z A R D S

There is little good evidence relating to thedistribution of psychosocial hazards acrossdifferent types and levels of work andacross different countries. There have beenfew, if any, surveys which provide an ade-quate comparison of a wide range of dif-ferent types and levels of work (EuropeanFoundation, 1996).

A survey in the early 1990s attempted tomap the physical and organisational con-straints of work16 in the [then] twelvemember states of the European Communi-ty and in former East Germany (EuropeanSurvey on the Work Environment 1991-1992). Briefly, organisational problems af-fected a higher proportion of workers thandid physical problems. The main organisa-

16 The organisational constraints referred to in the Eu-ropean Survey on the Work Environment 1991-1992are equivalent to the psychosocial hazards referred toin this report.


85■

use of skills and by low social contact withothers at work. To a somewhat lesser ex-tent, it also involves low control and lack oftask variety. Interestingly, within this par-ticular organisational context, the work ofthe higher grades is also characterised bylow social contact and under use of skillsbut to a lesser extent. The most obviousdifferences between lower and highergrades relate to former’s lack of controland variety in work.

Our knowledge of how the distribution ofpsychosocial hazards relates to occupa-tional risk is somewhat complicated bysuggestions that it is particular synergisticcombinations of such hazards that carrythe greatest threat to health (Levi, 1984).Evidence of such synergy is claimed fromthe work of Karasek, but –as discussed insection 3.2.2– the evidence for such a syn-ergistic effect is weak. Another example isprovided by Martin & Wall (1989), whohave described a case study where the in-troduction of computer-based technologyinto the workplace resulted a high level ofstress reflecting the combination of in-creased cost responsibility with increasedattentional demands.


■86

5.5S U M M A R Y

It is possible from the available literature toexplore the effects of the more tangiblehazards of work on the experience ofstress and on health, and to identify thosepsychosocial hazards which pose a threatto employees. Most literature reviews haveidentified the need for further researchand development to translate this informa-tion into a form which can be used in theauditing and analysis of workplaces andorganisations. Such a model, togetherwith practical implementation strategies,has been provided by Cox et al. (2000) andis described in more detail in section 7.2.


87■

6.S T R E S S A N D H E A L T H

Over the past two decades, there has beenan increasing belief that the experience ofstress necessarily has undesirable conse-quences for health. It has become a com-mon assumption, if not a “cultural truism”(Leventhal & Tomarken, 1987), that it is as-sociated with the impairment of health.Despite this, the evidence is that the expe-rience of stress does not necessarily havepathological sequelae. Many of the per-son’s responses to that experience, both

psychological and physiological, are com-fortably within the body’s normal homeo-static l imits and, while taxing thepsychophysiological mechanisms involved,need not cause any lasting disturbance ordamage. However, it is also obvious thatthe negative emotional experiences whichare associated with the experience ofstress detract both from the general quali-ty of life and from the person’s sense ofwell-being. Thus the experience of stress,while necessarily reducing that sense ofwell-being, does not inevitably contributeto the development of physical or psycho-logical disorder. For some, however, theexperience may influence pathogenesis:stress may affect health. At the same time,however, a state of ill health can both actas a significant source of stress, and mayalso sensitise the person to other sourcesof stress by reducing their ability to cope.Within these limits, the common assump-tion of a relationship between the experi-ence of stress and poor health appearsjustified (Cox, 1988a).

This section presents a brief overview of thebroad range of health and health-relatedeffects which have been variously associat-ed with the experience of stress. More de-tailed discussions are available elsewhere(for example, Cox, 1978; Kristensen, 1996;Cincirpini et al.., 1984; Stainbrook & Green,1983; Millar, 1984, 1990). It focuses onchanges in health and health-related be-haviours and physiological function, whichtogether may account for any linkage be-tween that experience and psychologicaland physical health (Cox et al., 1983).

This Report also refers to the concept oforganisational healthiness. This concept

RE

SE

AR

CH


■88

(see Cox & Thomson, 2000) is based on ananalogy with individual health and is a de-rivation of sociotechnical systems thinking.It concerns the nature and viability of or-ganisations as systems, and includes mea-sures of the perceived quality of the socialorganisation and its relationships with thetechnical organisation. The term ‘thehealth of the organisation’ can be thoughtof as referring to its condition, in the samesense that the parallel term ‘the health ofthe individual’ refers to the general condi-tion of the person. In itself introducing thenotion of the ‘condition’ of the organisa-tion is intellectually insufficient, and fur-ther refinements need to be made: thehealth of the individual is often defined interms of their condition of body, mind andspirit (Longman’s Dictionary of the EnglishLanguage, 1992). In parallel terms, it hasbeen suggested by Smewing & Cox (1996)that the health of the organisation is “thegeneral condition of its structure and func-tion, management systems and culture.”This may be re-phrased as the quality of itsstructure and function, management sys-tems and culture. Additionally, a distinc-tion needs to be made between what ishealthy and what is not, in terms of ‘gen-eral condition’. Healthy individuals, andhealthy organisations, are those which areseemingly sound, that is fit-for-purpose,thriving and able to adapt in the longerterm. Expanding on this, a healthy organi-sation is “an organisation in which the dif-ferent components, which define itsgeneral condition, sum to it being ‘fit-for-purpose’, thriving and adaptable, andwhich is perceived positively by its employ-ees.” This is the definition adopted for thisReport.


89■

It is convenient to summarise the possiblehealth and health-related effects of stressunder two headings: psychological and so-cial effects, and physiological and physicaleffects. 6.1

E F F E C T S O F S T R E S S : A N

O V E R V I E W

The experience of stress can alter the waythe person feels, thinks and behaves, andcan also produce changes in their physio-logical function (Stansfeld et al., 1999;Sauter & Murphy, 1995; Cincirpini et al..,1984; Stainbrook & Green, 1983). Many ofthese changes simply represent, in them-selves, a modest dysfunction and possiblysome associated discomfort. Many areeasily reversible although still damaging tothe quality of life at the time. However, forsome workers and under some circum-stances, they might translate into poorperformance at work, into other psycho-logical and social problems and into poorphysical health (e.g., Devereux et al.,1999). Nevertheless, the overall strengthof the relationship between the experienceof stress and its antecedents on one handand health on the other is consistent butmoderate (Baker, 1985; Kasl, 1980a,1984).


■90

sonal relations, may be impaired by the ex-perience of stress, possibly reflecting morefundamental psychological changes in, forexample, irritability, attention span andmemory. Stress-related impairments of so-cial relations may both create secondaryproblems and reduce the availability of so-cial support.

Interestingly, the literature which describesthe translation from a normal psychologi-cal reaction to events to psychological ill-ness is not well formed, except in the caseof post traumatic stress and related disor-ders (see, for example, Figley, 1985; Hillas& Cox, 1987). A variety of psychologicalsequelae have been related to exposure toextremely threatening situations such ascatastrophes and disasters (Logue, 1980;Logue et al., 1981), war (Blank, 1981; Mil-gram, 1982) and terrorism (Bastiaans,1982).

Psychological ill health has also been asso-ciated with work stress (e.g., Stansfeld etal., 1999). One of the classical studies inthis area is that by Colligan et al. (1977).They conducted a survey, by occupation,of all first admissions to 22 of the 27 com-munity mental health centres in Tennessee(USA), from January 1972 through June1974. 8,450 cases were considered from130 different occupational groups. Occu-pations were ranked according to estimat-ed admission rate per 1000 workers andby z scores. Z scores were calculated forobserved against expected frequencies ofadmission on the basis of the relative fre-quency of members of the groups in thepopulation. These rates were then com-pared and the top 30 ranks reported. Thegroup with by far the highest rate was

6.2P S Y C H O L O G I C A L A N D S O C I A L

E F F E C T S

The psychological effects of stress may beexpressed in a variety of different ways,and involve changes in cognitive-perceptu-al function, emotion and behaviour. Someof these changes may represent attemptsto cope, including changes in health-relat-ed behaviours. There is evidence that somehealth-promoting behaviours, such as ex-ercise and relaxation, sleep and good di-etary habits, are impaired by theexperience of stress, while other health-risk behaviours, such as smoking anddrinking, are enhanced. Other behaviours,such as sexual behaviour, which may behealth-neutral, can also be impaired andthat impairment become a secondarycause of stress. Similarly, increases inhealth-risk behaviours can also becomesecondary causes of stress if sustained.Particular reference may be made to psy-chological dependency on alcohol orsmoking. Social behaviour, and interper-


91■

health technology technicians, and fiveothers in the top 30 were relatively lowstatus health care occupations. Many ofthe occupations which were representedin the top 30 also involved continual inter-action with others (patients, clients, cus-tomers, etc.), including human serviceoccupations. It has been argued that thepresence of so many health care occupa-tions in the top 30 is an artefact and sim-ply reflects their better knowledge ofpsychological health issues and of appro-priate health care facilities. However, thiscriticism cannot be so readily applied tothe personal service groups represented inthe survey. Operatives ranked 28th (out of130).


■92

els (Pollard, 1997). Research on peopleconducting their everyday lives, both inand out of work, is necessary to establishwhether the same responses are shown ona day to day basis. Such research requiresnew methodologies and careful data col-lection. So far, it has been shown thatadrenaline and blood pressure do seem tovary in expected ways. Other responses ineveryday life, including those of choles-terol, cortisol and the immune system, areless well characterised.

6 . 3 . 1 M e c h a n i s m s o f S t r e s s - r e l a t e dP h y s i o p a t h o l o g y

Zegans (1982) has suggested that thereare three different ways in which the phys-iological changes associated with the ex-perience of stress occur: as a concomitantphysiological response to an appraisal ofthreat or a failure of coping; as a physio-logical response to an appraisal of threatwhen active coping is not possible; and, asa non-specific response during the initialorientation-alarm state. Zegans (1982) hasalso suggested a number of ways in whichsuch physiological responses might con-tribute to pathology. The acute responsemay itself cause damage, particularly if analready compromised organ system is in-volved. If this is not the case, then repeat-ed occurrence of that insult might causemore permanent damage. The experienceof stress and the physiological insult itcauses might become chronic and againcause more permanent damage. Togetherthese three cover the often cited condi-tions for increased wear and tear on thebody (Selye, 1950): exposure to stressorswhich is severe, frequent or of long dura-

6.3P H Y S I O L O G I C A L A N D

P H Y S I C A L E F F E C T S

Contemporary research into physiologicaland physical health correlates of stress be-gan in the 1920s and 1930s with the workof Cannon (1929, 1931) and Selye (1936).Since then much has been published inthis area (e.g., Landsbergis et al., 1995;Meijman et al. , 1995; Kawakami &Haratani, 1999).

A large body of data has been accumulat-ed concerning physiological responses inpeople exposed to stressors in laborato-ries. Adrenaline and cortisol have becomeknown as stress hormones because, inmen, levels of both hormones consistentlyrise in response to stress in laboratory-based investigations. If chronically repeat-ed, elevation of adrenaline and cortisol islikely to have long-term consequences forhealth, especially cardiovascular health,partly via the effects of the hormones onblood pressure and serum cholesterol lev-


93■

tion. However, Zegans (1982) has also ar-gued that there are other mechanismswhich might contribute to the translationof a normal transient physiological re-sponse into one of pathological signifi-cance. Most appear to relate to theinteraction between stress responses andother physiological systems, particularlycontrol systems. First, the experience ofstress might result in an inappropriately se-vere response because either a deficiencyin relevant control systems or the stress re-sponse might stimulate other less benignreactions, again because of the lack ofcontrol elsewhere.

Zegans (1982) has also argued that the po-tentially pathogenic effects of the stress re-sponse express themselves by challengingthe various body systems which integrateand defend physiological function, andwhich underpin its link with behaviour.These systems include the hypothalamic-pituitary-adrenal cortical axis, the auto-nomic nervous system-adrenal medullaryaxis, the immune system, the reticular acti-vating system, and the cognitive-affectivecentres of the brain (Zegans, 1982). Muchattention, in the past, has been focused onthe role of the adrenal glands in stressphysiology and there are several reviewsavailable (for example, Selye, 1950; Levi,1972; Cox & Cox, 1985; Szabo et al.,1983). Stress can cause endocrine hypoac-tivity and hyperactivity (Lipton, 1976) andalter the balance of autonomic control al-tering function in the cardiovascular, respi-ratory, secretory and visceral systems(Lisander, 1979). It appears to impair ordistort the immune response (Stein et al.,1981; Kawakami & Haratani, 1999). It can

distort visceral perception (Brener, 1978),alter sleep patterns with knock-on effectson a variety of other activities (Weitzmanet al., 1975), and induce changes in otherbehaviours, some of which have signifi-cance for health (Antelman & Caggiula,1977).

There have been a small number of studiesthat have exposed subjects to stressful sit-uations and measured a wide range ofphysiological, largely biochemical, re-sponses and subsequently factor analysedthese data. Given that such studies requiremuch control and resources, it is often dif-ficult to capture sufficient data (by case) tosatisfy the requirements of factor analyticprocedures (see, for example, Ferguson &Cox, 1993). However, these studies are ofinterest, and those that have been report-ed have similar findings. Rose et al. (1967)analysed circulating hormone levels in 46men undergoing basic military training.They found five factors: a cortisol factor, acatecholamine factor, two factors relatedto androgens and oestrogen, and one re-lated to thyroid function. A study of 115military trainees by Ellertsen et al. (1978)identified three factors: a cortisol factor, acatecholamine factor and a testosterone-free fatty acid factor. Ryman & Ursin(1979) studied 31 American Navy compa-ny commanders in stressful conditions andagain reported a factor model of theirphysiological responses consistent withthat reported by Ellertsen et al. (1978).Ursin (1979) has suggested that thesethree physiological response factors mightbe differentially related to pathology. Sub-jects who respond with a predominantcortisol response might be more prone, ac-


■94

cording to the model of Henry & Stephens(1977), to depression, disorders of the im-mune system and gastric or duodenal ul-cers. Using the same argument, Ursin(1979) linked catecholamine responders tocardiovascular problems and possibly renalconditions.

Turkkan et al., (1982) have reviewed theavailable evidence from animal studies andhave come to a conclusion not inconsis-tent with that expressed by Zegans (1982).From the animal evidence, there appear tobe four physiological systems which areparticularly vulnerable to stress. The fourare: the cardiovascular system (Brady &Harris, 1977; Schneiderman, 1978; Kris-tensen, 1996 for a recent review); en-docrine system (for example, Mason,1968; Stone, 1975); gastro-intestinal func-tion (see Turkkan et al., 1982) and immunesystem (for example, Monjan, 1981;Kawakami & Haratani, 1999). Stress-relat-ed dysfunction in these systems is poten-tially significant for physical health.

Given this consensus, it is not surprisingthat the literature on stress and physicalhealth largely focuses on a number of par-ticular conditions, although a large num-ber of others are commonly cited as being,to some extent, stress-related (see, for ex-ample, Cox, 1978; Millar, 1984). It hasbeen suggested (Cox, 1978) that, undercertain circumstances, all physical condi-tions are potentially susceptible to stresseffects. If this is true, then questions mustbe asked about which are the more sus-ceptible or the most directly susceptible,and how that susceptibility is affected bythe nature of work and the workplace. Themore susceptible conditions appear to be

those relating to the cardiovascular andrespiratory systems (for example, coronaryheart disease and asthma: Marmot & The-orell, 1988; Kristensen, 1996, Bosma &Marmot, 1997; Stansfeld et al., 1995,1999), the immune system (for example,rheumatoid arthritis and possibly someforms of cancer), and the gastro-intestinalsystem (for example, gastric and peptic ul-cers), and those relating to the endocrine,autonomic and muscular systems. Amongthis group, most attention is currently be-ing focused on the immune system (e.g.,Peters et al., 1999; Borella et al., 1999;Kawakami & Haratani, 1999). There areseveral reviews available concerning thegeneral relationship between stress, emo-tion and immune function (for example,O’Leary, 1990) but few, if any, overviews ofthe effects of work-related stress on thatfunction.


95■

some aspects of perceived work stress. Theimmunological measures correlated withthe measures of health complaints relatedto immune system activity.

Endresen et al. (1991) have reported asomewhat similar study of Norwegianbank workers. Their data suggested that T-cell number (not examined in the Vaerneset al. (1991) study) and C3 (both cellular),and also IgM (humoral), were sensitive toboth perceived work stress and associatedemotional distress. There are a number ofother studies from the Norwegians whichsupport the finding of a linkage betweenthe experience of work stress and immunesystem activity. These include studies on:offshore divers (Bergan et al., 1987), sub-marine officers (Vaernes et al., 1987),nurses (Endresen et al., 1987; Arnestad &Aanestad, 1985), primary school teachers(Ursin et al., 1984) and shift workers in theprocessing industry (Vaernes et al., 1988).While it may be safe to conclude that sucha linkage exists, particularly in relation tocellular mechanisms, the direction of thisrelationship is not yet clear (the data arecorrelational) nor is its significance forhealth. Animal studies do, however, sug-gest that environmental stimuli (stressors)can alter the effectiveness of the immunesystem and reduce, in some circumstances,its ability to defend against both externalinfective agents and tumour growth (e.g.,Van Raaij et al., 1996). Much of this evi-dence has been usefully summarised in re-views of the role of psychosocial factorsand psychophysiological processes in can-cer(s) (Ader, 1981; Fox, 1981; Sklar & Anis-man, 1981; Cox & Mackay, 1982; Irwin &Anisman, 1984; Cox, 1984).

6.4W O R K - R E L A T E D

P S Y C H O I M M U N O L O G Y

There are a number of studies, many ofthem Norwegian, which demonstrate a linkbetween the experience of work stress andchanges in immune system activity, bothcellular and humoral. Vaernes et al. (1991)have reported a study of Norwegian airforce personnel in which they showed sig-nificant correlations between perceivedwork stress and immunoglobulin levels, andalso complaints related to immune systemactivity. Levels of complement componentC3 (humoral immunity) appeared particu-larly sensitive to variations in perceivedwork stress, and 31% of the variance in thismeasure could be accounted for by threework stress items relating to: taking the jobhome, having to lead other people, andproblems with subordinates. Interestingly,levels of IgM and IgG (cellular) did not cor-relate in any substantial way with the workstress measures. There was weak evidenceof a linkage between IgA (cellular) and


■96

6 . 4 . 1 M e c h a n i s m s

Work by Riley (1981) provides one possibleaccount of the way the experience of(work) stress may influence the develop-ment of cancers. Riley (1981) has arguedthat stress-associated pathologies will notbe observed, despite the presence ofstress, if there is no disease process alreadyin existence. He is arguing here for a rolefor stress in the development of existingcancers rather than in the aetiology of newcancers. Second, even if there is an existinglatent pathology, the effects of stress willnot be observed unless the disease is un-der the control of the immune system. Thismay account for stress effects on the de-velopment of some cancers and not oth-ers. Third, the effects of stress will only beobserved if there is some functional bal-ance between the individual’s defencesand the developing cancer. Where one orother is obviously dominant, any addition-al effects of stress may be impossible todetect. This means that the effects ofstress may not be detectable in the earlyand terminal stages of cancer develop-ment. This model was largely developedfrom Riley’s studies on rodents to accountfor cancer development (see Riley, 1979,1981; Riley et al., 1981) but might be use-fully applied to other diseases which in-volve the immune system activity (see, forexample, Cox, 1988b).

6 . 4 . 2 O t h e r P a t h o l o g i e s

A considerable variety of different patholo-gies, both psychological and physical, havebeen associated with the experience ofstress through work (Holt, 1982). Thosedisorders usually cited as being stress-relat-

ed include: bronchitis, coronary heart dis-ease, mental illness, thyroid disorders, skindiseases, certain types of rheumatoidarthritis, obesity, tuberculosis, headachesand migraine, peptic ulcers and ulcerativecolitis, and diabetes (Cox, 1978; Cooperand Marshall, 1976; Kroes, 1976, Selye,1976; Bosma & Marmot, 1997; Stansfeldet al., 1995, 1999; Kristensen, 1996).

According to Selye (1956) repeated, in-tense or prolonged elicitation of this phys-iological response, it has been suggested,increases the wear and tear on the body,and contributes to what he has called the‘diseases of adaptation’. This apparentlyparadoxical term arises from the contrastbetween the immediate and short-termadvantages bestowed by physiological re-sponse to stress (energy mobilisation foran active behavioural response) to thelong-term disadvantages (increased risk ofcertain ‘stress-related’ diseases).

Furthermore, the general occurrence ofphysical ill-health has also been related tothe experience of stress. For example,Nowack (1991) has reported on the rela-tionship between perceived stress andcoping style, on the one hand, and self-re-ported ill health, on the other. The fre-quency and severity of physical ill health(Wyler et al., 1968) were measured. Aftercontrolling for demographic variables andfor psychological well-being, perceivedstress was shown to be a strong predictorof both the frequency and severity of phys-ical ill health. About 30% of the variancein the latter was accounted for by per-ceived stress. However, there is the prob-lem of the direction of effect given that the


97■

study was correlational in nature, as manyin this area are.

Attention focused, in earlier years, on pep-tic ulcers as the prototypical work stressdisease (Holt, 1982). However, despite thisattention, opinion is divided on whether ornot the condition is stress related. In 1967,Susser concluded, from a review of the lit-erature, that there is a link, while some-what later Weiner (1977) stated that nosuch link had yet been proved. However, atthe same time, House et al. (1979) report-ed a link between work stress –particularlystressful relations with others– and ulcers,after controlling for seven possibly con-founding variables.

Much attention has also been focused oncardiovascular diseases, especially coro-nary heart disease. The origin of coronaryheart disease, like many chronic degenera-tive conditions, is multifactorial but workfactors and stress have clearly been indi-cated (see, for example, Poppius et al.,1999; Kristensen, 1996; Cooper & Mar-shall, 1976; House, 1974; Jenkins et al.,1976). However, the evidence is not com-pletely unequivocal and negative findingshave been reported (see, for example,Haynes et al., 1978a, 1978b). One well-es-tablished and frequently replicated findingis the link between type A behaviour pat-tern and cardiovascular disease (see, forexample, Jenkins et al., 1968).

There has been evidence for a long timethat the experience of stress can con-tribute to an acceleration of the diseaseprocess in at least one particular type ofrheumatoid arthritis (see Genest, 1983,1989). Rimon & Laakso (1985) have sug-

gested that there are two separate types ofrheumatoid arthritis: one, a disease formless connected with genetic factors andpotentially more influenced by stress, anda second form more associated withheredity disposition and less influenced bypsychosocial processes. These groups mayoverlap with those described by Crown etal. (1975). These authors distinguished be-tween patients on the basis of the pres-ence or absence of rheumatoid factor (RF).The sero-positive group, with RF, showed amore negative psychopathological profilethan those without RF. Such findings havebeen replicated by other workers such asGardiner (1980) and Volhardt et al. (1982).


