Occupational Burnout and Health

Home , Ergophobia, Occupational burnout

Occupational burnout and health

Kirsi Ahola

People and Work Research Reports 81

Finnish Institute of Occupational Health Helsinki, Finland 2 CONTENTS

ACKNOWLEDGEMENTS ...... 6

ABSTRACT ...... 8

TIIVISTELMÄ ...... 10

LIST OF ORIGINAL PUBLICATIONS ...... 12

1. INTRODUCTION ...... 14 1.1 Concept of occupational burnout ...... 15 1.2 Socio-demographic factors and occupational burnout 18 1.3 Job strain and occupational burnout ...... 20 1.4 Occupational burnout and health problems ...... 23 Mental health problems ...... 24 Musculoskeletal problems ...... 33 Cardiovascular diseases ...... 33 Summary and limitations of previous research .. 35 1.5 Occupational burnout and sickness absence ...... 36

2. PRESENT STUDY ...... 39 2.1 Framework of the study ...... 39 2.2 Aims of the study ...... 41

3. METHODS ...... 43 3.1 Procedure ...... 43 3.2 Participants ...... 44 3.3 Measures ...... 46 Occupational burnout ...... 46 Mental disorders ...... 49 Depressive symptoms ...... 50 Physical illnesses ...... 51 Sickness absence ...... 51 Job strain ...... 52 Socio-demographic factors ...... 52 Confounding factors ...... 52 3.4 Statistical analyses ...... 53

3 CONTENTS

4. RESULTS ...... 55 4.1 Association between socio-demographic factors and occupational burnout ...... 55 Gender, age and burnout ...... 55 Education, occupational grade and burnout ...... 58 Marital status and burnout ...... 58 Brief summary ...... 59 4.2 Job strain in relation to occupational burnout and depressive disorders ...... 59 4.3 Association between occupational burnout and disorders and illnesses ...... 61 Burnout and mental disorders ...... 61 Burnout and musculoskeletal disorders ...... 65 Burnout and cardiovascular diseases ...... 67 Co-occurrence of burnout with mental disorders and physical illnesses ...... 68 Brief summary ...... 70 4.4 Contribution of occupational burnout to long sickness absences ...... 71

5. DISCUSSION ...... 74 5.1 Synopsis of the main ﬁ ndings ...... 74 5.2 Occupational burnout in the arena of occupational health psychology ...... 75 Burnout and ill health ...... 75 Depressive and anxiety disorders ...... 75 Alcohol dependence ...... 78 Musculoskeletal disorders ...... 79 Cardiovascular diseases ...... 80 Summary ...... 81 Distinction between burnout and ill health ...... 82 Association with work characteristics ...... 82 Association with work disability ...... 84 Summary ...... 84 Role of individual factors in burnout ...... 85 Gender and the level of burnout ...... 85 Gender differences in the association between burnout and ill health ...... 86 Age and the level of burnout ...... 89 Summary ...... 90

4 CONTENTS

5.3 Evaluation of the study ...... 91 Assessment of burnout ...... 91 Major strengths ...... 92 Study limitations ...... 93 5.4 Conclusions ...... 94 5.5 Policy implications ...... 96

REFERENCES ...... 97

ORIGINAL PUBLICATIONS ...... 117

5 ACKNOWLEDGEMENTS

Th is study was carried out at the Finnish Institute of Occupational Health (FIOH). I would like to express my sincere gratitude to the Director General of FIOH, Professor Harri Vainio, for the facilities provided to me for my work. I also extend my warmest thanks to the Director of the Centre of Expertise for Work Organisations, Professor Kari Lind- ström, for the opportunity to carry out my research at FIOH. As an academic dissertation, this work is being presented in the Department of Psychology at the University of Helsinki, the opportunity for which I am much obliged. Th is project was a part of the Health 2000 Study which was organized by the National Public Health Institute. I am very grateful to the Chair- man of the Health 2000 Study Project Group, Professor Arpo Aromaa, and the Chairman of the Mental Health Working Group of the Health 2000 Study, Professor Jouko Lönnqvist, for giving me the opportunity to participate in the Health 2000 Study. I wish to express my deepest appreciation to the reviewers of this thesis, Professor Ulla Kinnunen, from the University of Tampere, and Professor Jyrki Korkeila, from the University of Turku, for their construc- tive comments, which signiﬁ cantly improved this thesis. Most of all, I am immensely indebted to my supervisors, docent Teija Honkonen, M.D., Ph.D., and Professor Mika Kivimäki, for sharing their expertise with me. I extend my warmest thanks to the co-authors of the original publications of this thesis, Professor Emerita Raija Kalimo, Pro- fessor Jouko Lönnqvist, Professor Erkki Isometsä, Professor Jussi Vahtera, docent Marianna Virtanen, Ph.D., docent Seppo Koskinen, M.D., Ph.D., docent Marja Pertovaara, M.D., Ph.D., docent Sami Pirkola,

6 ACKNOWLEDGEMENTS

M.D., Ph.D., and mathematician Erkki Nykyri, Lic.Sc., for their eff ort and valuable comments. It was Raija Kalimo who originally introduced me to the fi eld of burnout research, for which I am deeply grateful. I express my special thanks to my team leader, Sirkku Kivistö, M.A, Lic., for her endless support and encouragement. I have been very privi- leged in having such stimulating co-workers in the Centre of Expertise in Work Organisations, and especially in the Work and Mental Health Team at FIOH. My thanks to you all! I extend warm thanks to all the participants, fi eld workers, and project staff of the Health 2000 Study for their eff ort and assistance. I warmly thank Georgianna Oja, E.L.S., for the linguistic editing of the original publications and this thesis. Finally, I am so very grateful to my friends and relations, and especially to my children Jonni and Sanja, for my most valuable resource during this project: continuous companionship in everyday life. Th is study was supported by the Finnish Work Environment Fund (project no. 106382) and the Academy of Finland (projects no. 105195 and 117604), for which I am deeply indebted.

Helsinki, September 2007

Kirsi Ahola

7 ABSTRACT

Occupational burnout is assumed to be a negative consequence of chronic work stress. In this study, it was explored in the framework of occupational health psychology, which focusses on psychologically mediated processes between work and health. Th e objectives were to examine the overlap between burnout and ill health in relation to mental disorders, musculoskeletal disorders, and cardiovascular diseases, which are the three commonest disease groups causing work disability in Finland; to study whether burnout can be distinguished from ill health by its relation to work characteristics and work disability; and to determine the socio-demographic correlates of burnout at the population level. A nationally representative sample of the Finnish working population aged 30 to 64 years (n = 3151–3424) from the multidisciplinary epidemiological Health 2000 Study was used. Burnout was measured with the Maslach Burnout Inventory – General Survey. Th e diagnoses of common mental disorders were based on the standardized mental health interview (the Composite International Diagnostic Interview), and physical illnesses were determined in a comprehensive clinical health examination by a research physician. Medically certiﬁ ed sickness absences exceeding 9 work days during a 2-year period were extracted from a register of Th e Social Insurance Institution