■98

not be personally or professionally accept-able: people may continue to turn up forwork under stress but perform poorly: pre-senteeism17.6.5

O R G A N I S A T I O N A L E F F E C T S

If significant numbers of workers are expe-riencing and expressing the effects ofstress at work, then the problem assumesorganisational proportions. There hasbeen some (unsupported) suggestion thatif (about) 40% of workers in any group(department or organisation) are facingstress related problems, then that group ororganisation can also be said to be un-healthy in some way. From the literature,there appear to be several effects of stresswhich may be of more direct concern toorganisations. The most frequently citedappear to be: reduced availability for workinvolving high turnover, absenteeism andpoor time keeping (all essentially ‘escape’strategies), impaired work performanceand productivity, an increase in client com-plaints (cf: Jones et al., 1988) and an in-crease in employee compensation claims(Barth, 1990; Lippe, 1990; Neary et al.,1992). For some, escapist strategies may

17 Presenteeism is a term used to refer to “being phys-ically present at work but mentally absent” (e.g.,Cooper et al., 1996). It is contrasted with absenteeism.


99■

6.6S U M M A R Y

There is evidence that the experience ofstress at work is associated with changes inbehaviour and physiological function, bothof which may be detrimental to employ-ees’ health. Much is known about the pos-sible mechanisms underpinning sucheffects, and particular attention has beenpaid to pathologies possibly associatedwith impaired immune activity as well asthose more traditionally linked to stress,such as ulcers, coronary heart disease andrheumatoid arthritis.


■100


101■

7.T H E A S S E S S M E N T A N D

M A N A G E M E N T O F

W O R K - R E L A T E D S T R E S S

The European Agency’s Topic Centre onGood Practice – Stress at Work (TC/GP-ST)collects and evaluates existing good prac-tice information about stress at workacross the EU and beyond. Consequently,the present Report will not examine actual

practice, but –having reviewed the re-search into the nature, causes and effectsof work-related stress in the preceding sec-tions– will deal briefly with the research ev-idence regarding the assessment andmanagement of stress at work.

RE

SE

AR

CH


■102

that most “stress management” interven-tions target the individual rather than theorganisation (the former is usually seen ascheaper and less cumbersome: see section7.5), are often off-the-shelf designs, andare entirely divorced from the process ofdiagnosis of the problems -if diagnosistakes place at all (Cox, 1993). A differenttype of approach is therefore required inorder to carry out risk assessments whichcan then inform the design of interven-tions -in other words, a strategy that actu-ally asks the question before giving theanswer. Such a strategy has already beensuggested for the control of physical haz-ards (e.g., Council Directive 89/391/EEC[“Framework Directive”]; European Com-mission, 1996): the control cycle, whichhas been defined as “the systematicprocess by which hazards are identified,risks analysed and managed, and workersprotected” (Cox & Griffiths, 1995) andcomprises 6 steps:

1. Identification of hazards

2. Assessment of associated risks

3. Implementation of appropriate controlstrategies

4. Monitoring of effectiveness of controlstrategies

5. Re-assessment of risk

6. Review of information needs, andtraining needs of employees exposedto hazards

Steps 1 through 5 are recursive and de-signed to ensure continuous improvementof occupational health and safety at work.Each step can be conceptualised as a fur-ther cycle of activities similar to a goal-seeking process as described by Schott(1992). As a systematic and comprehen-

7.1T H E A S S E S S M E N T O F

W O R K - R E L A T E D S T R E S S :

T H E C O N T R O L C Y C L E

As discussed earlier, there are numerousreviews of research into psychosocial haz-ards and stress (e.g., Cox, 1993;Cartwright & Cooper, 1996; Borg, 1990;Hiebert & Farber, 1984; Kasl, 1990; Coop-er & Marshall, 1976), and a large numberof papers dealing with stressors in almostevery conceivable work setting and occu-pation. However, as Cox (1993) indicates,“research into the nature and effects of ahazard is not the same as assessment ofthe associated risk”. Indeed, most pub-lished studies would provide very little datathat could be used for a risk assessment.Many “stress surveys” tend to identify onlyhazards or only outcomes, whereas theobject of a risk assessment is to establishan association between hazards andhealth outcomes, and to evaluate the riskto health from exposure to a hazard.

An almost unavoidable corollary of thepaucity of adequate risk assessments is


103■

sive approach to assessing the risks withinthe work environment, the control cyclesatisfies current legal requirements. How-ever, it is sti l l necessary to evaluatewhether it represents a scientifically validand reliable strategy to assess psychosocialhazards.

Occupational health psychology borrowedthe concept of risk assessment from thefield of physical hazard control (Cox & Cox,1993). The formalised approach requiredby EU legislation on physical hazards (e.g.,Council Directive 98/24/EC) is ideally im-plemented through a problem-solving ap-proach such as the control cycle. Forexample, the EC guidance documentclearly subscribes to the notion of the con-trol cycle as the favoured approach in itsdefinition of risk assessment: “a systemat-ic examination of all aspects of the workundertaken to consider what could causeinjury or harm, whether the hazards couldbe eliminated, and if not what preventiveor protective measures are, or should be,in place to control the risks” (EuropeanCommission, 1996 § 3.1). The risk assess-ment approach also has in its favour theadvantage of being an already familiarstrategy with employers. It also provides anintegrated framework which could osten-sibly accommodate psychosocial hazardsas an additional category of hazards to befound in the workplace.

However, considerable difficulties emergewhen trying to broaden the risk assess-ment approach to include psychosocialhazards: The first task is to achieve a defi-nition of the terms used in risk assessment.This is far from straight-forward and hasoften proved difficult even in the more

tangible area of physical hazards18. A re-view of the literature suggests that there isreasonable consensus on the definitions ofthe basic terminology. For example, the EUMember States have agreed on “acceptedand practical” definitions for the followingfundamental terms:

Hazard: The intrinsic property or ability ofsomething (e.g. work materials, equip-ment, work methods and practices) withthe potential to cause harm.

Risk: The likelihood that the potential forharm will be attained under the conditionsof use and/or exposure, and the possibleextent of the harm.

(European Commission, 1996, § 1.2)

Although these are acceptable at a basiclevel and as a guideline for employers,from a scientific perspective there is a dan-ger of stretching the parallel too far whenthe need arises to operationalise those de-finitions. For instance, there remain somedoubts as to whether the above definitionof “hazard” would include some charac-teristics of the work environment such as“broad corporate policies: paid leaves ofabsence, promotion, health insurance cov-erage, etc.” (Landy et al., 1994). More-over, it is not possible to establish an exactconceptual or practical symmetry betweenphysical and psychosocial hazards. Expo-sure to certain levels of radiation is known

18 See, for example, the Internet-based project spon-sored by the Organization for Economic Cooperationand Development to harmonise the definitions of thebasic generic terms involved in the risk assessment ofchemical hazards (Organization for Economic Cooper-ation and Development, 1997)


■104

to be an indisputable risk to every worker’shealth, while one can be very confidentthat other substances are safe for every-one. However, it is not obvious that suchstatements can be put forward with anyconfidence for most –if not all– psychoso-cial hazards. Could anything within thework environment be a potential psy-chosocial hazard? If so, the definition ofhazard could become meaningless. If not,what aspects of work could never be haz-ardous, and why? Similarly, whereas psy-chosocial hazards can be conceptualisedas part of a continuum that is representedby “psychosocial hazard” at one end and“salutogenic factor” at the other (e.g.from very low to very high job control),physical hazards such as asbestos wouldseem to be negative per se and lacking apotential salutogenic role (even its absencewould not be health-enhancing but mere-ly neutral).

A study by Kang et al. (1999) in the physi-cal hazards field illustrates these conceptu-al and practical differences betweenphysical and psychosocial hazards. Theyexamined the usefulness of an automatichazard analyser (AHA). This system per-forms hazard analysis in terms of bothfunctional failure and variable deviation inthe search for possible causes of accidents.The result of analysis provides a pathwayleading to an accident, and, therefore,gives not only clear understanding of theaccident, but useful information for haz-ard assessment. Kang et al. applied AHA tothe feed section of an olefin dimerizationplant, and the system performed betterthan traditional qualitative hazard analysismethods. Research into the assessment of

psychosocial hazards is clearly in too earlya stage to permit the use of an expert sys-tem such as that described by Kang et al.

With regard to “harm”, in order to cate-gorise “the extent of harm” referred to inthe definition of risk, the EC guidance doc-ument suggests the following range ofoutcomes:

Minor damageNon-injury accidentMinor injury (bruise, laceration)Serious injury (fracture, amputation,chronic ill-health)FatalMultiple-fatality

(European Commission, 1996§ 4.8.3)

It would not be a simple task to achieve aconsensus on a hierarchy of “degrees ofpsychological harm” similar to that whichis easily available for physical harm. More-over, a number of studies (Landy et al.,1994; Kasl, 1987, 1990; Johnson, 1996)have identified the difficulties encounteredwhen researchers and practitioners have todecide on what indicators of both physicaland psychological well-being they shoulduse: “In a scenario which repeated itselfover and over, a particular approach wasseen as pretty reasonable for surveillanceof injuries, somewhat useful for a narrowband of work-related diseases, but inade-quate for the intended broader spectrumof such diseases, and by implication ines-timably useless for surveillance of psycho-logical disorders” (Landy et al., 1994).

It is clearly not merely a matter of agreeingon what the appropriate indices are (indi-vidual health or organisational function-


105■

ing? Both? Why? Should the selected in-dices take into account the culture of theorganisation and / or occupational group,or should the culture itself be an index oforganisational healthiness?). It is arguablymore difficult to find reliable and validsources of information for the indices: Psy-chiatric diagnoses, treatment and care-seeking records, symptom checklists,indices of functional effectiveness, “posi-tive mental health” measures, indicatorsof “quality of life”, health-related behav-iours, employers’ and trade unions’records (where they exist at all), data onuse of occupational health services, anddata on compensation and litigation are alleither seriously or fatally flawed due toself-selection, recording and reportingproblems, complex operationalisation, orconfounding variables. To be fair, these dif-ficulties also arise for the assessment ofphysical hazards (e.g. inaccurate organisa-tional records, unwillingness of companiesor individuals to report accidents or “near-misses” which may reveal possible defi-ciencies in their control systems, etc.),However, the problems for psychosocialhazards are compounded by the difficul-ties intrinsic to monitoring outcomeswhich are less perceptually obvious thanphysical injuries or fatalities.

This elusive nature of psychosocial hazardsalso contributes to making causal relation-ships between hazard and harm consider-ably more difficult to establish (Johnson &Hall, 1996). One only needs to considerthe differential effort required to prove be-yond doubt the effects of asbestos or radi-ation on individual health and those ofmost of the psychosocial hazards men-

tioned earlier (e.g. the vast literature accu-mulated on the effects of job control oncardiovascular disease; see section 6.3).

Finally, much of the difficulty in drawing ascientifically valid and exact parallel withthe risk assessment of physical hazards liesin the problems encountered by re-searchers when trying to measure thework environment. These were examinedin detail in section 4.

To summarise, the risk assessment model isvery helpful as an analogy and represents auseful strategy for the assessment of psy-chosocial hazards at work. However, thereare a number of issues to bear in mind: (a)the operationalisation of definitions ofhazard, (b) the identification of adequateindices of harm that can also be reliablymonitored, (c) satisfactory proof of acausal relationship, and (d) problems ofmeasurement of the work environment.


■106

ment and reduction of psychosocial haz-ards. At the heart of the risk managementdescribed by Cox et al. (2000) are two dis-tinct but intimately related cycles of activi-ty: risk assessment and risk reduction.These form the basic building blocks forthe staged model of risk management.However, in addition to risk assessmentand risk management, three other compo-nents are specified. These include “evalua-tion” and “organisational learning andtraining”. The model also introduces anew linking stage between risk assessmentand risk reduction, that of “the translationprocess”. Because all aspects of the riskmanagement process should be evaluated,and not just the outcomes of the risk re-duction stage, the “evaluation” stage istreated as all encompassing and supra-or-dinate to the other stages. This model ofrisk management is shown below (Figure4). The risk reduction stage, in practice,tends to involve not only prevention butalso actions more orientated towards indi-vidual health and welfare.

7.2A R I S K M A N A G E M E N T

A P P R O A C H T O W O R K - R E L A T E D

S T R E S S

Cox et al. (2000) have described a frame-work that takes into consideration theproblems outlined in the previous sectionand aims to overcome the difficulties ofadapting the control cycle to the assess-

F i g u r e 4 : A f r a m e w o r k m o d e l o f r i s k m a n a g e m e n t f o r w o r k s t r e s s

EVALUATION

RISK ASSESSMENT(including AUDIT) TRANSLATION RISK REDUCTION

ORGANISATIONALLEARNING & TRAINING

FEEDBACK


107■

There are parallels between this model andthe organisational intervention process be-ing developed by applied researchers inthe USA. The “interventions team” work-ing as part of NORA (National Institute,

1999) also emphasise the need for evalua-tion and the feedback of evaluation datato inform earlier stages in the overall analy-sis-intervention cycle (Goldenhar et al.,1998) (see Figure 5 below).

F i g u r e 5 : I n t e r v e n t i o n r e s e a r c h i n o c c u p a t i o n a l s a f e t y a n d h e a l t h : A c o n c e p t u a lm o d e l ( f r o m G o l d e n h a r e t a l . , 1 9 9 8 )

Cox et al. (2000) have also described afive-step strategy to carry out a risk assess-

ment process in practice. The differentphases are summarised in Figure 6 below.

The five steps for the risk assessment for work stress:

• Step 1: Familiarisation • Step 2: Work Analysis Interviews• Step 3: Assessment Survey• Step 4: Audit of Existing Management Control and Employee Support Systems• Step 5: Analysis and Interpretation of Assessment Data

F i g u r e 6 : T h e f i v e s t e p s f o r r i s k a s s e s s m e n t f o r w o r k s t r e s s

EffectivenessResearch

ImplementationResearch

DevelopmentalResearch

1 Gather BackgroundInformation (ConductNeeds Assessment)

2 DevelopPartnerships

3 ChooseMethods or

Designs4 Complete

Development,Implementation,

or Evaluation

5 Report &Disseminate


■108

Each step builds on information collectedduring any preceding step. The initial steps(Steps 1, and 2) are designed to build amodel of the work and working conditionsof the assessment group that is goodenough to support the design and lateruse of the assessment instrument (Step 3).This instrument is used to quantify theworkers’ exposure (at group level) to allthe significant stressors associated withtheir work and working conditions, andassess their health.

The five steps are largely sequential withone possible exception. The Audit of Exist-ing Management Control and Employee

Support Systems (step 4) can be conduct-ed either in parallel with the Work AnalysisInterviews, or following the Analysis andInterpretation of Assessment Data. It is of-ten most convenient to conduct it in paral-lel with the Work Analysis Interviews. Inthis case, the information collected canusefully contribute to the working modelof the assessment group’s situation that isbuilt up in the early stages of the assess-ment. Finally, all information is analysedand interpreted (step 5).

These five steps can be mapped onto anoverall assessment strategy as shown inFigure 7 below.

[1] Familiarisation

[2] Work Analysis Interviews

[3] Assessment Survey

Identify & AssessGroup Exposure toStressful Hazards

Identify & Assess KeyMarkers of Employee

& OrganisationalHealth

Audit ExistingManagement Control& Employee Support

Systems

Identify LikelyRisk Factors

[4] Audit

[5] Analysis & Interpretationof Data

MakeRecommendationson Residual Risks

F i g u r e 7 : R i s k a s s e s s m e n t s t r a t e g y a n d p r o c e d u r e s


109■

cursory consideration to risk managementin Section 5 under the heading “Actions asa result of risk assessment at work”. Al-though useful as a tool for organisationsundertaking a risk assessment, the docu-ment –as would be expected given its pur-pose– only offers a generic flowchart ofoptions to choose from depending on theresults of the assessment. Furthermore,the lack of examination of the effective-ness of stress management programmesremains one of the main shortcomings inthe scientific literature (van der Hek &Plomp, 1997).

7.3T H E M A N A G E M E N T O F

W O R K - R E L A T E D S T R E S S

The scientific literature on risk manage-ment is even more sparse than that on riskassessment. Exhaustive literature reviewshave failed to produce more than a hand-ful of studies (e.g. Jackson, 1983; Israel etal., 1996). Apart from reviews of stressmanagement interventions (e.g., van derHek & Plomp, 1997; Dollard & Winefield,1996; Burke, 1993; International LabourOrganization, 1992; Murphy, 1984 &1988; Cox, 1993), much of what is pub-lished is limited to prescriptions and rec-ommendations (e.g., Briner, 1997;Kompier et al., 1998), guidance for whatamounts to “good management practice”with some psychological content (e.g., In-ternational Federation, 1992, and variouspublications by NIOSH in the USA) orgeneric standard recipes for a healthierwork environment (e.g. Landy, 1992;Locke, 1976). The EC’s 1996 Guidance onRisk Assessment at Work document gives


■110

worker training to reduce the likeli-hood of those workers experiencingstress.

2. Timely reaction, often based on man-agement and group problem-solving,to improve the organisation’s (or man-agers’) ability to recognise and dealwith problems as they arise.

3. Rehabilitation, often involving offer-ing enhanced support ( includingcounselling) to help workers copewith and recover from problemswhich exist.

Within this model, many authors make adistinction between those objectiveswhich concern or focus on the organisa-tion (organisational stress management)and those that concern and focus on theindividual (personal stress management)(for example, De Frank & Cooper, 1987;Ivancevich & Matteson, 1986; Ivancevichet al., 1990; Keita & Sauter, 1992; Matte-son & Ivancevich, 1987; Murphy, 1984,1988; Murphy & Hurrell, 1987; Newman &Beehr, 1979; Quick & Quick, 1984; Quicket al., 1992a; Schwartz, 1980).

While equal attention is now being paidto both in the literature (and in legisla-tion), much practice is biased towards thepersonal (and more clinical) approach. Atthe same time, while attention is beingpaid to preventive and rehabilitativestrategies, less attention is being focusedon reactive strategies. One exception isthat of Cox & Cox (1992) who describe a‘stress tool kit’ for line and specialist man-agers to help them recognise and dealwith employees’ problems which arestress-related.

7.4P R I N C I P L E S O F S T R E S S

M A N A G E M E N T

In one of the early papers in this area,Newman and Beehr (1979) suggested thatstress management can be classified interms of its objectives and strategies, its fo-cus or target, and the agent throughwhich it is carried out. This section exam-ines each in turn.

7 . 4 . 1 O b j e c t i v e s

While only a minority of organisations ap-pear to be directly and deliberately ad-dressing the management of occupationalstress, those actions which are being takencan be classified in terms of their impliedobjectives. There are, at least, three dis-tinct sets of objectives which have beenadopted by organisations in managingwork stress and its health effects (Cox etal., 1990; Dollard & Winefield, 1998):

1. Prevention, often control of hazardsand exposure to hazards by design and


111■

7 . 4 . 2 A g e n c y a n d T a r g e t

Given that a clear distinction is made be-tween the different possible objectives,Cox et al. (1990) have suggested that theissues of agency and focus or target, asraised by Newman and Beehr (1979), canbe paired and effectively dealt with interms of three questions:• Organisation as agent and target: what

can the organisation do to put its ownhouse in order?

• Organisation as agent and workers astarget: what can the organisation do toenhance the support it offers workers?

• Employee as agent and target: what canindividual workers do better to managetheir work and any associated experi-ence of stress?

It was pointed out by Cox et al. (1990) thatthe second and third questions overlap. Inreality, they question whether the organi-sation can help the individual to helpthemselves. This is often the explicit goalof employee support programmes. Logi-cally, there is a fourth pairing (worker asagent and organisation as target) whichdescribes the involvement of workers inorganisational development.


■112

on the subject, the relative effectiveness ofsuch programmes has been difficult to de-termine, largely because of methodologi-cal deficiencies inherent in much of therelevant research and lack of adequateevaluations (e.g. Briner, 1997; van der Hek& Plomp (1997); see also section 7.1).Murphy et al. (1992), Kompier et al. (1998)and van der Hek & Plomp (1997), for ex-ample, consider that evaluations shouldinclude cost-benefit analyses and assess-ments of employee satisfaction, job stres-sors, performance, absenteeism andhealth status. However, they rarely do so.Van der Hek & Plomp (1997) found that,out of 342 scientific papers on stress man-agement interventions, only 37 referred tosome kind of evaluation research, of which7 were ‘evaluated’ on the basis of anecdo-tal comments from participants.

Beehr & O’Hara (1987), Burke (1993), Dol-lard & Winefield (1996) have reviewed thedifficulties involved in the design andevaluation of stress management inter-ventions. Most designs are either ‘pre-ex-perimental’ or ‘quasi-experimental’ (trueexperiments being difficult to conduct inorganisations) and vary considerably intheir ability to control for the various‘threats’ to validity. For example, in thestudy of the effects of counselling on em-ployees’ anxiety levels, anxiety scores mayappear to return to normal over repeatedtesting but, if employees were initially se-lected (or volunteered) on the basis of ex-treme scores, this may simply reflect aregression to the mean. Since most sec-ondary and tertiary stress managementprogrammes are voluntary, selection ef-fects may operate: the characteristics of

7.5C O M M O N I N T E R V E N T I O N S :

T H E I R E F F E C T I V E N E S S

To summarise, there are three commontypes of intervention to be found in the lit-erature on stress management (see, for ex-ample, Murphy, 1988; Cooper &Cartwright, 1997; Dollard & Winefield,1996; Kompier et al., 1998):

1. Primary: some form of organisationalor work development which attemptsto reduce stressors (control hazards),including work design and ergonomics(e.g., Jones et al., 1988; Golembiewskiet al., 1987)

2. Secondary: worker training either inthe form of health promotion or psy-chological skills (e.g., Lindquist &Cooper, 1999)

3. Tertiary: employee assistance (largelyfocused on the provision of coun-selling).

These are described more fully below.However, despite a burgeoning literature


113■

participants and non-participants may bequite different. Selection effects have beendiscussed in detail in the evaluation of em-ployee fitness programmes (see, for exam-ple, Jex, 1991).

One of the advantages of stressor reduc-tion interventions is that they attempt tochange stressors common to all, therebyside-stepping selection effects (Burke,1993). Further, many studies claim to showimprovements as a result of interventionsthat may in fact be due to non-specific ef-fects such as treatment credibility, expecta-tions or even just sitting quietly. On therare occasions that control groups havebeen included in occupational stress inter-ventions, it is not uncommon for bothtreatment and control groups to show im-provements. Similarly, where differenttypes of stress management programmeshave been compared it is not unusual forall to produce similar improvements (forexample, Hart, 1987). These reservationsand others have been echoed in publica-tions by Keita & Sauter (1992) and Quick etal. (1992b). With the paucity of sounddata on the outcomes of such interven-tions, it is not surprising that it has beenvery difficult to make judgements concern-ing the cost benefits (the merits of an in-tervention in financial terms) or costeffectiveness (merits in comparison withavailable alternatives). This issue is also re-ferred to later in this section when dealingwith employee assistance programmes.

Many reviews (Murphy, 1988; Ivancevichet al., 1990; Burke, 1993; Dollard & Wine-field, 1996; Cooper & Williams, 1997) findmost stress management interventions areindividually focused, designed for man-

agerial and white-collar workers and con-cerned with changing the worker as op-posed to work or the work environment.For example, Williamson (1994) foundthat out of 24 evaluative studies of stressinterventions being conducted at the time,21 focused on the individual, (e.g., stressmanagement programmes, relaxation,etc.) and only 3 focused on change at theorganisational level. Kompier et al. (1998)offer four main reasons why interventionsthat target the individual appear to bemore numerous in the scientific literature:“the opinions and interests of companymanagement, the nature of psychology,the difficulty of conducting methodologi-cally ‘sound’ intervention studies and thedenominational segregation of stress re-search”. Briner (1997) has also noted that“primary” interventions are the least pop-ular, and has suggested that “in an orga-nizational context […] changing thenature of the job or the organization maybe considered more daunting and complexthan simply buying-in some of the othertypes of interventions”.

This may be a reflection of the nature andinfluence of management views in somecountries. Surveys in the United Statesamong management and union groupshave revealed clear differences in theirviews of stress (for example, Singer et al.,1986). Whilst management emphasise in-dividual (secondary and tertiary) interven-tions, seeing personality, family problemsor lifestyle as being prominent sources ofstress, union groups consider social and or-ganisational factors such as job design andmanagement style as being both more re-sponsible and more suitable targets for in-


■114

tervention. Dollard & Winefield (1996)suggested that “the politics involved inconceptualizing the stress problem and inrecognizing psychological disorder as aleading occupational health issue in Aus-tralia has impaired advances towards itsprevention and treatment and the statusof occupational stress as a national policyissue”. It has been suggested that in Scan-dinavia, where responsibility for workingconditions is shared equally betweenlabour and management groups, organi-sational approaches to stress managementare generally more common than else-where (Landsbergis, 1988). The domi-nance of management views, particularlyin the United States, has contributed tothe development of Employee AssistanceProgrammes and Stress ManagementTraining ahead of stressor reduction / haz-ard control techniques.

S t r e s s o r R e d u c t i o n ( H a z a r dC o n t r o l ) I n t e r v e n t i o n s

Murphy (1988) identified and reviewedseveral interventions which addressed thenature and design of the work environ-ment or organisation (Jackson, 1983; Wall& Clegg, 1981; Pierce & Newstrom, 1983).His interpretation of these studies wasframed by the concept of control in rela-tion to stress and health (see Averill, 1973;Miller, 1979; Thompson, 1981; Cox & Fer-guson, 1991). The issue of control is a per-vasive one throughout the stress literature.

The study by Wall & Clegg (1981) manipu-lated worker control over significant as-pects of the work process; themanipulation in the Jackson (1983) studyproduced modest increases in worker con-

trol; the manipulation in the study byPierce & Newstrom (1983) –introduction offlexitime systems– could also be said to in-crease worker control over some aspectsof their work. All three studies demon-strated the effectiveness of the control re-lated manipulations in reducing workers’report of stressors and aspects of their ex-perience of stress.

Jackson (1983) reported a well-designedevaluation of an intervention study con-ducted amongst staff working in 25 out-patient clinics in hospitals in the UnitedKingdom and designed to reduce role am-biguity and conflict. The hypothesis undertest was that increased participation in de-cision-making would decrease the experi-ence of role problems. Clinic supervisorswere given appropriate training on partici-pation and the number of staff meetingsheld in the clinics was increased. The ef-fects of these interventions were evaluatedagainst a number of outcome measuresusing a Solomon 4 group design. Signifi-cant reductions in role ambiguity and roleconflict were observed in the interventionclinics after 6 months follow up.

A study by Jones et al. (1988), which fo-cused on the number of malpractice com-plaints received by a medical practice,produced positive results. Four studieswere conducted to examine both the rela-tion between stress and medical malprac-tice and the impact of stress managementprograms in reducing malpractice risk. 76hospitals and more than 12,000 individu-als participated. In study 1, hospital de-partments with a current record ofmalpractice reported higher levels of on-the-job stress than did matched low risk


115■

departments. In study 2, workplace stresslevels of 61 hospitals correlated signifi-cantly with frequency of malpracticeclaims. In study 3, a longitudinal investiga-tion was conducted to evaluate the impactof an organisation-wide stress manage-ment programme on the frequency of re-ported medication errors. Resultssuggested a significant drop in averagemonthly medication errors as a result ofthe program. Study 4 was a 2-year longi-tudinal investigation that compared thefrequency of medical malpractice claims.Twenty-two hospitals that implementedan organisation-wide stress managementprogramme had significantly fewer claimscompared with a matched sample that didnot participate.

MacLennan (1992) presents several prob-lem situations facing institutions in the USand details some of the organisationalremedies instituted to tackle them. Al-though these interventions were not evalu-ated, they provide useful examples of thetype of approach considered promising. TheFirst American Bankcorp of Nashville, Ten-nessee (which has 150 banks) experiencedproblems with high turnover, sickness ab-sence and low productivity. They formed‘action teams’ from each area of operationwho were trained in problem identificationand problem-solving. Employees rotated onand off the teams with the result that manypeople had the opportunity to participate.In the first year, turnover was cut from 50%to 25%. MacLennan details several inter-ventions undertaken by other US bankinginstitutions designed to reduce work-familyconflicts including onsite day centres forpre-school and school children, maternity

leave arrangements, job protectionschemes, arrangements for part-time workfor returning mothers and fathers, flexitimeand working at home, the provision of‘family sick days’ and unpaid leave to beused for children, spouses or elderly par-ents. Other organisational interventions(MacLennan, 1992) concerned sexual ha-rassment and work flow problems in gov-ernment and stressors facing long distancelorry drivers, air traffic controllers and AM-TRAK (railway) engineers. In the latter case,for example, management had reduced thenumber of engineers driving fast trains fromtwo to one, with no consideration given tothe fact that most of the drivers had beenused to working in pairs for some time, northat the seating and instrumentation incabs may have needed alteration. Followingcomplaints of increased levels of stress, theunion hired consultants to identify the rele-vant stressors facing solitary drivers of high-speed trains, many of which could betackled by organisational interventions.

Murphy & Hurrell (1987) describe the de-velopment of a worker-management‘stress reduction committee’, as a possiblefirst step in any stress management inter-vention. In their study, the results of astress management workshop providedthe information required for an employeesurvey. The committee then reviewed andprioritised the identified sources of stress,planned organisational interventions de-signed to address them and presentedthem to management, recommending anannual audit. Such approaches acknowl-edge the importance of the process as wellas the content of interventions by the in-volvement of employees.


■116

Golembiewski et al. (1987) describe an in-tervention in which a programme of or-ganisational development wasimplemented. The programme affected allthe members of the organisation and tookplace over 13 months. The authors mea-sured levels of burnout, job involvementand turnover rates. They found that thelevel of burnout decreased (and remainedlow for at least four months), and de-creased somewhat in the following ninemonths. There were also improvements inthe rates of turnover, which persisted afterthe initial implementation of the pro-gramme.

Finally, Landsbergis and Vivona-Vaughan(1997) carried out, and evaluated, an in-tervention based on organisational devel-opment, action research and Karasek’s(1979) job strain model. In this study, em-ployee committees conducted problem di-agnosis, action planning and action takingin two departments in a public agencyover a period of one year (there were alsotwo waiting-list control departments). Pre-and post-intervention measures were ob-tained from workers in all four depart-ments via a standardised surveyinstrument, and qualitative informationwas also obtained at a four-month follow-up by telephone interviews with membersof the problem-solving committees.

The results obtained by the evaluation sur-vey were mixed: for members in Interven-tion Department 1, values at post-testwere nearly all in a more negative directionthan at pre-test. However, for InterventionDepartment 2 all study variables were in amore positive direction. The feedback fromthe evaluation questionnaire was also

somewhat critical: 52% (Department 1)and 39% (Department 2) of staff memberswho did not participate in the committeesfelt that the intervention had been either“ineffective” or only “slightly effective”.Nevertheless, over two-thirds of them feltthat the programme should be initiated inother departments. The follow-up inter-views revealed a possible explanation forthis pattern of results: a divisional reorgan-isation begun by the agency one week be-fore post-test had affected all 4departments and resulted in feelings offrustration and disappointment. Workersfeared that the reorganisation would resultin the loss of the gains achieved by theproblem-solving committees. As a result,in Department 1 many of the proposedchanges (e.g., a policy and proceduresmanual and an associated committee) hadnot been completed. The authors discussthe reasons for the mixed results both interms of the methodological limitationsand the inevitable constraints of appliedresearch.

S t r e s s M a n a g e m e n t T r a i n i n g

In 1984, Murphy reviewed thirteen pub-lished and unpublished studies on person-al stress management for NIOSH.Although the programmes varied consid-erably in terms of the work groups in-volved, the nature of the techniques andthe outcome measures used, Murphy(1984) was able to make several generalobservations on those programmes andtheir effectiveness.

The majority of the programmes focused ontraining in techniques such as relaxationand other behavioural skills, meditation,


117■

biofeedback, and cognitive restructuring.All the studies reviewed involved some formof relaxation training, and in all but onecase (Peterson, 1981) in combination withcognitive or behavioural skills training. Thisgenerally consisted of a mixture of severaldifferent techniques including assertivenessand personal effectiveness training, cogni-tive restructuring and the reshaping of per-sonal perceptions by logical reasoning. Alltechniques seemed to involve, to some de-gree, strengthening the person’s self es-teem or sense of personal worth. Of the 32outcome measures used in the thirteenstudies, 27 clearly related to the individualand only 3 to the organisation.

Murphy (1984) concluded that a numberof significant benefits accrued to individu-als, including reductions in physiologicalarousal levels, in tension and anxiety, insleep disturbances and in somatic com-plaints. A number of workers also reportedan increased ability to cope with work andhome problems following completion oftheir programme. Not all of these effectswere maintained at follow up testing whichwas usually between 3-9 months later.

Many of these studies are solely reliant onself-report measures and there has been arelative paucity of more objective data inevaluation studies. A study by Ganster etal. (1982) employed both self-report mea-sures of psychological and somatic com-plaints and measures of adrenaline andnoradrenaline levels. In that study, a stressmanagement training program was evalu-ated in a field experiment with 79 publicagency employees who were randomly as-signed to treatment (n = 40) and control (n= 39) groups. The training program con-

sisted of 16 hours of group exposure dis-tributed over 8 weeks. Using proceduresbased on those developed by Meichen-baum (1977), treatment subjects weretaught to recognise and alter their cogni-tive interpretations to stressful events atwork. Subjects were also taught progres-sive relaxation techniques to supplementthis process. Dependent variables wereadrenaline and noradrenaline excretion atwork, anxiety, depression, irritation andsomatic complaints, all measured threetimes (pre-test, post-test and 4 months af-ter treatment). Treatment subjects exhibit-ed significantly lower adrenaline anddepression levels than did controls at thepost-test, and 4 month follow up levels didnot regress to pre-test levels. However,treatment effects were not replicated in asubsequent intervention on the originalcontrol group. The authors did not recom-mend the general adoption of such stressmanagement training programmes.

Murphy (1984) also listed a number of ad-vantages to adopting personal stress man-agement programmes, beyond those forindividual participants:1. They can be established and evaluated

quickly without major disruption towork routines.

2. They can be tailored to individual work-ers’ needs and also contribute to thecontrol of non-work problems.

3. They can link into worker assistanceprogrammes (counselling).

He concluded that the major disadvantageof such programmes is that they are notdesigned to reduce or eliminate sources ofstress at work but only to teach workersmore effective coping strategies. A num-


■118

ber of likely reasons for the imbalance be-tween the number of individual- and or-ganisation-focused stress preventionprogrammes carried out have been out-lined earlier.

The cost-benefit considerations of person-al stress management programmes werenot directly addressed by Murphy in 1984,although he did point out the deliverycosts of the various techniques considered.He concluded that biofeedback was prob-ably the most expensive while meditationwas probably the least expensive. A cost-benefit ratio has been attempted for suchtechniques by Manuso (cited in Schwartz,1980). He calculated that every dollarspent on personal stress management pro-grammes might realise $5.52 in benefitsfor the organisation as a result of de-creased symptom activity and increasedperformance.

E m p l o y e e A s s i s t a n c eP r o g r a m m e s

Employee Assistance Programmes (EAPs),whose origins can be seen in organisa-tions’ concerns over to the cost of alco-holism in the workplace, have flourished inthe United States and Europe. In their nar-rower form, such programmes focus on‘picking up the pieces’ (with counsellingand helplines) for the ‘troubled employee’,addressing drug abuse, personal crises,and marital and family problems. Some arebroader, embracing concerns such as im-pending retirement and relocation. In ef-fect, the range of benefits that could beoffered is infinite. The service may be pro-vided in-house, by consortia or by special-ist EAP contractors.

For example, Cooper et al. (1992a, 1992b)have described the evaluation of a pilotscheme for individual based stress coun-selling in the United Kingdom Post Office.The evaluation was based on a simple pre-/post-test design which compared the psy-chological health and absence behaviourof those using the scheme with a broadlymatched control group of non-partici-pants. Measures of job satisfaction and or-ganisational commitment were also taken.While the authors recognised weaknessesin the design, the data suggested thatcounselling was effective in improving self-reported psychological health and absencefrom work, but not job satisfaction and or-ganisational commitment.

Murphy et al. (1992) point out that the pro-vision and management of such pro-grammes is not as straightforward as it mayappear at first sight: there is a delicate bal-ance between assisting individuals and pro-tecting and promoting the interests oforganisations. Berridge & Cooper (1993) re-fer to this as an “uneasy alliance” where“the balance of interests may well only bemaintained because of the lack of funda-mental analysis of either group’s functionand activities on the part of the other”. Asfar as stress management is concerned,data from an EAP (with individuals’anonymity guaranteed) could be a usefulsource of information, enabling an organi-sation to identify ‘high stress’ departmentsor procedures (Murphy et al., 1992) andperhaps to plan organisational interven-tions.

There has been much interest in the legalimplications of EAP provision in the UnitedStates. Whilst some commentators view


119■

EAPs as a reflection of a general ‘helping’trend in labour relations, and some ques-tion how far EAP provision would haveprogressed without compensation legisla-tion and the union movement (Berridge &Cooper, 1993), others suggest that EAPsrepresent a “legal expedient of providingemployees with a chance, so that the em-ployer who follows EAP to the letter meetsarbitration criteria when firing becomes in-evitable” for problem employees (Nobile,1991). Some have argued that EAP provi-sion may actually increase an organisa-tion’s legal l iability by, for example,opening itself up to accusations of incor-rect assessment of a problem, of inade-quately trained or qualif ied serviceproviders or of unequal access.

Although such programmes have beenlimited by methodological difficulties andby issues of confidentiality, there havebeen claims for considerable financial ad-vantages. In the United States, the GMprogramme, which assists some 100,000employees each year, has been said to savethe company $37 million per year (Feld-man, 1991). A study by the Paul RevereLife Insurance Company claims to show asaving of $4.23 for every dollar spent(Intindola, 1991). Reviewing this area,Berridge and Cooper (1993) point out thatthere has been much criticism of the basisof such claims and much argument as tothe most appropriate method of evalua-tion: cost-benefit analysis, cost effective-ness analysis, utility analysis, peer review,employee attitude surveys or statisticalcase sampling. “In all such evaluation theindependence of the evaluator needs to becombined with the maintenance of confi-

dentiality and the integrity of programmedata. The reconciliation of these require-ments, along with the demands of man-agement, renders the evaluation of EAPsextremely problematic and open to criti-cism from all concerned” (Berridge andCooper, 1993).

One component of broadly based EAPs isoften stress management training. Howev-er, such interventions are usually offeredwithout any link in to counselling or otherforms of employee assistance.


■120

tion), agency (organisation and/or employ-ees) and target (organisation and/or indi-viduals). The scientific literature suggeststhat organisational-level interventions (or,at least, intervention programmes that tar-get the organisation as well as the individ-ual employees) may be the most beneficialfor both individuals and organisations. It isalso frequently argued that stress manage-ment interventions should be evaluated.This is essential for the proper develop-ment of the area. However, a review of thestress management literature reveals thatmost interventions are weak, targetingonly the individual, and that very few areadequately designed or evaluated in scien-tific terms. This section has discussed someof the reasons for this disparity betweenscientific requirements and actual practice.Finally, the three types of interventions(primary, secondary and tertiary) havebeen evaluated here in terms of their ef-fectiveness. The available evidence –de-scribed in detail– suggests that, althoughfew in number, organisational-level inter-ventions that aim to eliminate or controlthe hazards within the work environmenthave significant advantages and representthe best way forward.

7.6S U M M A R Y

Many existing off-the-shelf “stress” sur-veys fail to provide a sufficiently detailedbasis for sound intervention programmes.This section has considered the advan-tages and difficulties in extending existingrisk management paradigms from the fieldof physical hazards to cover psychosocialhazards. Recent studies in the EU and theUSA have described an assessment and in-tervention framework which takes intoconsideration the problems identified inprevious section of this Report and aims toovercome the difficulties of adapting acontrol cycle approach to the manage-ment of work-related stress. This frame-work also includes aspects oforganisational learning and training whichmay bring additional benefits to organisa-tions. Stress management programmeshave been classified according to some ba-sic principles of intervention: objective(prevention, timely reaction, or rehabilita-


121■

8.C O N C L U S I O N S

This section attempts to summarise thefindings of the present Report and identifyareas in need of further research.

RE

SE

AR

CH


■122

vidual, but others relate to the design andmanagement of work and interventions toimprove the work environment.8.1

D E F I N I N G S T R E S S

There is a growing consensus on the defi-nition of stress as a negative psychologicalstate with cognitive and emotional com-ponents, and on its effects on the health ofboth individual employees and their or-ganisations. Furthermore, there are nowtheories of stress which can be used to re-late the experience and effects of workstress to exposure to work hazards and tothe harmful effects on individual and or-ganisational health that such exposuremight cause. Applying such theories to theunderstanding of stress at work allows anapproach to the management of workstress through the application of the no-tion of the control cycle. Such an approachhas proved effective elsewhere in healthand safety. It offers a systematic problem-solving system for continuous improve-ment in relation to work stress. There areseveral distinct areas in which more re-search is required: some relate to the indi-


123■

paradigm for individual differences maynot be capable of providing the necessaryprogress. Could, for example, the conceptof coping be replaced in the literature by,say, that of control? Is all coping an at-tempt to establish perceived control with-in one or more domains of experience–cognition, emotion, physiology or behav-iour? What are the effects of ontologicalvariables such as ageing on coping and theexperience of stress?

8.2I N D I V I D U A L D I F F E R E N C E S :

W O R K A B I L I T Y A N D C O P I N G

The experience of stress is partly depen-dent on the individual’s ability to cope withthe demands placed on them by theirwork, and on the way in which they sub-sequently cope with those demands, andrelated issues of control and support. Moreinformation is required on the nature,structure and effectiveness of individuals’abilities to meet work demands and tocope with any subsequent stress. The needfor more information on coping is widelyrecognised (see, for example, Dewe,2000), but relatively less attention hasbeen paid to the need better to under-stand this concept in relation to those ofwork ability and competence, althoughthis is being flagged in relation to ageingresearch (e.g., Griffiths, 1999a; Ilmarinen& Rantanen, 1999).

It is suggested here that a more radical ap-proach is required, as the present research


■124

ing argued for throughout is better mea-surement procedures, conforming torecognised good practice in relevant areas,and applied within a declared theoreticalcontext.8.3

M E A S U R I N G S T R E S S

More research and development are re-quired in relation to the measurement ofthe experience of stress and related emo-tion and the overall stress process. The in-adequacy of single one-off measures iswidely recognised in the literature but, de-spite this, they continue to be used, andacross studies focused on different aspectsof the stress process. This diversity may ac-count for much of the disagreement with-in stress research. Part of the solution tothis problem lies with agreeing the theo-retical framework within which measure-ment is made, but part lies with thedevelopment of a more adequate technol-ogy of measurement based in ‘good prac-tice’ in a number of areas includingpsychometrics, knowledge elicitation andknowledge modelling. A forced standardi-sation of measurement is not being ar-gued for here and should be resisted for itseffects on scientific progress. What is be-


125■

been too strong a focus on ‘caring for orcuring’ the individual. In many situations,this has reduced the whole issue to one ofpersonnel administration, welfare andcounselling. Second, much of what hasbeen offered, even in this narrow respect,has either a weak theoretical base or hasbeen developed from theory outside occu-pational stress research. Third, there hasbeen a tendency to treat the application ofstress management strategies as a self-contained action and to divorce that appli-cation from any preceding process ofproblem diagnosis. Fourth, stress manage-ment strategies often focus on single typesof intervention and multiple strategies arerarely offered. Last, such interventions arerarely offered for evaluation beyond partic-ipants’ immediate reactions or measures offace validity (see section 8.5).

There are several overarching reasons whythe practice of stress management hasbeen so poor: most relate to the lack of im-pact of contemporary stress theory onpractice. Theory informs practice, andwithout progress in the development oftheory there cannot be a strong logical de-velopment of practice. The lack of impact,in turn, may be accounted for partly by thestagnation of theory referred to in section3, and partly by the lack of a frameworkwhich allows the translation of theory intopractice. As Kurt Lewin put it, “there’snothing so practical as a good theory”.

8.4S T R E S S M A N A G E M E N T

I N T E R V E N T I O N S

There have been a wide variety of differentinterventions which have been advancedas ‘stress management’, and many otherswhich could have been so labelled butwhich have not been. A basic distinctioncan be made between those targeted onthe organisation and those targeted on in-dividual workers, and, among the latter,interventions concerned with white-collarand managerial workers are more com-mon than those concerned with blue-col-lar workers. Various explanations, largelyfocused on economic and political issues,have been advanced to account for thisdifference (see also section 8.6).

A review of the scientific literature sug-gests that there are a number of problemswith research into the management ofwork-related stress. First, too narrow aview has often been taken of what consti-tutes stress management and there has


■126

Evaluation data on stress managementprogrammes are relatively rare. There arerelatively fewer cost-benefit and cost-ef-fectiveness studies compared to studies onthe overall effectiveness of programmes orthe relative effectiveness of their compo-nent parts (see section 7.5). What there issuggests that stress management pro-grammes may be effective in improvingthe quality of working life of workers andtheir immediate psychological health, al-beit self-reported. The evidence relatingsuch interventions to improvements inphysical health is weaker, largely formethodological reasons. There have beenseveral authoritative reviews of organisa-tional and personal stress managementprogrammes in the last ten years reachingbroadly similar conclusions. The publica-tion by the International Labour Organiza-tion in 1992, titled ‘Preventing Stress atWork’, reviews a wide range of differentinterventions, both completed and inprogress, and summarises them in terms ofKarasek’s (1979) model of jobdemands/job decision latitude.

There is an obvious need to encouragetheoretically exciting and methodological-ly adequate research in this area of prac-tice. The main problems, which again arewidely recognised, relate to: the lack of ap-plication of theory to practice, the lack ofa framework for practice, the lack of ade-quately designed and meaningful evalua-tion studies, and the lack of balancebetween the number of individually- andorganisationally-focused interventions.

In summary, it must be concluded that“the jury is still out” on stress manage-ment training: whilst it seems logical that

8.5E V A L U A T I O N O F


Evaluation has been variously defined.Nutt (1981) has talked of it in terms of themeasurement of the degree to which ob-jectives have been achieved, and Green(1974) as the comparison of an object ofinterest against a standard of acceptability.In contrast to basic research, evaluationimplies and requires from the onset criteriaand procedures for making judgements ofmerit, value or worth (Scriven, 1967).

There are three common purposes forevaluations of stress management pro-grammes. The first is to ask whether theprogramme is effective; specifical lywhether the programme objectives are be-ing met. A second purpose is to determinethe efficiency or comparative effectivenessof two or more programmes or methodswithin a programme. The third purpose isto assess the cost-benefit or the cost-ef-fectiveness of the programme.


127■

such interventions should promote em-ployee health, there are not yet sufficientdata to be confident that they do. Howev-er, the evidence for employee assistanceprogrammes –particularly those broadlyconceived to include health promotion inthe workplace– may be more encourag-ing, although that which relates to coun-selling alone is weak. The provision ofcounselling is largely designed to assistemployees who are already experiencingproblems, and is, in that sense, post hoc.Stressor reduction / hazard control is, forseveral reasons, the most promising areafor interventions, although, again, there isnot yet sufficient information to be confi-dent about the nature and extent of theireffectiveness. To date, such conclusionsare based more on moral and strategic rea-soning than on empirical data, althoughthe data that do exist are supportive. Whatcan be firmly concluded, however, is thatthere is still a need for further and moreadequate evaluation studies. Perhaps thekey to the methodology and evaluation ofintervention studies is a re-appraisal of thevalue of the natural science paradigm infield research (Griffiths, 1999b). One issuemight be the inappropriateness of theevaluation paradigm itself.


■128

tions focused on the design of the workenvironment, and Murphy (1988) notesthat given the varieties of work stressorsthat have been identified, many othertypes of action relating to organisationaland work development should be effectivein reducing work stress. Van der Hek &Plomp (1997) also concluded that “there issome evidence that organization-wide ap-proaches show the best results on individ-ual, individual-organizational interfaceand organizational parameters [outcomemeasures]; these comprehensive pro-grammes have a strong impact on the en-tire organization, and require the fullsupport of management”. The emergingevidence is strong enough for the UnitedStates’ National Institute for OccupationalSafety and Health (NIOSH) to have identi-fied “the organization of work” as one ofthe national occupational safety andhealth priority areas (Rosenstock, 1997).As part of their National Occupational Re-search Agenda (NORA), NIOSH intend tofocus research on issues such as the impactof work organisation on overall health, theidentification of healthy organisation char-acteristics and the development of inter-vention strategies.

What is not clear from the evaluation liter-ature is the exact mechanism by whichsuch interventions, and particularly thosefocused on the individual, might affecthealth. Often, where different types of in-dividually focused interventions have beencompared, there is no evidence that anyone or any combination is better than anyother. This indicates that there may be ageneral, non-specific effect of intervening.The fact of an intervention may be benefi-

8.6I N D I V I D U A L - A N D

O R G A N I S A T I O N A L - L E V E L


Unfortunately, there are very few well de-signed and evaluated interventions avail-able in the literature to date (see section 7).Nonetheless, Murphy et al. (1992) con-clude that “job redesign and organisation-al change remain the preferred approachesto stress management because they focuson reducing or eliminating the sources ofthe problem in the work environment”.However, they also point out that such ap-proaches require a detailed audit of workstressors and a knowledge of the dynamicsof organisational change if unwelcomeoutcomes are to be minimised. Moreover,such interventions can be expensive andmore difficult and disruptive to design, im-plement and evaluate –factors which maymake them less popular alternatives to sec-ondary and tertiary interventions.

Nonetheless, Landy (1992) has sum-marised a number of possible interven-


129■

cial, rather than its exact content. Inter-views with managers responsible for intro-ducing such interventions suggest thatthey are aware of such effects (see, for ex-ample, Cox et al., 1988). It is thereforepossible that at least part of the effects ofstress management programmes is due tothe way they alter workers’ perceptions ofand attitudes to their organisations, andhence organisational culture. It was ar-gued earlier that poor organisational cul-ture might be associated with an increasedexperience of stress, while a good organi-sational culture might weaken or “buffer”the effects of stress on health. A definingfactor for organisational culture is the sizeof the enterprise, and this should be bornein mind when considering interventionand evaluation issues, together with thewider context in terms of the socio-eco-nomic environment in the Member States.


■130

adaptation and application of a control cy-cle approach such as that made explicit incontemporary models of risk management(see section 7.1). This is already happeningin several countries of the EuropeanUnion, for example in the United Kingdom(Cox et al., 2000; Griffiths et al., 1996), theNetherlands (Kompier et al., 1998) andFinland (Elo, 1994). In different countriesthis approach is given different names, anda wide variety of local arguments are de-ployed to support its use. However, the un-derlying philosophy is the same, and thisapproach offers the best way forward.

The final comment concerns the maturityof stress research as an area of applied sci-ence. Two things must be apparent to theinformed reader of this Report. First, thereis a wealth of scientific data on workstress, its causes and effects, and on someof the mechanisms underpinning the rela-tionships among these. More general re-search is not needed. What is required isan answer to the outstanding method-ological questions, and to more specificquestions about particular aspects of theoverall stress process and its underpinningmechanisms. Second, although this wealthof scientific data exists, it still needs to betranslated into practice, and the effective-ness of this practice evaluated. This is an-other set of needs, and one that will onlybe settled outside the laboratory andthrough the development of consensusand eventually common practice.

While stress at work will remain a majorchallenge to occupational health, our abil-ity to understand and manage that chal-lenge is improving. The future looksbright.

8.7O V E R A L L C O N C L U S I O N S

The evidence on the effectiveness of stressmanagement interventions reviewed inthis Status Report is promising. The avail-able data, although sparse, suggest thatinterventions, especially at the organisa-tional level (e.g., Cox et al., 2000; Gansteret al., 1982; Shinn et al., 1984; Dollard &Winefield, 1996; Kompier et al., 1998), arebeneficial to both individual and organisa-tional health and should be investigated–and evaluated– further.

The strategic argument for the manage-ment of work stress advanced in this Re-port on the basis of the available scientificevidence and current legal thinking in Eu-rope is that work stress is a current and fu-ture health and safety issue, and, as such,should be dealt with in the same logicaland systematic way as other health andsafety issues. That is, the management ofstress at work should be based on the


131■

9.R E F E R E N C E S

Abel, T.M., Metraux, R., & Roll, S. (1987)Psychotherapy and Culture. University ofNew Mexico Press, Albuquerque.

Ader, R. A. (1981) Psychoneuroimmunolo-gy. Academic Press, New York.

Akerstedt, T., & Landstrom, U. (1998)Work place countermeasures of night shiftfatigue International Journal Of IndustrialErgonomics, Vol.21, No.3-4, pp.167-178

Ahasan M.R., Mohiuddin G., Vayrynen S.,Ironkannas H., & Quddus R. (1999) Work-related problems in metal handling tasks inBangladesh: obstacles to the developmentof safety and health measures. Ergonom-ics, Vol.42, No.2, pp.385-396

Ahlbom, A., Karasek, R.A., & Theorell, T.(1977) Psychosocial occupational de-mands and risk for cardio-vascular death.Lakartidningen, 77, 4243-4245.

Althouse, R., & Hurrell, J. J. (1977) AnAnalysis of Job Stress in Coal Mining. De-partment of Health, Education and Wel-fare (NIOSH) publication no: 77-217, USGovernment Printing Office, WashingtonDC.

Angus, R. G., & Heslegrave, R. J. (1983)The effects of sleep loss and sustainedmental work: implications for commandand control performance. In: J. Ernsting(ed) Sustained Intensive Air Operations:Physiological and Performance Aspects.NATO-AGARD Cp - 338. Technical Editing& Reproduction Ltd., London.

Anisman, H., Pizzion, A., & Sklar, L.S.(1980) Coping with stress, norepinephrinedepletion, and escape performance. BrainResearch, 191, 583-588.

Antelman, S.M. & Caggiula, A.R. (1977)Norephinephrine-dopamine interactionsand behaviour. Science, 195, 646-653.

Appley, M. H., & Trumbull, R. (1967) Psy-chological Stress. Appleton-Century-Crofts, New York.

Arnestad, M., & Aanestad, B. (1985) Workenvironment at a psychiatric ward: stress,

RE

SE

AR

CH


■132

health and immunoglobulin levels. Unpub-lished PhD thesis, University of Bergen,Bergen.

Arroba, T., & James, K. (1990) Reducingthe cost of stress: an organizational mod-el. Personnel Review, 19, 21-27.

Arthur H.M., Garfinkel P.E., Irvine J. (1999)Development and testing of a new hostili-ty scale. Canadian Journal Of Cardiology,Vol.15, No.5, pp.539-544

Aspinwall,L.G., Taylor,S.E. (1997) A stitchin time: Self-regulation and proactive cop-ing. Psychological Bulletin, 121, 417-436

Averill, J.R. (1973) Personal control overaversive stimuli and its relationship tostress. Psychological Bulletin, 80, 286-303.

Bacharach, S. B., Bamberger, P.B., & Con-ley, S. (1991) Work-home conflict amongnurses and engineers: mediating the im-pact of role stress on burnout and satisfac-tion at work. Journal of OrganizationalBehaviour, 12, 39-53.

Baddeley, A. D. (1972) Selective attentionand performance in dangerous environ-ments. British Journal of Psychology, 63,537-546.

Bailey, J. M., & Bhagat, R. S. (1987) Mean-ing and measurement of stressors in thework environment. In: S. V. Kasl & CLCooper (eds) Stress and Health: Issues inResearch Methodology. Wiley & Sons,Chichester.

Baker, D. B. (1985) The study of stress atwork. Annual Review of Public Health, 6,367 -381.

Bandura, A. (1977) Self-efficacy: towards aunifying theory of behavioural change.Psychological Review, 84, 191-215.

Barefoot, J. C., Dahlstrom, W. G., &Williams, R. B. (1983) Hostility, CHD inci-dence, and total mortality: a 25 year fol-low up study of 255 physicians.Psychomatic Medicine, 45, 83-90.

Barreto S.M., Swerdlow A.J., Smith PG,Higgins CD (1997) Risk of death from mo-tor-vehicle injury in Brazilian steelworkers:A nested case-control study. InternationalJournal Of Epidemiology, Vol.26, No.4,pp.814-821

Barth, P.S. (1990) Workers’ compensationfor mental stress cases. Behavioural Sci-ences and the Law, 8, 358.

Baruch, G. K., & Barnett, R. E. (1987) Rolequality and psychological well-being. In: F.J. Cobb (ed) Spouse, Parent, Worker: OnGender and Multiple Roles. Yale UniversityPress, New Haven, Conneticut.

Bastiaans, J. (1982) Consequences ofmodern terrorism. In L. Goldberger &S.Breznitz (eds) Handbook of Stress: Theo-retical and Clinical Aspects. Free Press,New York.

Beale, D., Clarke, D., Cox, T., Leather, P. &Lawrence, C. (1999) System memory in vi-olent incidents: Evidence from patterns ofreoccurrence. Journal of OccupationalHealth Psychology, 4(3), 233-244.

Beale, D., Cox, T., Clarke, D., Lawrence, C.& Leather, P. (1998) Temporal architectureof violent incidents. Journal of Occupa-tional Health Psychology, 3, 65-82.


133■

Beattie, R. T., Darlington, T. G., & Cripps,D. M. (1974) The Management Threshold.British Institute of Management Paper, no:OPN 11., BIM, London.

Beckham, E., & Adams, R. (1984) Copingbehavior in depression: report on a newscale. Behavioral Research and Therapy,22, 71-75.

Beehr, T.A. (1995) Psychological Stress inthe Workplace. New York: Routledge.

Beehr, T.A. & Newman, J.E. (1978) Jobstress, employee health, and organization-al effectiveness: a facet analysis, modeland literature review. Personnel Psycholo-gy, 31, 665-699.

Beehr, T.A., & O’Hara, K. (1987) Method-ological designs for the evaluation of oc-cupational stress interventions. In: S. Kasl& C. Cooper (eds) Stress and Health: Issuesin Research Methodology. Wiley & Sons,Chichester.

Bergan, T., Vaernes, R. J., Ingebrigsten, P.,Tonder, O., Aakvaag, A., & Ursin, H. (1987)Relationships between work environmen-tal problems and health among Norwe-gian divers in the North Sea. In: A. Marroni& G. Oriani (eds) Diving and HyperbaricMedicine. Academic Press, New York.

Berridge, J. & Cooper, C.L. (1993) Stressand coping in US organizations: the role ofthe Employee Assistance Programme.Work & Stress, 7, 89-102.

Bettenhausen, K.L. (1991) 5 years ofgroups research - what we have learnedand what needs to be addressed. JournalOf Management, 17 (2), pp.345-381

Bhagat, R. S., & Chassie, M. B. (1981) De-terminants of organizational commitmentin working women: some implications fororganizational integration. Journal of Oc-cupational Behaviour, 2, 17-30.

Bhalla, S., Jones, B., & Flynn, D. M. (1991)Role stress among Canadian white-collarworkers. Work & Stress, 5, 289-299.

Biersner, R.J., Gunderson, E.K., Ryman,D.H., & Rahe, R.H. (1971) Correlations ofPhysical Fitness, Perceived Health Status,and Dispenssary Visits with Performance inStressful Training. USN Medical Neuropsy-chioatric Research Unit. Technical reportno: 71-30. US Navy, Washington DC.

Blank, A.S. (1981) The price of constantvigilance: the Vietman era veteran. Fron-tiers of Psychiatry, 11, Feb.

Blohmke, M., & Reimer, F. (1980)Krankheit und Beruf. Alfred Huthig Verlag,Heidelberg.

Boggild H. & Knutsson A. (1999) Shiftwork, risk factors and cardiovascular dis-ease. Scandinavian Journal Of Work Envi-ronment & Health, Vol.25, No.2, pp.85-99

Borella P., Bargellini A., Rovesti S., PinelliM., Vivoli R., Solfrini V., & Vivoli G. (1999)Emotional stability, anxiety, and naturalkiller activity under examination stress Psy-choneuroendocrinology, Vol.24, No.6,pp.613-627

Borg, M.G. (1990) Occupational stress inBritish educational settings: A review. Edu-cational Psychology, 10 (2) 103-126.

Bosma, H., & Marmot, M.G. (1997) Lowjob control and risk of coronary heart dis-


■134

ease in Whitehall II (prospective cohort)study. British Medical Journal, 314, no.7080

Bowers, K.S. (1973) Situationalism in psy-chology: an analysis and critique. Psycho-logical Review, 80, 307-335.

Bradley, G. (1989) Computers and the Psy-chological Work Environment. Taylor andFrancis, London.

Brady, J.V. (1958) Ulcers in ‘executive’monkeys. Scientific American, 199, 95-100.

Brady, J.V. (1975) Toward a behavioural bi-ology of emotion. In L. Levi (ed) Emotions:Their Parameters and Measurement.Raven, New York.

Brady, J.V. & Harris, A.H. (1977) The exper-imental production of altered physiologicalstates. In W. Honig & J.E.R. Staddon (eds)Handbook of Operant Behaviour. Prentice-Hall, Engelwood Cliffs, California.

Brener, J. (1978) Visceral perception. In: J.Beatty (ed) Biofeedback and Behaviour: ANATO Symposium. Plenum Press, NewYork.

Breslow, L., & Buell, P. (1960) Mortalityfrom coronary heart disease and physicalactivity of work in California. Journal ofChronic Diseases, 22, 87-91.

Briner, R. (1997) Improving stress assess-ment: Toward an evidence-based ap-proach to organizational stressinterventions. Journal of PsychosomaticResearch, 43 (1), 61-71

British Psychological Society (1992) Guide-lines for the Prevention and Managementof Violence at Work. British PsychologicalSociety, Leicester.

Broadbent, D.E. (1971) Decision andStress. Academic Press, New York.

Broadbent, D.E., & Gath, D. (1981) Illhealth on the line: sorting myth from fact.Employment Gazette, 89, no. 3.

Brook, A. (1973) Mental stress at work.Practitioner, 210, 500-506.

Buck, V. (1972) Working Under Pressure.Staples Press, London.

Bundesministerium für Arbeit undSozialordnung, 1999. Reported occupa-tional diseases in Germany (1960-1997).Bundesministerium für Arbeit undSozialordnung.

Burke, R.J. (1986) Occupational and lifestress and family: conceptual frameworksand research findings. International Re-view of Applied Psychology, 35, 347-369.

Burke, R.J. (1993) Organizational-level in-terventions to reduce occupational stres-sors. Work and Stress, 7(1), 77-87.

Canadian Mental Health Association(1984) Work and Well-being: The Chang-ing Realities of Employment. Toronto, On-tario.

Cannon, W.B. (1929) Bodily Changes inPain, Hunger, Fear and Rage: An Accountof Recent Researches in the Function ofEmotional Excitement. Appleton, NewYork.


135■

Cannon, W.B., (1931) The Wisdom of theBody. Norton, New York.

Caplan, R.D., Cobb, S., French, J. R. P., vanHarrison, R., & Pinneau, S. R. (1975) JobDemands and Worker Health. US Depart-ment of Health, Education and WelfarePublication no: NIOSH 75-160, US Govern-ment Printing Office, Washington DC.

Carayon, P. (1993) A longitudinal test ofKarasek’s Job Strain model among officeworkers. Work & Stress, Vol.7, No.4,pp.299-314

Cartwright, S. & Cooper, CL (1996) Publicpolicy and occupational health psychologyin Europe. Journal of Occupational HealthPsychology, I (4), 349-361.

Chappell, D. & Di Martino, V. (1998) Vio-lence at Work. Geneva: InternationalLabour Office.

Chatterjee, D.S. (1987) Repetition straininjury - a recent review. Journal of the So-ciety of Occupational Medicine, 37, 100-105.

Chatterjee, D.S. (1992) Workplace upperlimb disorders : a prospective study withintervention. Occupational Medicine, 42,129-136.

Chen, P.Y. & Spector, P.E. (1991) Negativeaffectivity as the underlying cause of cor-relations between stressors and strains.Journal of Applied Psychology, 76, 398-407.

Cincirpini, P. M., Hook, J. D., Mendes deLeon, C. F., & Pritchard, W. S. (1984) A Re-view of cardiovascular, Electromyographic,Electrodermal and Respiratory Measures of

Psychological Stress. National Institute forOccupational Safety and Health, contractno: 84-257, Cincinnati, Ohio.

Cobb, S., & Kasl, S. V. (1977) Termination:the Consequences of Job Loss. US Depart-ment of Health, Education and Welfare,Cincinnati.

Cohen, A. (1969) Effects of noise on psy-chological state. In: W. D. Ward & J. E.Fricke (eds) Noise as a Public Health Haz-ard. American Speech and Hearing Associ-ation, Washington DC.

Cohen, A. (1974) Industrial noise andmedical, absence and accident record dataon exposed workers. In: W. D. Ward (ed)Proceedings of the International Congresson Noise as a Public Health Problem. USEnvironmental Protection Agency, Wash-ington DC.

Cohen, A. (1976) The influence of a com-pany hearing conservative program on ex-tra-auditory problems in workers. Journalof Safety Research, 8, 146-162.

Cohen, S. (1980) After effects of stress onhuman performance and social behaviour:a review of research and theory. Psycho-logical Bulletin, 88, 82-108.

Cohen, S.G. & Ledford, G.E. (1994) the ef-fectiveness of self-managing teams - aquasi-experiment. Human Relations, 47(1), pp.13-43

Cohen, S., & Willis, T. A. (1985) Stress, so-cial support and the buffering hypothesis.Psychological Bulletin, 98, 310-357.

Colligan, M.J., Smith, M.J., & Hurrell, J.J.(1977) Occupational incidence rates of


■136

mental health disorders. Journal of HumanStress, 3, 34-39.

Confederation of British Industry [CBI](1999) Absence Bill Of £10 Billion For Busi-ness In 1998 - CBI Survey. Confederationof British Industry, London.

Cooper, C.L. (1978) Work stress. In: P. B.Warr (ed) Psychology at Work. Penguin,Harmondsworth.

Cooper, C.L. (1981) Executive Families Un-der Stress. Prentice Hall, Englewood Cliffs,New Jersey.

Cooper, C.L. & Cartwright, S. (1997) An in-tervention strategy for workplace stress.Journal of Psychosomatic Research, 43 (1),7-16

Cooper, C.L., & Davidson, M. (1982) HighPressure: Working Lives of Women Man-agers. Fontana, London.

Cooper, C.L., & Hingley, P. (1985) TheChange Makers. Harper and Row, London.

Cooper, C.L., & Marshall, J. (1976) Occu-pational sources of stress: a review of theliterature relating to coronary heart diseaseand mental ill health. Journal of Occupa-tional Psychology, 49, 11-28.

Cooper, C.L., & Smith, M. J. (1986) JobStress and Blue Collar Work. Wiley & Sons,Chichester.

Cooper, C.L. & Williams, S. (1997) Creat-ing healthy work organizations. Chich-ester: John Wiley & Sons

Cooper, C.L., Liukkonen, P. & Cartwright,S. (1996) Stress prevention in the work-

place: assessing the costs and benefits toorganisations. Dublin: European Founda-tion for the Improvement of Living andWorking Conditions.

Cooper, C.L., Allison, T., Reynolds, P., &Sadri, G. (1992a) An individual-basedcounselling approach for combating stressin British Post Office employees. In Interna-tional Labour Office, Conditions of WorkDigest (Vol. 11): Preventing Stress at Work.International Labour Office, Geneva.

Cooper, C.L., Sadri, G., Allison, T., &Reynolds, P. (1992b) Stress counselling inthe Post Office. Counselling PsychologyQuarterly, 3, 3-11.

Corey, D.M. & Wolf, G.D. (1992) An inte-grated approach to reducing stress in-juries. In J.C. Quick, L.R. Murphy, & J.J.Hurrell (eds) Stress and Well-being atWork.

Council Directive 89/391/EEC of 12 June1989 on the introduction of measures toencourage improvements in the safety andhealth of workers at work. Official JournalL 183, 29/06/1989 p. 0001 – 0008 InternetWWW page at http://europa.eu.int/eur-lex/en/lif/dat/1989/en_389L0391.html

Council Directive 98/24/EC of 7 April 1998on the protection of the health and safetyof workers from the risks related to chemi-cal agents at work (fourteenth individual Di-rective within the meaning of Article 16(1)of Directive 89/391/EEC). Official Journal L131, 05/05/1998 p. 0011 – 0023 InternetWWW page at http://europa.eu.int/eur-lex/en/lif/dat/1998/en_398L0024.html


137■

Cox, S., & Tait, R. (1991) Safety, Reliabilityand Risk Management. ButterworthHeinemann, London.

Cox, S., Cox, T., Thirlaway, M., & Mackay,C. J. (1985) Effects of simulated repetitivework of urinary catecholamine excretion.Ergonomics, 25, 1129-1141.

Cox, T. (1978) Stress. Macmillan, London.

Cox, T. (1980) Repetitive work. In: CLCooper & R. Payne (eds) Current Concernsin Occupational stress. Wiley & Sons,Chichester.

Cox, T. (1984) Stress: a psychophysiologi-cal approach to cancer. In: CL Cooper (ed)Psychosocial Stress and Cancer. Wiley &Sons, Chichester.

Cox, T. (1985a) The nature and measure-ment of stress. Ergonomics, 28, 1155-1163.

Cox, T. (1985b) Repetitive work: occupa-tional stress and health. In: CL Cooper &M. J. Smith (eds) Job Stress and Blue CollarWork. Wiley & Sons, Chichester.

Cox, T. (1987). Stress, coping and problemsolving. Work & Stress, 1, 5-14.

Cox, T. (1988a) Psychobiological factors instress and health. In S. Fisher & J. Reason(eds) Handbook of Life Stress, Cognitionand Health. Wiley & Sons, Chichester.

Cox, T. (1988b) AIDS and stress. Work &Stress, 2, 109-112.

Cox, T. (1990) The recognition and mea-surement of stress: conceptual andmethodological issues. In: E. N. Corlett & J.

Wilson (eds) Evaluation of Human Work.Taylor & Francis, London.

Cox, T. (1993) Stress Research and StressManagement: Putting theory to work.Sudbury: HSE Books.

Cox, T., & Cox, S. (1985) The role of theadrenals in the psychophysiology of stress.In: E. Karas (ed) Current Issues in ClinicalPsychology. Plenum Press, London.

Cox, T. & Cox, S. (1992) Mental health atwork: assessment and control. In R. Jenk-ins & N. Coney (eds) Prevention of MentalIll Health at Work. HMSO, London.

Cox, T., & Cox, S. (1993) Psychosocial andOrganizational Hazards: Monitoring andControl. Occasional Series in OccupationalHealth, No.5. World Health Organization(Europe), Copenhagen, Denmark.

Cox, T., & Ferguson, E. (1991) Individualdifferences, stress and coping. In: CLCooper & R. Payne (eds) Personality andStress. Wiley & Sons, Chichester.

Cox, T., & Ferguson, E. (1994) Measure-ment of the subjective work environment.Work & Stress, 8 (2), 98-109

Cox, T. & Griffiths, A.J. (1994) The natureand measurement of work stress: Theoryand practice. In N. Corlett & J. Wilson(Eds.) Evaluation of Human Work: A Prac-tical Ergonomics Methodology. London:Taylor and Francis.

Cox, T. & Griffiths, A.J. (1995) The assess-ment of psychosocial hazards at work. InM.J. Shabracq, J.A.M. Winnubst & CLCooper (Eds.) Handbook of Work and


■138

Health Psychology. Chichester: Wiley &Sons

Cox, T., & Howarth, I. (1990) Organiza-tional health, culture and helping. Work &Stress, 4, 107-110.

Cox, T., & Kuk, G (1991) Healthiness ofschools as organizations: teacher stressand health. Paper to: International Con-gress, Stress, Anxiety & Emotional Disor-ders, University of Minho, Braga, Portugal.

Cox, T., & Leather, P. (1994) The preventionof violence at work: application of a cogni-tive behavioural theory. In: CL Cooper & I.Robertson (eds) International Review of In-dustrial and Organizational Psychology,Wiley & Sons, Chichester.

Cox, T., & Leiter, M. (1992) The health ofhealthcare organizations. Work & Stress,6, 219-227.

Cox, T., & Mackay, C. J. (1981) A transac-tional approach to occupational stress. In:E. N. Corlett and J. Richardson (eds) Stress,Work Design and Productivity. Wiley &Sons, Chichester.

Cox, T., & Mackay, C. J. (1982) Psychoso-cial factors and psychophysiological mech-anisms in the aetiology and developmentof cancers. Social Science and Medicine,16, 381-396.

Cox, T., & Mackay, C.J. (1985) The mea-surement of self-reported stress andarousal. British Journal of Psychology, 76,183-186.

Cox, T. & Thomson, L. (2000) Organisa-tional healthiness: work-related stress andemployee health. In P. Dewe, M. Leiter & T.

Cox (Eds.) Coping, Health and Organisa-tions. London: Taylor & Francis.

Cox, T., Cox, S., & Thirlaway, M. (1983)The psychological and physiological re-sponse to stress. In: A. Gale & J. A. Ed-wards (eds) Physiological Correlates ofHuman Behaviour. Academic Press, Lon-don.

Cox, T., Ferguson, E., & Farnsworth, W. F.(1993) Nurses’ knowledge of HIV and AIDSand their perceptions of the associated riskof infection at work. Paper to: VI EuropeanCongress on Work and Organizational Psy-chology, Alicante.

Cox, T., Leather, P., & Cox, S. (1990) Stress,health and organizations. OccupationalHealth Review, 23, 13-18.

Cox, T., Watts, C., & Barnett, A. (1981) TheExperience and Effects of Task-InherentDemand. Final technical report to the USArmy Research, Development and Stan-dardization Group (UK).

Cox, T., Griffiths, A.J., Barlow, C.A., Ran-dall, R.J., Thomson, L.E. & Rial-Gonzalez,E. (2000) Organisational interventions forwork stress: a risk management approach.HSE Books, Sudbury.

Crown, S., Crown, J.M., & Fleming, A.(1975) Aspects of the psychology and epi-demiology of rheumatoid disease. Psycho-logical Medicine, 5, 291-299.

Danna, K. & Griffin R.W. (1999) Health andwell-being in the workplace: A review andsynthesis of the literature. Journal Of Man-agement, Vol.25, No.3, pp.357-384


139■

Davidson, M.J., & Cooper, C.L. (1981) Amodel of occupational stress. Journal ofOccupational Medicine, 23, 564-570.

Davidson, M.J., & Cooper, C.L. (1983)Working women in the European Commu-nity – the future prospects. Long RangePlanning, 16, 49-54.

Davidson, M. J., & Earnshaw, J. (1991) Vul-nerable Workers: Psychosocial and LegalIssues. Wiley & Sons, Chichester.

Davies, N.V. & Teasdale, P (1994) The coststo the British economy of work accidentsand work-related ill health. Sudbury: HSEBooks.

DeFrank, R.S., & Cooper, C.L. (1987)Worksite management interventions: theireffectiveness and conceptualization. Jour-nal of Managerial Psychology, 2, 4-10.

Deloitte & Touche Consulting Group(1999) Call Centre Agent Report: A surveyon Profit and Productivity. Deloitte ToucheTohmatsu, Melbourne. Internet WWWpage at http://www.deloitte.com.au/con-tent/call_centre_survey98.asp

Dembrowski, T.M., MacDougall, J. M.,Williams, R. B., Haney, T. L., & Blumenthal,J. A. (1985) Components of Type A hostili-ty and anger-in: relationship to angio-graphic findings. Psychomatic Medicine,47, 219-233.

de Rijk, A. J., le Blanc, P. M., Schaufeli, W.B., & de Jonge, J. (1998) Active coping andneed for control as moderators of the jobdemand-control model: effects onburnout. Journal of Occupational and Or-ganizational Psychology, 71, 1-18.

Devereux J., Buckle P. & Vlachonikolis I.G.(1999) Interactions between physical andpsychosocial risk factors at work increasethe risk of back disorders: an epidemiolog-ical approach. Occupational and Environ-mental Medicine, vol. 56, no. 5, pp.343-353

Dewe, P. (1987) New Zealand ministers ofreligion: identifying sources of stress andcoping strategies. Work & Stress, 1, 351-363.

Dewe, P. (1991) Measuring work stressors:the role of frequency, duration and de-mand. Work & Stress, 5, 77-91.

Dewe, P. (1993) Work, stress and coping:common pathways for future research?Work & Stress, 7 (1), 1-3.

Dewe, P. (2000) Measures of coping withstress at work: a review and critique. In P.Dewe, M. Leiter & T. Cox (Eds.) Coping,Health and Organisations. London: Taylor& Francis.

Dewe, P., Cox, T., & Ferguson, E. (1993) In-dividual strategies for coping with stress atwork: a review of progress and directionsfor future research. Work & Stress, 7 (1), 5-15.

Dewe, P., Leiter, M & Cox, T (Eds.) (2000)Coping, Health and Organisations. Lon-don: Taylor & Francis.

Diament, J., & Byers, S. O. (1975) A precisecatecholamine assay for small samples.Journal of Laboratory and Clinical Medi-cine, 85, 679-693.

Dimsdale, J. E., & Moss, J. (1980a) Plasmacatecholamines in stress and exercise.


■140

Journal of the American Medical Associa-tion, 243, 340-342.

Dimsdale, J. E., & Moss, J. (1980b) Short-term catecholamine response to psycho-logical stress. Psychosomatic Medicine, 42,493-497.

Dohrenwend, B. S., & Dohrenwend, B. P.(1974) Stressful Life Events: Their Natureand Effects. Wiley & Sons, New York.

Dohrenwend, B.S., Krasnoff, L., Askenasy,A.R. & Dohrenwend, B.P. (1982) The psy-chiatric epidemiology research interviewlife events scale

Dohrenwend, B. S., Krasnoff, L., Askenasy,A. R., & Dohrenwend, B. P. (1988) The Psy-chiatric Epidemiology Research InterviewLife Events Scale. In: L. Goldberg & S.Breznitz (eds) Handbook of Stress: Theo-retical and Clinical Aspects. Free Press,New York.

Dollard, M.F. & Winefield, A.H. (1996)Managing occupational stress: a nationaland international perspective. Internation-al Journal of Stress Management, 3 (2),69-83.

Donaldson, J., & Gowler, D. (1975) Perog-atives, participation and managerial stress.In: D. Gowler & K. Legge (eds) ManagerialStress. Gower Press, London.

Douglas, M. (1992) Risk and Blame. Rout-ledge, London.

Eaton, W.W., Anthony, J.C., Mandel, W., &Garrison, R. (1990) Occupation and preva-lence of major depressive disorder. Journalof Occupational Medicine, 32, 1079-1086.

Edwards, J.R. & Cooper, C.L. (1990) Theperson-environment fit approach to stress:recurring problems and some suggestedsolutions. Journal of Organizational Be-haviour, 11, 293-307.

Einhorn, H.J., & Hogarth, R.M. (1981) Be-havioural decision theory: processes ofjudgement and choice. Annual Review ofPsychology, 32, 53-88.

Ekehammer, B. (1974) Interactionism inpersonality from a historical perspective.Psychological Bulletin, 81, 1026.

Elkin, A.J. & Rosch, P.J. (1990) Promotingmental health at the workplace: the pre-vention side of stress management. Occu-pational Medicine State of the Art review.5(4), 739-754

Ellertsen, B., Johnsen, T. B., & Ursin, H.(1978) Relationship between the hormon-al responses to activation and coping. In:H. Ursin, E. Baade, and S. Levine (eds) Psy-chobiology of Stress: A Study of CopingMen. Academic Press, New York.

Elo, A.-L. (1986) Assessment of PsychicStress Factors at Work. Institute of Occu-pational Health, Helsinki.

Elo, A.-L. (1994) Assessment of mentalstress factors at work. Occupational Medi-cine, 945-959

Endresen, I.M., Ellertsen, B., Endresen, C.,Hjelmen, A.M., Matre, R., & Ursin, H.(1991) Stress at work and psychologicaland immunological parameters in a groupof Norwegian female bank employees.Work & Stress, 5, 217-227.


141■

Endresen, I. M., Vaernes, R. J., Ursin, H., &Tonder, O. (1987) Psychological stress fac-tors and concentration of immunoglobu-lins and complement components inNorwegian nurses. Work & Stress, 1, 365-375.

Ertel, M., Junghanns, G., Pech, E., &Ullsperger, P. (1997) Auswirkungen derBildschirmarbeit auf Gesundheit undWohlbefinden. Wirtschaftsverlag NW, Bre-merhaven. (Schriftenreihe der Bunde-sanstalt für Arbeitsmedizin: Forschung, Fb762)

European Agency for Safety and Health atWork (1998) Priorities and Strategies inOSH Policy in the Member States of theEU. European Agency for Safety andHealth at Work, Bilbao. Internet WWWpage at http://agency.osha.eu.int/re-ports/priorities (available in 8 EU lan-guages)

European Agency for Safety and Health atWork (1999) Work-related neck and upperlimb musculoskeletal disorders. Luxem-bourg: Office for Official Publications ofthe European Communities.

European Commission (1996) Guidanceon risk assessment at work. EuropeanCommission, Brussels.

European Foundation for the Improve-ment of Living and Working Conditions(1992) European Survey on the Work Envi-ronment . Dublin, Ireland.

European Foundation for the Improve-ment of Living and Working Conditions(1996) Second European Survey on Work-

ing Conditions in the European Union.Dublin, Ireland

European Foundation for the Improve-ment of Living and Working Conditions(1997) European Working Environment inFigures. Dublin, Ireland

Eysenck, M. W. (1983) Anxiety and individ-ual differences. In: G. R. J. Hockey (ed)Stress and Fatigue in Human Performance.Wiley & Sons, Chichester.

Feather, N.T. (1990) The Psychological Im-pact of Unemployment. Springer-Verlag,New York.

Feldman, S. (1991) Today’s EAP’s make thegrade. Personnel, 68, 3-40.

Ferguson, D. (1973) A study of occupa-tional stress and health. Ergonomics, 16,649-663.

Ferguson, E., & Cox, T. (1993) Exploratoryfactor analysis: a user’s guide. Internation-al Journal of Selection and Assessment, 1(2), 84-94

Fielden S.L & Peckar C.J. (1999) Workstress and hospital doctors: a comparativestudy. Stress Medicine, vol. 15, no. 3, pp.137-141

Figley, C.R. (1985) Trauma and Its Wake:The Study of Treatment of Post TraumaticStress Disorder. Brunner/Mazel, New York.

Fisher, S. (1986). Stress and Strategy.Lawrence Erlbaum Associates, London.

Fisher, S. (1996) Life change, personal con-trol and disease. South African Journal ofPsychology, Vol.26, No.1, pp.16-22


■142

Flanagan, P., McAnally, K.I., Martin, R.L.,Meehan, J.W., & Oldfield S.R. (1998) Au-rally and visually guided visual search in avirtual environment. Human Factors,Vol.40, No.3, pp.461-468.

Fletcher, B. C. (1988) The epidemiology ofoccupational stress. In: C.L. Cooper & R.Payne (eds) Causes, Coping and Conse-quences of Stress at Work. Wiley & Sons,Chichester.

Folger, R., & Belew, J. (1985) Nonreactivemeasurement: a focus for research on ab-senteeism and occupational stress. In: L. L.Cummings & B. M. Straw (eds) Organiza-tional Behaviour. JAI Press Inc., Greenwich,Connecticut.

Folkard, S., & Monk, T. H. (1985) Hours ofWork - Temporal Factors in Work Schedul-ing. Wiley & Sons, Chichester.

Folkman, S. (1984). Personal control andstress and coping processes: a theoreticalanalysis. Journal of Personality and SocialPsychology, 46, 839-852.

Folkman, S., & Lazarus, R. (1986). Stressprocess and depressive symptomology.Journal of Abnormal Psychology, 95 , 107-113.

Folkman, S., Lazarus, R., Greun, R., & De-Longis, A. (1986b). Appraisal, coping,health status, and psychological symp-toms. Journal of Personality and Social Psy-chology, 50, 571-579.

Folkman, S., Lazarus, R.S., Dunkel-Schet-ter, C,. DeLongis, A., & Gruen, R. (1986a).Dynamics of a stressful encounter: cogni-tive appraisal, coping, and encounter out-

comes. Journal of Personality and SocialPsychology. 50, 992-1003.

Forsythe, R.P., & Harris, R.E. (1970) Circula-tory changes during stressful stimuli in rhe-sus monkeys. Circulation Research (supp.1) 26-27, I.B-I.20.

Fox, B. H. (1981) Psychosocial factors andthe immune system in human cancer. In: R.Ader (ed) Psychoneuroimmunology. Acad-emic Press, New York.

Frankenhauser, M. (1975) Sympathetic-adreno-medullary activity, behaviour andthe psychosocial environment. In: P. H.Venables & M. Christie (eds) Research inPsychophysiology. Wiley & Sons, Chich-ester.

Frankenhauser, M., & Gardell, B. (1975)Underload and overload in working life: amultidisciplinary approach. Reports fromthe Department of Psychology, no: 460,University of Stockholm, Stockholm.

Frankenhauser, M., & Gardell, B. (1976)Underload and overload in working life:outline of a multidisciplinary approach.Journal of Human Stress, 2, 15-23.

French, J. R. P., & Caplan, R. D. (1970) Psy-chosocial factors in coronary heart disease.Industrial Medicine, 39, 383-397.

French, J. R. P., & Caplan, R. D. (1972) Or-ganizational stress and individual strain. In:A. Marrow (ed) The Failure of Success.AMACOM, New York.

French, J. R. P., Caplan, R. D., & van Harri-son, R. (1982) The Mechanisms of JobStress and Strain. Wiley & Sons, New York.


143■

French, J. R. P., Rogers, W., & Cobb, S.(1974) A model of person-environment fir.In: G.W. Coehlo, D.A. Hamburg, & J.E.Adams, (eds) Coping and Adaptation. Ba-sic Books, New York.

Frese, M. & Zapf, D. (1988) Methodologi-cal issues in the study of work stress: Ob-jective vs. subjective measurement of workstress and the question of longitudinalstudies. In CL Cooper & R. Payne (Eds.)Causes, Coping and Consequences ofStress at Work. Chichester: John Wiley.

Friedman, M., & Rosenman, R.H. (1974)Type A: Your Behaviour and Your Heart.Knoft, New York.

Friedman, M., & Ulmer, D. (1984) TreatingType A Behaviour and Your Behaviour.Knoft, New York.

Gael, S. (1988) The Job Analysis Handbookfor Business, Industry and Government.Wiley & Son, New York.

Ganster, D.C., & Fusilier, M.R. (1989) Con-trol in the workplace. In: CL Cooper & I.Robertson (eds) International Review of In-dustrial and Organizational Psychology.Wiley & Sons, Chichester.

Ganster, D.C., Mayes, B.T., & Fuselier, M.R.(1986) Role of social support in the experi-ence of stress at work. Journal of AppliedPsychology, 71, 102-110.

Ganster, D.C., Mayes B.T., Sime W.E., &Tharp GD (1982) Managing occupationalstress: a field experiment. Journal of Ap-plied Psychology, 67, 533-542.

Gardell, B. (1971) Alienation and mentalhealth in the modern industrial environ-

ment. In: L. Levi (ed) Society, Stress andDisease Vol I. Oxford University Press, Ox-ford.

Gardell, B. (1973) Quality of Work andNon-work Activities and Rewards in Afflu-ent Societies. Reports from PsychologicalLaboratories no: 403, University of Stock-holm, Stockholm.

Gardell, B. (1982) Work participation andautonomy: A multilevel approach todemocracy at the workplace. InternationalJournal of Health Services, 12, 31-41.

Gardiner, B.M. (1980) Psychological as-pects of rheumatoid arthritis. Psychologi-cal Medicine, 10, 150-163.

Genest, M. (1983) Coping with rheuma-toid arthritis. Canadian Journal of Behav-ioural Science, 15, 392-408.

Genest, M. (1989) The relevance of stressto rheumatoid arthritis. In R.W.J. Neufeld(ed) Advances in the Investigation of Psy-chological Stress.Wiley & sons, New York.

Geurts S., Rutte C., & Peeters M. (1999)Antecedents and consequences of work-home interference among medical resi-dents. Social Science & Medicine, Vol.48,No.9, pp.1135-1148

Glass, D. C. (1977) Behaviour Patterns,Stress and Coronary Disease. Erlbaum, Hill-sadle, New Jersey.

Glass, D. C., & Singer, J. E. (1972) UrbanStress: Experiments on Noise and SocialStressors. Academic Press, New York.

Gobel M, Springer J, Scherff J (1998) Stressand strain of short haul bus drivers: Psy-


■144

chophysiology as a design orientedmethod for analysis. Ergonomics, Vol.41,No.5, pp.563-580

Goldberg, R.J. & Novack, D.H. (1992) Thepsychosocial review of systems. Social Sci-ence & Medicine, Vol.35, No.3, pp.261-269

Goldenhar, L. M., Swanson, N. G. HurrellJnr., J. J., Ruder, A. & Deddens, J. (1998).Stressors and adverse outcomes for femaleconstruction workers. Journal of Occupa-tional Health Psychology, 3, 19 - 32.

Golembiewski, R.T., Hilles, R, & Daly, R(1987) Some effects of multiple OD inter-ventions on burnout and work site fea-tures. Journal of Applied BehavioralScience, 23, 295-313

Green, L. W. (1974) Towards cost-benefitevaluations of health education. HealthEducation Monographs, 1 (Supplement),34-36.

Griffiths, A.J. (1999a) Work design andmanagement - The older worker. Experi-mental Aging Research, 25 (4), pp.411-420

Griffiths, A.J. (1999b) Organizational inter-ventions: facing the limits of the naturalscience paradigm. Scandinavian Journal ofWork and Environmental Health, 25(6),pp. 589-596.

Griffiths, A.J., Cox, T. & Auty A. (1998)Work stress: a brief guide for line man-agers. Loss Prevention Council, Boreham-wood, UK.

Griffiths, A.J., Cox, T. & Barlow, C.A.(1996) Employers’ responsibilities for the

assessment and control of work-relatedstress: a European perspective. Health andHygiene, 17, 62-70.

Guppy, A., Weatherstone, L. (1997) Cop-ing strategies, dysfunctional attitudes andpsychological well-being in white collarpublic sector employees. Work And Stress,11,58-67

Gutek, B. A., Repetti, R. L., & Silver, D. L.(1988) Nonwork roles and stress at work.In: CL Cooper & R. Payne (eds) Causes,coping and Consequences of Stress atWork. Wiley & Sons, Chichester.

Hacker, W. (1991) Objective work environ-ment: analysis and evaluation of objectivework characteristics. Paper presented to: AHealthier Work Environment: Basic Con-cepts & Methods of Measurement. Hog-berga, Lidingo, Stockholm.

Hacker, W., Iwanova, A., & Richter, P.(1983) Tatigkeits-bewertungssystem (TBS-L). Hogrefe, Gottingen.

Hall, E.M. (1991) Gender, work control andstress: a theoretical discussion and an em-pirical test. In J.V. Johnson & G. Johansson(eds) The Psychosocial Work Environment:Work Organization, Democratization andHealth. Baywood Publishing, New York.

Handy, C. (1975) Difficulties of combiningfamily and career. The Times, Sept 22, 16.

Harrington, J. M. (1978) Shift Work andHealth: A critical review of the Literature.HMSO, London.

Harris, L. & Associates (1985) Poll conduct-ed for the Metropolitan Life Foundation


145■

Hart, K.E. (1987) Managing stress in occu-pational settings: a selective review of cur-rent research and theory. In CL Cooper (ed)Stress Management Interventions at Work.MCB University Press Ltd.

Haslam, D. R. (1982) Sleep loss, recoverysleep and military performance. Ergonom-ics, 25, 163-178.

Haynes, S.G., Feinleib, M. , Levine, S.,Scotch, N., & Kannel, W.B. (1978b) The re-lationship of psychosocial factors to coro-nary heart disease in the Framinghamstudy, II. Prevalence of coronary heart dis-ease. American Journal of Epidemiology,107, 384-402.

Haynes, S.G., Levine, S., Scotch, N., Fein-leib, M., & Kannel, W.B. (1978a) The rela-tionship of psychosocial factors tocoronary heart disease in the Framinghamstudy, I. American Journal of Epidemiology,107, 362-383.

Health & Safety Executive (1990a) WorkRelated Upper Limb Disorders : A Guide toPrevention. HSE Books, Sudbury.

Health & Safety Executive (1990b) MentalHealth at Work.HSE Books, Sudbury.

Heinisch, D.A. & Jex, S.M. (1998) Mea-surement of negative affectivity: a com-parison of self-reports and observerratings. Work & Stress, 12 (2), 145-160.

Henry, J. P., & Stephens, P. M. (1977)Stress, Health and the Social Environment.A Sociobiologic Approach to Medicine.Springer, New York.

Hiebert, B. & Farber, I. (1984) Teacherstress: A literature survey with a few sur-

prises. Canadian Journal of Education, 9(1), 14-27.

Hillas, S., & Cox, T. (1987) Post TraumaticStress Disorder in the Police. OccasionalPaper. Police Scientific Research and Devel-opment Branch, Home Office, London.

Hingley, P., & Cooper, C.L. (1986) Stressand the Nurse Manager. Wiley & Son,Chichester.

Holmes, T. H., & Rahe, R. H. (1967) The So-cial Readjustment Rating Scale. Journal ofPsychosomatic Research, 11, 213-218.

Holroyd, K.A., & Lazarus, R.S. (1982)Stress, coping and somatic adaptation. In:L. Goldberger & S. Breznitz (eds) Hand-book of Stress: Theoretical and Clinical As-pects. Free Press, New York.

Holt, R. R. (1982) Occupational stress. In: L.Goldberger & S. Breznitz (eds) Handbookof Stress: Theortical and Clinical Aspects.Free Press, New York.

House, J. (1974) Occupational stress andcoronary heart disease: a review and theo-retical integration. Journal of Health andSocial Behaviour, 15, 12-27.

House, J. S., & Wells, J. A. (1978) Occupa-tional stress, social support and health. In:A. McLean, G. Black, & M. Colligan (eds)Reducing Occupational Stress: Proceed-ings of a Conference. DWEH (NIOSH) Pub-lication no: 78-140, 8-29.

House, J.S., McMichael, A.J., Wells, J.A.,Kaplan, B.H., & Landerman, L.R. (1979)Occupational stress and health among fac-tory workers. Journal of Health and SocialBehaviour,20, 139-160.


■146

Houts, P. S., & McDougall, V.C. (1988) Ef-fects of informing workers of their healthrisks from exposure to toxic materials.American Journal of Industrial Medicine,13, 271-279.

Hurrell, J. J., & McLaney, M. A. (1989) Con-trol, job demands and job satisfaction. In:S. L. Sauter, J. J. Hurrell, & CL Cooper (eds)Job Control and Worker Health. Wiley &Sons, Chichester.

Idzikowski, C., & Baddeley, A. D. (1983)Fear and dangerous environments. In: G.R. J. Hockey (ed) Stress and Fatigue in Hu-man Performance. Wiley & Sons, Chich-ester.

Ilmarinen J., & Rantanen J. (1999) Promo-tion of work ability during ageing. Ameri-can Journal of Industrial Medicine, No.S1,pp.21-23

Ingersoll G.L., Cook J.A., Fogel S., Apple-gate M, Frank B (1999) The effect of pa-tient-focused redesign on midlevel nursemanagers’ role responsibilities and workenvironment. Journal Of Nursing Adminis-tration, Vol.29, No.5, pp.21-27

International Federation of Commercial,Clerical and Technical Employees [FIET](1992) Resolutions adopted by the 22ndFIET World Congress (San Fransisco, Au-gust 1991). Geneva.

International Labour Organization [ILO](1986) Psychosocial Factors at Work:Recognition and Control. OccupationalSafety and Health Series no: 56, Interna-tional Labour Office, Geneva.

International Labour Organization [ILO](1992) Preventing Stress at Work. Condi-tions of Work Digest, 11, InternationalLabour Office, Geneva.

Intindola, B. (1991) EAP’s still foreign tomany small businesses. National Under-writer, 95, 21.

Irwin, J., & Anisman, H. (1984) Stress andpathology: immunological and central ner-vous system interactions. In: CL Cooper(ed) Psychosocial Stress and Cancer. Wiley& Sons, Chichester.

Israel, B.A., Baker, E.A., Goldenhar, L.M.,Heaney, C.A. & Schurman, S.J. (1996) Oc-cupational stress, safety and health: Con-ceptual framework and principles foreffective prevention interventions. Journalof Occupational Health Psychology, I (3),261-286.

Ivancevich, J.M., & Matteson, M.T. (1980)Stress and Work. Scott Foresman, Glen-view, Illinois.

Ivancevich, J.M., & Matteson, M.T. (1986)Organizational level stress managementinterventions: review and recommenda-tions. Journal of Organizational Behaviourand Management, 8, 229-248.

Ivancevich, J.M., Matteson, M.T., Freed-man, S.M., & Phillips, J.S. (1990) Worksitestress management interventions. Ameri-can Psychologist, 45, 252-261.

Jackson, P. R., & Warr, P. B. (1984) Unem-ployment and psychological ill health: themoderating role of duration and age. Psy-chological Medicine, 14, 610-614.


147■

Jackson, S. (1983) Participation in deci-sion-making as a strategy for reducing job-related strain. Journal of AppliedPsychology, 68, 3-19.

Jackson, S., & Schuler, R. S. (1985) A meta-analysis and conceptual critique of re-search on role ambiguity and role conflictin work settings. Organizational Behaviourand Human Decision Processes, 36, 16-78.

Jenkins, C.D., Rosenman, R. H., & Fried-man, M. (1968) Replicability of rating thecoronary prone behaviour pattern. Journalof Chronic Diseases, 20, 371-379.

Jenkins, C.D., Zyzanski, S.J., & Rosenman,R.H. (1976) Risk of new myocardial infarc-tion in middle-aged men with manifestcoronary heart disease. Circulation, 53,342-347.

Jenkins, R. (1992) Prevalence of Mental Ill-ness in the Workplace. In R. Jenkins & N.Coney (eds) Prevention of Mental Ill Healthat Work. HMSO, London.

Jex, S.M. (1991) The psychological benefitsof exercise in work settings: a review, cri-tique, and dispositional model. Work &Stress, 5, 133-147.

Jex, S.M. & Spector, P.E. (1996) The impactof negative affectivity on stressor-strain re-lations: a replication and extension. Work& Stress, 10 (1), 36-45.

Jick, T.D. (1979) Mixing qualitative andquantitative methods: Triangulation in ac-tion. Administrative Science Quarterly, 24,602-611.

Johansson, G. & Aronsson, G. (1984)Stress reactions in computerized adminis-

trative work. Journal of Occupational Be-haviour, 5, 159-181.

Johnson, J. V. (1989) Control, collectivityand the psychosocial work environment.In: S. L. Sauter, J. J. Hurrell Jr & C. L. Coop-er (eds) Job control and worker health.John Wiley & Sons, Chichester.

Johnson, J.V. (1996) Conceptual andmethodological developments in occupa-tional stress research. An introduction tostate-of-the-art reviews I. Journal of Occu-pational Health Psychology, I (1), 6-8.

Johnson, J.V. & Hall, E.M. (1996) Dialecticbetween conceptual and causal enquiry inpsychosocial work-environment research.Journal of Occupational Health Psycholo-gy, I (4), 362-374.

Johnson, J. V., Hall, E. M., Stewart, W.,Fredlund, P. & Theorell, T. (1991) Com-bined exposure do adverse work organiza-tion factors and cardiovascular disease:towards a life-course perspective. In: L. D.Fechter (ed.) Proceedings of the 4th Inter-national Conference on the Combined Ef-fects of Environmental Factors. JohnsHopkins University Press, Baltimore.

Johnson, L.C. (1981) Biological Rhythms,Sleep and Shift Work. Advances in SleepResearch, Vol 7. Spectrum, New York.

Jones, D.M. (1983) Noise. In: G. R. J. Hock-ey (ed) Stress and Fatigue in Human Per-formance. Wiley & Sons, Chichester.

Jones, D.M. (1999) The cognitive psychol-ogy of auditory distraction: The 1997 BPSBroadbent Lecture British Journal Of Psy-chology, 90 (2), 167-187.


■148

Jones, J.R., Hodgson, J.T., Clegg, T.A. & El-liot R.C. (1998) Self-reported work-relatedillness in 1995: results from a householdsurvey. Sudbury: HSE Books.

Jones, J.W., Barge, B.N., Steffy, B.D., Fay,L.M., Kunz, LK, & Wuebker, LJ (1988)Stress and medical malpractice: organiza-tional risk assessment and intervention.Journal of Applied Psychology, 73, 727-735.

Junghanns, G., Ertel, M., & Ullsperger, P.(1998) Anforderungsbewältigung undGesundheit bei computergestützterBüroarbeit. Wirtschaftsverlag NW, Bremer-haven. (Schriftenreihe der Bundesanstaltfür Arbeitsschutz und Arbeitsmedizin:Forschung, Fb 787)

Junghanns, G.; Ullsperger, P. & Ertel, M.(1999) Zum Auftreten von Gesundheits-beschwerden bei computergestützterBüroarbeit - eine multivariate Analyse aufder Grundlage einer fragebo-gengestützten Erhebung. Zeitschrift fürArbeitswissenschaft 53 (25. NF), 18-24.

Junghanns, G., Ullsperger, P., Ertel, M., &Pech, E. (1999, in press) Gesundheit undWohlbefinden bei moderner Büroarbeit -eine Studie zum “Anforderungs-Kontroll”-Modell. Ergo-Med.

Kahn, R. L. (1973) Conflict, ambiguity andoverload: three elements in job stress. Oc-cupational Mental Health, 31, 2-9.

Kahn, R. L. (1974) Conflict, ambiguity andover work: three elements in job stress. In:A. McLean (ed) Occupational Stress.Charles C. Thomas, Springfield, Illinois.

Kahn, R. L., & Byosiere, S. (1990) Stress inOrganizations. In: M. Dunnette (ed) Hand-book of Industrial and Organizational Psy-chology. Rand McNally, Chicago.

Kahn, R. L., Wolfe, D. M., Quinn, R. P.,Snoek, J. D., & Rosenthal, R. A. (1964) Or-ganizational Stress: Studies in Role Conflictand Ambiguity. Wiley & Sons, New York.

Kang B., Lee B., Kang K.W., Suh J.C., &Yoon ES TI (1999) AHA: a knowledgebased system for automatic hazard identi-fication in chemical plant by multimodelapproach. Expert Systems With Applica-tions, Vol.16, No.2, pp.183-195

Kanter, R.M. (1977) Work and family in theUnited States: a critical review and agendafor research and policy. Russell SageFoun-dation, New York.

Karasek, R. A. (1979) Job demands, jobdecision latitude and mental strain: impli-cations for job redesign. AdministrativeScience Quarterly, 24, 285-308.

Karasek, R. A. (1981) Job socialisation andjob strain: the implications of two psy-chosocial mechanisms for job design. In: B.Gardell & G. Johansson (eds) Working Life:A Social Science Contribution to Work Re-form. Wiley & Sons, Chichester.

Karasek, R., & Theorell, T. (1990) HealthyWork: Stress, Productivity and the Recon-struction of Working Life. Basic Books,New York.

Karasek, R.A., Schwartz, J., & Theorell, T.(1982) Job characteristics, occupation, andcoronary heart disease. (Final report onGrant No. R-01-OH00906). Cincinnati,


149■

OH: National Institute for OccupationalSafety and Health.

Karasek, R.A., Baker, D., Marxer, F.,Ahlbom, A., & Theorell, T. (1981) Job deci-sion latitude, job demands and cardiovas-cular disease. American Journal of PublicHealth, 71, 694-705.

Kasl, S. V. (1980a) Epidemiological contri-butions to the study of work stress. In: C.L.Cooper and R. Payne (eds) Stress at Work.Wiley & Sons, Chichester.

Kasl, S. V. (1980b) The impact of retire-ment. In: C.L. Cooper & R.L. Payne (eds)Current Concerns in Occupational Stress.Wiley & Sons, Chichester.

Kasl, S. V. (1984) Stress and health. Annu-al Review of Public Health, 5, 319-341.

Kasl, S. V. (1987) Methodologies in stressand health: past difficulties, present dilem-mas and future directions. In: S. Kasl & C.Cooper (eds) Stress and Health: Issues inResearch Methodology. Wiley & Sons,Chichester.

Kasl, S.V. (1989) An epidemiological per-spective on the role of control in health. InS.L. Sauter, J.J. Hurrell, & CL Cooper (eds)Job Control and Worker Health. Wiley &Sons, New York.

Kasl, S.V. (1990) Assessing health risks inthe work setting. In S. Hobfoll (Ed.) NewDirections in Health Psychology Assess-ment. Washington D.C.: H hemispherePublishing Corporation.

Kasl, S. V. (1992) Surveillance of psycho-logical disorders in the workplace. In: G. P.Keita & S. L. Sauter (eds) Work and Well-

Being: An Agenda for the 1990s. Ameri-can Psychological Association, Washing-ton DC.

Kasl, S. V., & Cobb, S. (1967) Effects ofparental status incongruence and discrep-ancy in physical and mental health of adultoffspring. Journal of Personality and SocialPsychology, monograph: 7, 1-5.

Kasl, S. V., & Cobb, S. (1980) The experi-ence of losing a job: some effects on car-diovascular functioning. Psychotherapyand Psychosomatics, 34, 88-109.

Kasl, S.V. and Cobb, S. (1982) Variability ofstress effects among men experiencing jobloss. In L. Goldberger & S. Breznitz (eds)Handbook of Stress: Theoretical and Clini-cal Aspects. Free Press, New York.

Kawakami N, & Haratani T. (1999) Epi-demiology of job stress and health inJapan: Review of current evidence and fu-ture direction. Industrial Health, Vol.37,No.2, pp.174-186

Kearns, J. (1986) Stress at work: the chal-lenge of change. BUPA series The Man-agement of Health: 1 Stress and the City,BUPA.

Kegeles, S.M., Coates, T., Christopher, A.,& Lazarus, J. (1989) Perceptions of Aids:the continuing saga of Aids-related stig-ma. Aids, 3 (supp 1), S253-S258.

Keita, G. P., & Sauter, S. L. (1992) Workand Well-Being: An Agenda for the 1990s.American Psychological Society, Washing-ton DC.

Kittel, F., Kornitzer, M., DeBacker, B., Dra-maix, M., Sobolski, J., Degre, J., Denolin,


■150

H. (1983) Type A in relation to job stress,social and bioclinical variables: the Belgianphysical fitness study. Journal of HumanStress, 9, 37-45.

Kobasa, S. (1979). Stressful life events,personality and health: an inquiry into har-diness. Journal of Personality and SocialPsychology, 37, 1-13.

Kobasa, S., & Puccetti, M. (1983). Person-ality and social resources in stress resis-tance. Journal of Personality and SocialPsychology, 45, 839-850.

Kobasa, S., Maddi, S., & Courington, S.(1981). Personality and constitution as me-diators in the stress-illness relationship.Journal of Health and Social Behaviour, 22,368-378.

Kobasa, S., Maddi, S., & Kahn, S., (1982).Hardiness and health: a prospective study.Journal of Personality and Social Psycholo-gy, 42, 168-177.

Kobayashi F., Furui H., Akamatsu Y.,Watanabe T., & Horibe H. (1999) Changesin psychophysiological functions duringnight shift in nurses - Influence of chang-ing from a full-day to a half-day work shiftbefore night duty. International Archives ofOccupational and Environmental Health,Vol.69, No.2, pp.83-90

Kompier, M.A.J., Geurts, S.A.E., Grunde-man, R.W.M., Vink, P. & Smulders, P.G.W.(1998) Cases in stress prevention: the suc-cess of a participative and stepwise ap-proach. Stress Medicine, 14, 155-168

Kornhauser, A. (1965) Mental Health ofthe Industrial Worker. University of Chica-go Press, Chicago, Illinois.

Kristensen, T.S. (1996) Job stress and car-diovascular disease: A theoretic critical re-view. Journal of Occupational HealthPsychology, I (3), 246-260.

Kroes, W.H. (1976) Society’s victim, the po-liceman: an analysis of job stress in polic-ing. Thomas, Springfield.

Kryter, K. D. (1972) Non auditory effects ofenvironmental noise. American Journal ofPublic Health, 62, 389-398.

Kuorinka, I. (1979) Occupational strainfrom working movments. Paper to: Inter-national Ergonomics Association, Warsaw(August). Abstract in: Ergonomics, 22,732.

Lacey, J. I. (1967) Somatic response pat-terning and stress: some revisions of acti-vation theory. In: M. H. Appley & R.Trumbull (eds) Psychological Stress. Apple-ton-Century-Crofts, New York.

Landsbergis, P.A. (1988) Occupationalstress among health care workers: a test ofthe job demands-control model. Journal ofOrganizational Behaviour, 9, 217-239.

Landsbergis, P.A. & Vivona-Vaughan, E.(1997) Evaluation of an occupational stressintervention in a public agency. Journal ofOrganizational Behavior, 16, 29-48.

Landsbergis, P.A., Schnall, P.L., Schwartz,J.E., Warren, K. & Pickering, T.G. (1995)Job strain, hypertentsion, and cardiovascu-lar disease: empirical evidence, method-ological issues, and recommendations for


151■

further research. In S.L. Sauter & L.R. Mur-phy (Eds.) Organizational risk factors forjob stress. Washington, DC: APA.

Landstrom, U., Holmberg, K., Kjellberg, A.,Soderberg, L., Tesarz, M. (1995) Exposuretime and its influence on noise annoyanceat work. Journal Of Low Frequency Noise& Vibration, Vol.14, No.4, pp.173-180

Landy, F. J. (1989) The Psychology of WorkBehaviour. Brooks/Cole, Monterey, Califor-nia.

Landy, F. J. (1992) Work design and stress.In: G. P. Keita & S. L. Sauter (eds) Work andWell-Being: An Agenda for the 1990s.American Psychological Association,Washington DC.

Landy, F.J., Quick, J.C. & Kasl, S. (1994)Work, stress and well-being. InternationalJournal of stress management. I (1), 33-73

Larwood, L., & Wood, M. M. (1979)Women in Management. LexingtonBooks, London.

Last, J. M. (1988) Dictionary of Epidemiol-ogy. Oxford University Press, New York.

Laville, A., & Teiger, C. (1976) Sante men-tale et conditions de travail. Therapeutis-che Umschau, 32, 152-156.

Lazarus, R. S. (1966) Psychological Stressand the Coping Process. McGraw-Hill,New York.

Lazarus, R. S. (1976) Patterns of Adjust-ment. McGraw-Hill, New York.

Lazarus, R.S. & Folkman, S. (1984). Stress,Appraisal and Coping. Springer Publica-tions, New York .

Leather, P., Lawrence, C., Beale, D., Cox, T.& Dickson, R. (1998) Exposure to occupa-tional violence and the buffering effects ofintra-organizational support. Work &Stress, 12(2), 161-178.

Leather, P., Brady, C., Lawrence, C., Beale,D. & Cox, T. (Eds.) (1999) Work-related vi-olence: Assessment and intervention. Lon-don: Routledge. ISBN 0415194156.

Leiter, M. (1991) The dream denied: pro-fessional burnout and the constraints ofhuman service organizations. CanadianPsychology, 32, 547-558.

Lennon, M.C. (1999) Work and unemploy-ment as stressors. In Horwitz, Allan V. (Ed);Scheid, Teresa L. (Ed) A handbook for thestudy of mental health: Social contexts,theories, and systems. (pp. 284-294). NewYork, NY, USA: Cambridge UniversityPress.

Leventhal, H., and Tomarken, A. (1987)Stress and illness: perspectives from healthpsychology. In: S. Kasl and C. Cooper (eds)Stress and Health: Issues in ResearchMethodology. Wiley & Sons, Chichester.

Levi, L. (1972) Stress and distress in re-sponse to psychosocial stimuli. Acta Med-ica Scandinavica, 191, supplement: 528.

Levi, L. (1981) Preventing Work Stress. Ad-dision-Wesley, Reading, Mass.

Levi, L. (1984) Stress in Industry: Causes,Effects and Prevention. Occupational Safe-


■152

ty and Health Series no. 51, InternationalLabour Office, Geneva.

Levi, L. (1992) Psychosocial, occupational,environmental and health concepts, re-search results and applications. In: G.P. Kei-ta and S.L. Sauter (eds) Work and WellBeing: An Agenda for the 1990s. Ameri-can Psychological Association, Washing-ton DC.

Levi, L., Frankenhauser, M., and Gardell, B.(1986) The characteristics of the work-place and the nature of its social demands.In: S. Wolf and A. J. Finestone (eds) Occu-pational Stress, Health and Performance atWork. PSG Pub. Co. Inc., Littleton, MA.

Lindquist T.L. & Cooper C.L. (1999) Usinglifestyle and coping to reduce job stressand improve health in ‘at risk’ office work-ers. Stress Medicine, Vol.15, No.3, pp.143-152

Lippe, K. (1990) Compensation for men-tal-mental claims under Canadian law. Be-havioural Sciences and the Law, 8,398-399.

Lipton, M.A. (1976) Behavioral effects ofhypothalamic polypeptide hormones inanimals and man. In E.J. Sachar (ed) Hor-mones, Behaviour and Psychopathology.Raven, New York.

Lisander, B. (1979) Somato-autonomic re-actions and their higher control. In C.Brooks, K. Koizumi, and A. Sato (eds) Inte-grative Functions of the Autonomic Ner-vous System. Elsevier, New York.

Lobban R.K., Husted J, & Farewell V.T.(1998) A comparison of the effect of job

demand, decision latitude, role and super-visory style on self-reported job satisfac-tion. Work and Stress, Vol.12, No.4,pp.337-350

Locke, A. A. (1976) The nature and causesof job satisfaction. In: M. D. Dunnette (ed)Handbook of Industrial and Organization-al Psychology. Rand McNally, Chicago.

Logue, J.N. (1980) Mental health aspectsof disaster. Paper presented to the fifthAnnual National Hazards Research Work-shop, Boulder.

Logue, J.N., Melick, M.E., and Struening,E. (1981) A study of health and mentalhealth status following a major natural dis-aster. In R. Simmons (ed) Research in Com-munity and Mental Health: An AnnualCompilation of Research, Vol. 2.Greewich:, JAI.

Loher, B. T., Noe, R. A., Moeller, N. L., andFitzgerald, M. P. (1985) A meta-analysis ofthe relation of job characteristics to jobsatisfaction. Journal of Applied Psycholo-gy, 70, 280-289.

Lu L., Tseng H.J. & Cooper C.L. (1999)Managerial stress, job satisfaction andhealth in Taiwan Stress Medicine, 1999,Vol.15, No.1, pp.53-64

Lundahl, A. (1971) Fritid Och Rekreation.All Manna Forlager. Laginkomstutrednin-gen, Stockholm.

Lundberg, U., and Forsman, L. (1979)Adrenal medullary and adrenal cortical re-sponses to understimulation and overstimulation: comparison between type A


153■

and type B persons. Biological Psychology,9, 79-89.

MacDougall, J. M., Dembrowski, T. M.,Dimsdale, J. E., and Hackett, T. P. (1985)Components of Type A, hostility andanger-in: further relationships to angio-graphic findings. Health Psychology, 4,137-152.

Mackay, C. (1987) Violence to Staff in theHealth Services. HMSO, London.

Mackay, C., Cox, T., Burrows, G., andLazzerini, T. (1978) An inventory for themeasurement of self-reported stress andarousal. British Journal of Social and Clini-cal Psychology, 17, 283-284.

MacLennan, B. W. (1992) Stressor reduc-tion: an organizational alternative to indi-vidual stress management. In J.C. Quick,L.R. Murphy, JJ. Hurrell (eds) Stress andWell-being at Work: Assessments and In-terventions for Occupational MentalHealth. American Psychological Associa-tion, Washington DC.

Mandler, G. (1982) Stress and thoughtprocesses. In: L. Goldberg and S. Breznitz(eds) Handbook of Stress: Theoretical andClinical Aspects. Free Press, New York.

Margolis, B. L., & Kroes, W. H. (1974) Workand the health of man. In: J. O’Toole (ed)Work and the Quality of Life. MIT Press,Cambridge, Mass.

Margolis, B. L., Kroes, W. H., & Quinn, R. P.(1974) Job stress; an unlisted occupationalhazard. Journal of Occupational Medicine,16, 652-661.

Marmot, M. G., & Madge, N. (1987) Anepidemiological perspective on stress andhealth. In: S.V. Kasl & CL Cooper (eds)Stress and Health: Issues in ResearchMethodology, Wiley & Sons, Chichester.

Marmot, M. and Theorell, T. (1988) Socialclass and cardiovascular disease. The con-tribution of work. International Journal ofHealth Services, 18, 659-674.

Marshall, J. (1977) Job pressures and satis-factions at managerial levels. UnpublishedPhD thesis, University of Manchester Insti-tute of Science and Technology, Manches-ter.

Martin, R., & Wall, T. (1989) Attentionaldemand and cost responsibility as stressorsin shop-floor jobs. Academy of Manage-ment Journal, 32, 69-86.

Martinez JMAG, & Martos MPB (1999) Themeaning of work in persons with type-Abehavior pattern. Psicothema, Vol.11,No.2, pp.357-366

Mason, J. W. (1968) A review of psy-choendocrine research on the pituitary-adrenal cortical system. PsychosomaticMedicine, 30, 576-607.

Mason, J. W. (1971) A re-evaluation of theconcept of non-specificity in stress theory.Journal of Psychiatric research, 8, 323.

Matteson, M.T., & Ivancevich, J.M. (1987)Controlling Work Stress. Jossey-Bass, SanFransisco.

Matthews, K. A., Glass, D. C., Rosenman,R. H., & Bortner, R. W. (1977) Competitivedrive, pattern A and coronary heart dis-ease: a further analysis of some data from


■154

the Western Collaborative Group Study.Journal of Chronic Diseases, 30, 489-498.

Meichenbaum, D. (1977) Cognitive-Be-haviour Modification. Plenum Press, NewYork.

Meichenbaum, D. (1983) Coping withStress. Century Publishing, London.

Meijman, T.F., Van Dormolen, M., Herber,R.F.M., Rongen, H. & Kuiper, S. (1995) Jobstress, neuroendocrine activation, and im-mune status. In SL Sauter & LR Murphy(Eds.) Organizational risk factors for jobstress. Washington, DC: APA.

Melamed S., Yekutieli D., Froom P., Kristal-Boneh E., Ribak J. (1999) Adverse workand environmental conditions predict oc-cupational injuries - The Israeli Cardiovas-cular Occupational Risk FactorsDetermination in Israel (CORDIS) study.American Journal Of Epidemiology,Vol.150, No.1, pp.18-26

Milgram, N.A. (1982) War-related stress inIsraeli children and youth. In L. Goldberger& S.Breznitz (eds) Handbook of Stress:Theoretical and Clinical Aspects. FreePress, New York.

Millar, D.J. (1984) The NIOSH suggestedlist of the ten leading work-related dis-eases and injuries. Journal of OccupationalMedicine, 26, 340-341.

Millar, D.J. (1990) Mental health and work-place: interchangeable partnership. Amer-ican Psychologist, 45, 1165-1166.

Miller, D.G., Grossman, Z.D., Richardson,R.L., Wistow, B.W., & Thomas, F.D. (1978)Effect of signalled versus unsignalled stress

on rat myocardium. Psychosomatic Medi-cine, 40, 432-434.

Miller, J. D. (1974) Effects of noise on peo-ple. Journal of the Acoustical Society ofAmerica, 56, 72-764.

Miller, S. (1979). Controllability and hu-man stress: method, evidence and theory.Behavioural Research and Therapy, 17, 28-304.

Miller, S., Brody, D., & Summerton, J.(1988). Styles of coping with threat: impli-cations for health. Journal of Personalityand Social Psychology, 54, 142-148.

Ministry of Labour (1987) Survey on Stateof Employees’ Health. Ministry of Labour,Tokyo, Japan.

Monjan, A.A. (1981) Stress and immuno-logic competence: studies in animals. In R.Ader (ed) Psychoneuroimmunology. Acad-emic Press, New York.

Monk, T. H., & Tepas, D. (1985) Shift Work.In: CL Cooper & M. J. Smith (eds) JobStress and Blue Collar Work. Wiley & Sons,Chichester.

Mudrack P.E. (1999) Time structure andpurpose, Type A behavior, and the Protes-tant work ethic. Journal Of OrganizationalBehavior, Vol.20, No.2, pp.145- 158

Murphy, L.R. (1984) Occupational stressmanagement: a review and appraisal.Journal of Occupational Psychology, 57, 1-15.

Murphy, L.R. (1988) Workplace interven-tions for stress reduction and prevention.In CL Cooper & R. Payne (Eds) Causes,


155■

Coping and Consequences of Stress atWork.

Murphy, L.R., & Hurrell, J.J. (1987) Stressmanagement in the process of occupa-tional stress reduction. Journal of Manage-rial Psychology, 2, 18-23.

Murphy, L.R., Hurrell, J.J. & Quick, J.C.(1992) Work and well-being: where do wego from here? In J.C. Quick, L.R. Murphy,& J.J. Hurrell (eds) Stress and Well-being atWork: Assessments and Interventions forOccupational Mental Health. AmericanPsychological Association, WashingtonDC.

Naitoh, P., Englund, C. E., & Ryman, D. H.(1983) Extending Human effectivenessDuring Sustained Operations ThroughSleep Management. US Naval Health Re-search Center, San Diego, California.

Narayanan, V. K., & Nath, R. (1982) A fieldtest of some attitudinal and behaviouralconsequences of flexitime. Journal of Ap-plied Psychology, 67, 214-218.

National Institute of Occupational Safetyand Health [NIOSH] (1988) PsychosocialOccupational Health. National Institute ofOccupational Safety and Health, Washing-ton, DC.

Neale, M.S., Singer, J., Schwartz, G.E., &Schwartz, J. (1983) Conflicting perspec-tives on stress reduction in occupationalsettings: a systems approach to their reso-lution. Report to NIOSH on P.O. No. 82-1058, Cincinnati, Ohio.

Neary, J., Elliott, K.V., & Toohey, J. (1992)The causes of workplace stress and strate-

gies for management. Paper presented atthe International Symposium on Work-re-lated Diseases: Prevention and Health Pro-motion, Linz, Austria.

Nerell, G. (1975) Medical complaints andfindings in Swedish sawmill workers. In: B.Thunell & B. Ager (eds) Ergonomics inSawmill and Woodworking Industries. Na-tional Board of Occupational Safety andHealth, Stockholm.

Neufeld, R. W. J., & Paterson, R. J. (1989)Issues concerning control and its imple-mentation. In: R. W. J. Neufeld (ed) Ad-vances in the Investigation of PsychologicalStress. Wiley & Sons, New York.

Newman, J. E., & Beehr, T. A. (1979) Per-sonal and organizational strategies forhandling job stress: a review of researchand opinion. Personnel Psychology, 32, 1-43.

Nobile, R.J. (1991) Matters of confidential-ity. Personnel, 68, 11-12.

Nordhus, I.H., & Fleime, A.M. (1991) Jobstress in two different care-giving contexts:a study of professional and semi-profes-sional health personnel in Norway. Work &Stress, 5, 229-240.

Nowack, K.M. (1991) Psychological predic-tors of health status. Work & Stress, 5,117-131.

Nutt, P. C. (1981) Evaluation Concepts andMethods: Shaping Policy for the HealthAdministrator. SP Medical and ScientificBooks, New York.

O’Brien, G. E. (1982) The relative contribu-tion of perceived skill-utilization and other


■156

perceived job attributes to the predictionof job satisfaction: cross validation study.Human Relations, 35, 219-237.

O’Hanlon, J. F. (1981) Boredom: practicalconsequences and a theory. Acta Psy-chologia, 49, 53-82.

O’Leary, A. (1990) Stress, emotion, andhuman immune function. PsychologicalBulletin, 108, 363-382.

Organization for Economic Cooperationand Development [OECD] (1997) Joint Pro-ject on the Harmonization of Chemical Haz-ard/Risk Assessment Terminology. InternetWWW page at: http://www.who.ch/pro-grammes/pcs/rsk term/cvr ltr.htm

Orpen, C. (1981) Effect of flexible workinghours on employee satisfaction and per-formance. Journal of Applied Psychology,66, 113-115.

Paffenbarger, R.S., Hale, W.E., Brand, R.J.,& Hyde, R.T. (1977) Work-energy level,personal characteristics and fatal heart at-tack: a birth cohort effect. American Jour-nal of Epidemiology, 105, 200-213.

Paffenbarger, R.S., Hyde, R.T., Wing, A.L.,& Steinmetz, C.H. (1984) A natural historyof athleticism and cardiovascular health.Journal of the American Medical Associa-tion, 252, 491-495.

Pahl, J. M., & Pahl, R. E. (1971) Managersand Their Wives. Allen Lane, London.

Parker, S.K., Chmiel, N. & Wall, T.D. (1997)Work characteristics and employee well-being within a context of strategic down-sizing. Journal of Occupational healthPsychology, 2(4) 289-303.

Patton, J. F., Vogel J. A., Damokosh, A. I.,& Mello, R. P. (1989) Effects of continuousmilitary operations on physical fitness ca-pacity and physical performance. Work &Stress, 3, 69-77.

Payne, R. (1988) Individual differences inthe study of occupational stress. In: CLCooper & R. Payne (eds) Causes, Copingand Consequences of Stress at Work. Wi-ley & Sons, Chichester.

Payne, R., & Fletcher, B. (1983) Job de-mands, supports and constraints as predic-tors of psychological strain among schoolteachers. Journal of Vocational Behaviour,22, 136-147.

Payne, R., & Hartley, J. (1987) A test of amodel for exlaining the affective experi-ence of unemployed men. Journal of Oc-cupational Psychology, 60, 31-47.

Pearlin, L., & Schooler, C. (1978). Thestructure of coping. Journal of Health andSocial Behavior, 19, 2-21.

Pearlin, L., Lieberman, M. L., Menaghan,E., & Mullan, J. T. (1981) The stressprocess. Journal of Health and Social Be-haviour, 19, 2 - 21.

Pearse, R. (1977) What Managers ThinkAbout Their Managerial Careers. AMA-COM, New York.

Perez A.D., Meizoso M.T.G., Gonzalez R.D.(1999) Validity of the structured interviewfor the assessment of Type A behavior pat-tern. European Journal Of PsychologicalAssessment, Vol.15, No.1, pp.39-48

Perkins, D. V. (1988) The assessment ofstress using life events scales. In: L. Gold-


157■

berg & S. Breznitz (eds) Handbook ofStress: Theoretical and Clinical Aspects.Free Press, New York.

Perrewe, P., & Ganster, D. C. (1989) Theimpact of job demands and behaviuralcontrol on experienced job stress. Journalof Organizational Behaviour, 10, 136-147.

Peters M.L., Godaert G.L.R., Ballieux R.E.,Brosschot J.F., Sweep F.C.G.J., SwinkelsL.M.J.W., vanVliet M., & Heijnen C.J.(1999) Immune responses to experimentalstress: Effects of mental effort and uncon-trollabil ity Psychosomatic Medicine,Vol.61, No.4, pp.513-524

Peterson, P. (1981) Comparison of relax-ation training, cognitive restructuring/be-havioural training and multimodal stressmanagement training seminars in an occu-pational setting. Dissertation submitted toFuller Theological Seminary, Los Angeles,California.

Pierce, J.L. & Newstrom, J.W. (1983) Thedesign of flexible work schedules and em-ployee responses: relationships andprocesses. Journal of Occupational Behav-iour, 4, 247-262.

Pollard T.M. (1997) Physiological conse-quences of everyday psychosocial stressCollegium Antropologicum, Vol.21, No.1,pp.17-28

Poppius E., Tenkanen L., Kalimo R., Hein-salmi P. (1999) The sense of coherence, oc-cupation and the risk of coronary heartdisease in the Helsinki Heart Study SocialScience & Medicine, Vol.49, No.1, pp.109-120

Porter, L. W. (1990) Commitment patternsin industrial organizations. Paper to: Soci-ety for Industrial and Organizational psy-chology, Miami Beach, Florida (April).

Powell, L. H. (1987) Issues in the measure-ment of the Type A behaviour pattern. In:S. V. Kasl, & C.L. Cooper (eds) Stress andHealth: Issues in Research Methodology.Wiley & Sons, Chichester.

Quick, J.C., & Quick, J.D. (1984) Organiza-tional Stress and Preventive Management.McGraw-Hill, New York.

Quick, J.C., Murphy, L.R., & Hurrell, J.J.(1992a) Stress and Well-being at Work:Assessments and Interventions for Occu-pational Mental Health. American Psycho-logical Association, Washington DC.

Quick, J.C., Joplin, J.R., Gray, D.A., & Coo-ley, E.C. (1993) The occupational life cycleand the family. In L.L’Abate (ed.), Hand-book of Developmental Family Psychologyand Psychopathology. Wiley & Sons, NewYork.

Quick, J. C., Murphy, L. R., Hurrell, J. J., &Orman, D. (1992b) The value of work, therisk of distress, and the power of preven-tion. In: J.C. Quick, L.R. Murphy, & J.J. Hur-rell (eds) Stress & Well Being at Work:Assessments and Interventions for Occu-pational Mental Health. American Psycho-logical Association, Washington DC.

Quick, J.C., Camara, W.J., Hurrell, J.J.,Johnson, J.V., Piotrkowski, C.S., Sauter,S.L., & Spielberger, C.D. (1997) Introduc-tion and historical overview. Journal of Oc-cupational Health Psychology, 2 (1), 3-6


■158

Rahe, R. H. (1969) Multi-cultural correla-tions of life change scaling: America,Japan, Denmark, and Sweden. Journal ofPsychosomatic Research, 13, 191-195.

Repetti, R. L. (1987) Linkages betweenwork and family roles. In: S. Oskamp (ed)Applied Social Psychology Annual Vol 7.Family Processes and Problems. Sage, Bev-erly Hills.

Repetti, R. L., & Crosby, F. (1984) Genderand depression: exploring the adult roleexplanation. Journal of Social and ClinicalPsychology, 2, 57-70.

Repetti, R. L., Matthews, R. A., & Waldron,I. (1989) Employment and women’shealth: effects of paid employment onwomen’s mental and physical health.American Psychologist, 44, 1394-1401.

Rice, H.K. (1963) The responding-rest ratioin the production of gastric ulcers in therat. Psychological Reports, 13, 11-14.

Richter, P., & Schmidt, C. F. (1988) Arbeit-sanforderungen und Beanspruchungsbe-wältigung bei Herzinfarkt-Patienten - eintätigkeitspsychologischer Diagnos-tikansatz. In: H. Schröder & J. Guthke(Hrsg.) Fortschritte der klinischen Persön-lichkeitspsychologie und klinischen Psy-chodiagnostik. Barth, Leipzig. pp 49-56(Psychotherapie und Grenzgebiete, 9).

Richter, P., Rudolph, M., & Schmidt, C. F.(1995) FABA: Fragebogen zur Analyse be-lastungsrelevanter Anforderungsbewälti-gung. Technische Universität, Institut fürArbeits-, Organisations- und Sozialpsy-chologie, Dresden (Methodensammlung,4).

Riley, V. (1979) Stress - cancer contradic-tions: a continuing puzzlement. CancerDetection and Prevention, 2, 159-162.

Riley, V. (1981) Psychoneuroendocrine in-fluences on immunocompetence and neo-plasia. Science, 212, 1100-1109.

Riley, V., Fitzmaurice, M. A., & Spackman,D. H. (1981) Psychoneuroimmunologicalfactors in neoplasia: studies in animals. In:R. Ader (ed) Psychoneuroimmunology.Academic Press, New York.

Rimon, R.A. & Laakso, R. (1985) Life stressand rheumatoid arthritis: a 15-year follow-up study. Psychotherapy and Psychoso-matics, 43, 38-43.

Robertson, I. T., & Cooper, CL (1983) Hu-man Behaviour in Organizations. MacDon-ald and Evans Ltd., London.

Ronen, S. (1981) Flexible Working Hours;An Innovation in the Quality of Work Life.McGraw Hill, New York.

Rosa, R.R. & Colligan, M.J. (1986) TheNIOSH Fatigue Test Battery: LaboratoryValidation of a Portable System for FieldStudy of Extended Workdays and WorkScheduling. NIOSH internal report. Nation-al Institute for Occupational Health,Cincinnati.

Rosa, R. R., Colligan, M. J., & Lewis, P.(1989) Extended workdays: effects of 8-hour and 12-hour rotating shifts scheduleson performance, subjective aleartness,sleep patterns and psychosocial variables.Work & Stress, 3, 2-32.

Rose, R. M., Poe, R. O., & Mason, J. W.(1967) Observations on the relationship


159■

between psychological state, 17-OHCS ex-cretion and epinephrine, norepinephrine,insulin, BEI, estrogen and androgen levelsduring basic training. Psychosomatic Med-icine, 29, 544.

Rosenman, R. H., Friedman, M., Straus, R.,Wurm, M., Kositchek, R., Hahn, W., &Werthessen, N. T. (1964a) A predictivestudy of coronary heart disease: appendix.Journal of the American Medical Associa-tion, 189, 1-4.

Rosenman, R. H., Friedman, M., Straus, R.,Wurm, M., Kositchek, R., Hahn, W., &Werthessen, N. T. (1964b) A predictivestudy of coronary heart disease. Journal ofthe American Medical Association, 189,113-124.

Rosenstock, L. (1997) Work organizationresearch at the National Institute for Occu-pational Safety and Health. Journal of Oc-cupational Health Psychology, 2(1), 7-100

Rotheiler, E., Richter, P. Rudolf, M. & Hin-ton, J. W. (1997) Further cross-cultural fac-tor validation on the FABA self reportinventory of coronary-prone behaviours.Psychology and Health 12, 505-512.

Russek, H. I., & Zohman, B. L. (1958) Rela-tive significance of heredity, diet and occu-pational stress in CHD of young adults.American Journal of Medical Sciences,235, 266-275.

Rutenfranz, J. (1982) Occupational healthmeasures for night and shift workers. Jour-nal of Human Ergology, 11 (supplement),67-86.

Rutenfranz, J., Haider, M., & Koller, M.(1985) Occupational health measures fornight workers and shift workers. In: S.Folkard & T. H. Monk (eds) Hours of Work:Temporal Factors in Work Scheduling. Wi-ley & Sons, Chichester.

Rutenfranz, J., Colquhoun, W. P., Knauth,P., & Ghata, J. N. (1977) Biomedical andpsychosocial aspects of shift work: a re-view. Scandinavian Journal of Work andEnvironmental Health, 3, 165-182.

Ryman, D.H., & Ursin, H. (1979) Factoranalyses of the physiological responses ofcompany commanders to stress. Quoted in: Ursin, H. (1979) Personality, activationand somatic health: a new psychosomatictheory. In: S. Levine & H. Ursin (eds) Cop-ing and Health. Plenum Press, New York.

Ryman, D.H., Naitoh, P., & Englund, C. E.(1989) Perceived exertion under conditionsof sustained work and sleep loss. Work &Stress, 3, 5-68.

Salo, K. (1995) Teacher stress and copingover an autumn term in Finland. Work &Stress, 9 (1), 55-66.

Salvendy, G., & Smith, M. (1981) MachinePacing and Occupational Stress. Wiley &Sons, Chichester.

Sandler, I N., And Lakey, B. (1982). Locusof control as stress moderator: the role ofcontrol perceptions and social support.American Journal of Community Psycholo-gy, 10, 65-79

Sarason, I. G., de Monchaux, C., & Hunt, T.(1975) Methodological issues in the as-sessment of life stress. In: L. Levi (ed) Emo-


■160

tions: Their Parameters and Measurement.Raven, New York.

Saunders, D. (1956). Moderator variablesin prediction.Educational and Psychologi-cal Measurement, 16, 209-222.

Sauter, S. L. (1992) Introduction to theNIOSH proposed National Strategy. In: G.P.Keita & S.L. Sauter (eds) Work and Well Be-ing: An Agenda for the 1990s. AmericanPsychological Association, WashingtonDC.

Sauter, S.L. & Murphy, L.R. (1995) Organi-zational risk factors for job stress. Wash-ington, DC: APA

Sauter, S.L., Hurrell, J.J., & Cooper, CL(1989) Job Control and Worker Health.Wiley & sons, Chichester.

Sauter, S. L., Murphy, L.R., & Hurrell, J.J.(1990) Prevention of work-related psycho-logical disorders: A national strategy pro-posed by the National Insitute forOccupational Safety and Health (NIOSH).American Psychologist, 45, 146-158.

Sauter, S. L., Murphy, L. R., & Hurrell, J. J.(1992) Prevention of work related psycho-logical disorders: a national strategy pro-posed by the National Institute forOccupational Safety and Health. In: G. P.Keita & S. L. Sauter (eds) Work and Well-Being: An Agenda for the 1990s. Ameri-can Psychological Association,Washington DC.

Sauter, S.L., Hurrell, J. J., Jr., Murphy, L.R.,& Levi, L. (Eds.). (1998). Psychosocial andorganizational factors. In J.M. Stellman(Ed.) Encyclopaedia of Occupational

Health and Safety, Fourth Edition, 2, pp.34.2-34.6. Geneva, Switzerland: Interna-tional Labour Organization

Schaubroeck, J. & Merritt, D.E. (1997) Di-vergent effects of job control on copingwith work stressors:The key role of self-ef-ficacy. Academy Of Management Journal,40, 738-754

Scheck,C.L., Kinicki,A.J., Davy, J.A. (1997)Testing the mediating processes betweenwork stressors and subjective well-being.Journal Of Vocational Behavior, 50, 96-123

Scheuch, K. (1990): Psychosoziale Fak-toren im Arbeitsprozeß und Gesundheit:Einführung. Z. ges. Hyg. 36, 403-407

Scheuch, K. (1996): Stress and resources atwork in a changing society. Bremerhaven:Wirtschaftsverlag NW, pp.95-109(Schriftenreihe der Bundesanstalt für Ar-beitsmedizin: Tagungsbericht 11)

Schneiderman, N. (1978) Animal modelsrelating behavioural stress and cardiovas-cular pathology. In T. Dembroski (ed) Pro-ceedings of the Forum on Coronary-ProneBehaviour. DHEW publication no.(NIH) 78-1451. US Government Printing Office,Washington DC.

Schonpflug, F. & Battmann, A. (1988) Thecosts and benefits of coping. In: S. Fisher &J. Reason (eds) Handbook of Life Stress,Cognition and Health. Wiley & Son, Chich-ester

Schott, F. (1992) Panel comments: workdesign. In: G. P. Keita & S. L. Sauter (eds)Work and Well-Being: An Agenda for the


161■

1990s. American Psychological Associa-tion, Washington DC.

Schriber, J.B. & Gutek, B.A. (1987) Sometime dimensions of work: measurement ofan underlying aspect of organizational cul-ture. Journal of Applied Psychology, 7,624-650.

Schrijvers C.T.M., van de Mheen H.D.,Stronks K, Mackenbach JP (1998) Socioe-conomic inequalities in health in the work-ing population: the contribution ofworking conditions. International JournalOf Epidemiology, Vol.27, No.6, pp.1011-1018

Schwartz, G. (1980) Stress management inoccupational settings. Public Health Re-ports, 95, 99-108.

Scott, R., & Howard, A. (1970) Models ofstress. In: S. Levine & N. Scotch (eds) SocialStress. Aldine, Chicago.

Scriven, M. (1967) The methodology ofevaluation. In: R. E. Stake (ed) Perspectivesof Curriculum Evaluation. AERA Mono-graph Series on Curriculum Evaluation no.1, Rand McNally, Chicago.

Sells, S. B. (1970) On the nature of stress.In: J. McGrath (ed) Social and Psychologi-cal factors in Stress. Holt, Rinehart & Win-ston, New York.

Selye, H. (1936) A syndrome produced bydiverse nocuous agents. Nature, 138, 32.

Selye, H. (1950) Stress, Acta Incorporated,Montreal.

Selye, H. (1956) Stress of Life. McGraw-Hill, New York.

Selye, H. (1976) Stress in Health and Dis-ease. Butterworths, Boston.

Sharit, J., & Salvendy, G. (1982) Occupa-tional stress: review and appraisal. HumanFactors, 24, 129-162.

Sheffield, D., Dobbie, D. & Carroll, D.(1994) Stress, social support, and psycho-logical wellbeing in secondary schoolteachers. Work & Stress, 8 (3), 235-243.

Shekelle, R. B., Ostfeld, A. M., & Paul, O.(1969) Social status and incidence of CHD.Journal of Chronic Disorders, 22, 381-394.

Shekelle, R. B., Gale, M., Ostfeld, A. M., &Paul, O. (1983) Hostility, risk of coronaryheart disease, and mortality. Psychosomat-ic Medicine, 45, 109-114.

Shilling, S., & Brackbill, R.M. (1987) Occu-pational health and safety risks and poten-tial health consequences perceived by USworkers. Public Health Reports, 102, 36-46.

Shinn, M, Rosario, M, Morch, H & Chest-nut, D.E. (1984) Coping with job stress andburnout in the human services. Journal ofPersonality and Social Psychology, 46, 864-876.

Shirom, A., Eden, D., Silberwasser, S., &Kellerman, J. J. (1973) Job stresses and riskfactors in CHD among occupational cate-gories in kibbutzim. Social Science andMedicine, 7, 875-892.

Siegrist, J. (1990) Chronischer Distress undkoronares Risiko: Neue Erkenntnisse undihre Bedeutung für die Pravention. In: M.Arnold, C. v. Ferber, & K.-D. Henke (Hrsg.)


■162

Ökonomie der Prävention. Bleicher, Gerlin-gen.

Sigman, A. (1992) The state of corporatehealth care. Personnel Management, 47-61.

Simon R.I. (1999) Chronic posttraumaticstress disorder: A review and checklist offactors influencing prognosis. Harvard Re-view Of Psychiatry, Vol.6, No.6, pp.304-312

Singer, J.A., Neale, M.S., Schwartz, G.E., &Schwartz, J. (1986) Conflicting perspec-tives on stress reduction in occupationalsettings: a systems approach to their reso-lution. In M.F. Cataldo & T.J. Coates (eds)Health and Industry: A Behavioural Medi-cine Perspective. Wiley & Sons, New York.

Sklar, L. S., & Anisman, H. (1981) Stressand cancer. Psychological Bulletin, 89, 36-406.

Sleeper, R. D. (1975) Labour mobility overthe life cycle. British Journal of IndustrialRelations, 13.

Smewing, C., and Cox, T. (1996) The orga-nizational health of health care institutionsin the United Kingdom. Proceedings of theIV Seminar on Organizational Psychologyof Health Care, European Network of Or-ganizational Psychologists, Munich.

Smith, A. (1991) A review of the non audi-tory effects of noise on health. Work &Stress, 5, 49-62.

Smith, M.J. (1985) Machine-paced workand stress. In CL Cooper & M.J. Smith (eds)Job Stress and Blue Collar Work. Wiley &sons, Chichester.

Smith, M.J., Hurrell, J.J., & Murphy, R.K.(1981) Stress and health effects in pacedand unpaced work. In G. Salvendy & M.J.Smith (eds) Pacing and OccupationalStress. Taylor and Francis, London.

Smith, R. P. (1981) Boredom: a review. Hu-man factors, 23, 329-340.

Spector, P. E. (1986) Perceived control byemployees: a meta analysis of studies con-cerning autonomy and participation in de-cision making. Human Relations, 39,1005-1016.

Spector, P.E. (1987a) Interactive effects ofperceived control and job stressors on af-fective reactions and health outcomes forclerical workers. Work & Stress, 1, 155-162.

Spector, P.E. (1987b) Method variance asan artifact in self-reported affect and per-ceptions at work: Myth or significant prob-lem? Journal of Applied Psychology, 72 (3),438-443.

Spielberger, C.D. (1976) The nature andmeasurement of anxiety. In C.D. Spielberg-er and R. Diaz-Guerrero (eds) Cross-Cultur-al Anxiety. Hemisphere, Washington DC.

Spurgeon, A., & Harrington, J. M. (1989)Work performance and health of juniorhospital doctors - a review of the literature.Work & Stress, 3, 117-128.

Spurgeon A., Harrington J.M., & CooperC.L. (1997) Health and safety problems as-sociated with long working hours: A re-view of the current position. OccupationalAnd Environmental Medicine, Vol.54,No.6, pp.367-375


163■

Stainbrook, G. L., & Green, L. W. (1983)Role of psychosocial stress in cardiovascu-lar disease. Houston Heart Bulletin, 3, 1-8.

Stampi, C. (1989) Polyphasic sleep strate-gies improve prolonged sustained perfor-mance: a field study on 99 sailors. Work &Stress, 3, 41-55.

Standing, H. & Nicolini, D. (1997) Reviewof Work-Related Violence. Health & SafetyExecutive Contract Research Report143/1997. Sudbury, Suffolk, U.K.: HSEBooks.

Stansfeld, S.A., Fuhrer, R., Shipley, M.J. &Marmot, M.G. (1999) Work characteristicspredict psychiatric disorder: prospective re-sults from the Whitehall II study. Occupa-tional and Environmental Medicine, 56,302-307

Stansfeld, S.A., North, F.M., White, I. &Marmot, M.G. (1995) Work characteristicsand psychiatric disorder in civil servants inLondon. Journal of epidemiology andCommunity Health, 49, 48-53.

Stein, M., Keller, S., & Schleifer, S. (1981)The hypothalamus and the immune re-sponse. In H. Weiner, M. Hofer, & A.Stunkard (eds) Brain, Behaviour and BodilyDisease. Raven, New York.

Stewart, R. (1976) Contrasts in Manage-ment. McGraw-Hill, New York.

Stone, E.A. (1975) Stress and cate-cholamines. In A. Friedhoff (ed) Cate-cholamiones and Behaviour, Vol 2.Plenum, New York.

Strauss, G. (1974) Workers: attitudes andadjustments. In the American Assesmbly,

Columbia University, The Worker and theJob: Coping with Change. Prentice-Hall,Englewood Cliffs, NJ.

Surtees P.G. & Wainwright N.W.J (1998)Adversity over the life course: assessmentand quantification issues. Stress Medicine,vol. 14, no. 4, pp. 205-211

Susser, M. (1967) Causes of peptic ulcer: aselective epidemiological review. Journalof Chronic Diseases, 20, 435-456.

Sutherland, V. J., & Cooper, CL (1990) Un-derstanding stress: psychological persepc-tive for health professionals. Psychology &Health, series: 5. Chapman and Hall, Lon-don.

Symonds, C.P. (1947) Use and abuse of theterm flying stress. In Air Ministry, Psycho-logical Disorders in Flying Personnel of theRoyal Air Force, Investigated during theWar, 1939-1945. HMSO, London.

Szabo, S., Maull, E.A., & Pirie, J. (1983) Oc-cupational Stress: understanding, recogni-tion and prevention. Experientia, 39,1057-1180.

Tavistock Institute of Human Relations(1986) Violence to Staff: A Basis for As-sessment and Prevention. HMSO, London.

Terry, D.J. & Jimmieson, N.L (1999) Workcontrol and employee well-being: Adecade review. In Cooper & Robertson(Eds) International review of industrial andorganizational psychology 1999, Vol. 14.(pp. 95-148). Chichester, England UK:American Ethnological Press.


■164

Theorell, T. (1997) Fighting for and losingor gaining control in life. Acta PhysiologicaScandinavica, 161, Suppl. 640, 107-111.

Thompson, S.C. (1981) Will it hurt less if Ican control it? A complex answer to a sim-ple question. Psychological Bulletin, 90,89-101.

Thomson, L., Griffiths, A., Cox, T. (1998)The psychometric quality of self-reportedabsence data. Proceedings of the Interna-tional Work Psychology Conference. Uni-versity of Sheffield, Institute of WorkPsychology. ISBN 0 9533504 0 1.

Turkkan, J.S., Brady, J.V., & Harris A.H.(1982) Animal studies of stressful interac-tions: a behavioural-physiologicaloverview. In L. Goldberger & S. Breznitz(eds) Handbook of Stress: Theoretical andClinical Aspects. Free Press, New York.

Turnage, J. J., & Spielberger, C.D. (1991)Job stress in managers, professionals andclerical workers. Work & Stress, 5, 165-176.

Ulrich, R.E., & Azrin, N.H. (1962) Reflexivefighting in response to aversive stimula-tion. Journal of the Experimental Analysisof Behaviour, 5, 511-520.

Uris, A. (1972) How managers ease jobpressures. International Management,June, 45-46.

Ursin, H. (1979) Personality, activation andsomatic health: a new psychosomatic the-ory. In: S. Levine & H. Ursin (eds) Copingand Health. Plenum Press, New York.

Ursin, H., Mykletun, R., Tonder, O.,Vaernes, R. J., Relling, G., Isaksen, E., &

Murisaon, R. (1984) Psychological stressfactors and concentrations of im-munoglobulins and complement compo-nents in humans. Scandinavian Journal ofPsychology, 23, 193-199.

United States Department of Health andHuman Services USDHHS (1980) New De-velopments in Occupational Stress: Pro-ceedings of a Conference. US Departmentof Health and Human Services, no: NIOSH81-102, US Government Printing Office,Washington DC.

Vaernes, R. J., Warncke, M., Eidsvik, S.,Aakvaag, A., Tonder, O., & Ursin, H. (1987)Relationships between perceived healthand psychological factors among subma-rine personnel: endocrine and immunolog-ical effects. In: A. Marroni & G. Oriani (eds)Diving and Hyperbaric Medicine. Academ-ic Press, New York.

Vaernes, R. J., Myhre, G., Aas, H., Homnes,T., Hansen, I., & Tonder, O. (1991) Rela-tionships between stress, psychologicalfactors, health and immune levels amongmilitary aviators. Work & Stress, 5, 5-16.

Vaernes, R. J., Knardahl, S., Romsing, J.,Aakvaag, A., Tonder, O., Walter, B., &Ursin, H. (1988) Relationships between en-vironmental problems, defense strategiesand health among shiftworkers in theprocess industry. Work & Stress, 1, 7-15.

Van der Hek, H. & Plomp, H.N. (1997) Oc-cupational stress management pro-grammes: a practical overview ofpublished effect studies. OccupationalMedicine, 47 (3), 133-141.


165■

Van Raaij, M.T.M., Oortgiesen, M., Tim-merman, H.H., Dobbe, C.J.G., & VanLov-eren, H. (1996) T ime-dependentdifferential changes of immune function inrats exposed to chronic intermittent noise.Physiology & Behavior, Vol.60, No.6,pp.1527-1533

Viswesvaran C., Sanchez J.I. & Fisher J.(1999) The role of social support in theprocess of work stress: A meta-analysis.Journal Of Vocational Behavior, Vol.54,No.2, pp.314-334

Volhardt, B.R., Ackerman, S.H., Grayzel,A.I., & Barland, P. (1982) Psychologicalydistinguishable groups of rheumatoidarthritis patients: a controlled single blindstudy. Psychosomatic Medicine, 44, 353-361.

Von Restorff, W., Kleinhanss, G., Schaad,G., & Gorges, W. (1989) Combined workstresses: effect of reduced air renewal onpsychological performance during 72hrsustained operations. Work & Stress, 3,15-20.

Voydanoff, P., & Kelly, R. F. (1984) Determi-nants of work-related family problemsamong employed parents. Journal of Mar-riage and the Family, 46, 881-892.

Wall, T.D. & Clegg, C.W. (1981) A longitu-dinal study of group work redesign. Jour-nal of Occupational Behaviour, 2 , 31-49.

Wallhagen M.I., Strawbridge W.J., CohenR.D., Kaplan G.A. (1997) An increasingprevalence of hearing impairment and as-sociated risk factors over three decades ofthe Alameda County Study. American

Journal Of Public Health, Vol.87, No.3,pp.440- 442

Wardell, W. I., Hyman, M., & Bahnson, C B.(1964) Stress and coronary heart disease inthree field studies. Journal of Chronic Dis-eases, 17, 73-84.

Warr, P. B. (1982) Psychological aspects ofemployment and unemployment. Psycho-logical Medicine, 12, 7-11.

Warr, P. B. (1983) Work, jobs and unem-ployment. Bulletin of the British Psycholog-ical Society, 36, 305-311.

Warr, P. B. (1987) Work, Unemploymentand Mental Health. Cambridge UniversityPress, Cambridge.

Warr, P. B. (1990) Decision latitude, job de-mands and employee well-being. Work &Stress, 4, 285-294.

Warr, P. B. (1992) Job features and exces-sive stress. In: R. Jenkins & N. Coney (eds)Prevention of Mental Ill Health at Work.HMSO, London.

Warshaw, L.J. (1979) Managing Stress. Ad-dison-Wesley, Reading, Mass.

Waterhouse, J.M., Folkhard, S., & Minors ,D.S. (1992) Shiftwork, Health and Safety:An Overview of the Scientific Literature1978-1990. HMSO, London.

Watson, D. & Clark, L.A. (1984) Negativeaffectivity: The disposition to experienceaversive emotional states. PsychologicalBulletin, 96 (3), 465-490.

Webb, E. J., Campbell, D. T., Schwartz, R.D. & Sechrest, L. (1966) Unobtrusive Mea-


■166

sures: Nonreactive Research in the SocialSciences. Rand McNally, Chicago.

Weinberg A., Cooper C.L., & Weinberg A.(1999) Workload, stress and family life inBritish Members of Parliament and thepsychological impact of reforms to theirworking hours. Stress Medicine, Vol.15,No.2, pp.79-87

Weiner, H. (1977) Psychobiology and Hu-man Disease. Elsevier, New York.

Weiss, J.M. (1972) Psychological factors instress and disease. Scientific American,226, 104-113.

Weitzman, E.D., Boyar, R.M., Kapen, S., &Hellman, L. (1975) The relationship ofsleep and sleep stages to neuroendocrinesecretion and biological rhythms in man.Recent Progress Hormone Research, 31,399-446.

Welford, A. T. (1973) Stress and perfor-mance. Ergonomics, 16, 567-580.

Wheaton, B. (1983) Stress, personal cop-ing resources and psychiatric symptoms:an investigation of interactive model. Jour-nal of Health and Social behaviour, 24.

Wilensky, H. (1960) Work, careers and so-cial integration. International Social Sci-ence Journal, 4, 54 -560.

Williams, R. B., Barefoot, J. C., & Shekelle,R. B. (1985) The health consequences ofhostility. In: M. A. Chesney & R. H. Rosen-man (eds) Anger and Hostility in Cardio-vascular and behavioural Disorders.Hemisphere Publishing Corp., WashingtonDC.

Williams, R. B., Haney, T. L., Lee, K. L.,Kong, Y., Blumenthal, J. A., & Whalen, R.E.(1980) Type A behaviour, hostility andcoronary atherosclerosis. PsychosomaticMedicine, 42, 539-549.

Windel A, Zimolong B. (1997) Group workand performance in business. Gruppendy-namik-Zeitschrift Fur AngewandteSozialpsychologie, 28 (4), pp.333-35

Winnubst, J. A. M., & Schabracq, M. J.(1996) Social Support, Stress and Organi-zation: Towards Optimal Matching. In : M.J. Schabracq, J. A. M. Winnubst, & C. L.Cooper (eds) Handbook of work andhealth psychology. John Wiley & Sons,Chichester.

World Health Organization [WHO] (1986)Constitution of the World Health Organi-zation. In: Basic Documents (36th ed).World Health Organization, Geneva.

Wykes, J. & Whittington, R. (1991) Copingstrategies used by staff following assaultby a patient: an exploratory study. Work &Stress, 5 (1), 37-48.

Wyler, A., Masuda, M., & Holmes, T.(1968) Seriousness of illness scale. Journalof Psychosomatic Research, 11, 363-375.

Wynne, R., Clarkin, N., Cox, T., & Griffiths,A. (1997). Guidance on the Prevention ofViolence at Work. Luxembourg: EuropeanCommission, DG-V.

Zegans, L.S. (1982) Stress and the Devel-opment of Somatic Disorders. In L.Gold-berger & S. Breznitz (eds) Handbook ofStress: Theoretical and Clinical Aspects.Free Press, New York.


167■

A g e n c y ’ s P r o j e c t M a n a g e r

Dr. M. AaltonenEuropean Agency for Safety and Health atWork Gran Via, 33E-48009 BilbaoSPAIN

P r o j e c t C o n s u l t a n t s

Prof. Tom Cox CBEDr. Amanda Griffiths Mr. Eusebio Rial-GonzálezInstitute of Work, Health and Organisa-tions (I-WHO)University of Nottingham Business SchoolNottingham NG8 1BBUNITED KINGDOM

P r o j e c t m e m b e r s w i t h i n t h e T o p i cC e n t r e o n R e s e a r c h - W o r k a n d H e a l t h

Dr. V. Borg (Task leader)National Institute of Occupational Health -Arbejdsmiljøinstituttet (AMI)

Lersoe Parkallé 105DK-2100 COPENHAGENDENMARK

Dr. A. BrouwersTNO Work and Employment (TNO)P.O. Box 7182130 AS HOOFDDORPThe NETHERLANDS

Dr. K. KuhnBundesanstalt für Arbeitsschutz und Ar-beitsmedizin (BAuA)Friedrich-Henkel-Weg 1-25D-44149 DORTMUNDGERMANY

Dr. M. Neboit Dr. P. MéreauInstitut National de Recherche et de Sécu-rité (INRS)Centre de Recherche et de FormationAvenue de Bourgogne, P.O. Box 2754501 VANDOEUVRE CEDEXFRANCE

Dra. S. NogaredaCentro Nacional de Condiciones de Traba-jo (INSHT)c/Dulcet 2-1008034 BARCELONA SPAIN

L e a d O r g a n i s a t i o n o f t h e T o p i c C e n t r eo n R e s e a r c h - W o r k a n d H e a l t h

Dr. J-L. Marié Dr. J-C. AndréInstitut National de Recherche et de Sécu-rité (INRS)30 rue Olivier NoyerF-75014 PARIS FRANCE

A P P E N D I X 1 . P R O J E C T

O R G A N I S A T I O N

European Agency for Safety and Health at Work

Research on Work-related Stress

Luxembourg: Office for Official Publications of the European Communities

2000 — 167 pp. — 14.8 x 21 cm

ISBN 92-828-9255-7

Price (excluding VAT) in Luxembourg: EUR 11

stress

Documents

unpublished phd thesis

administrative science quarterly

internet www page

council directive 89 391 eec

coronary heart disease

describing trends due

junior hospital doctors

autonomic nervous system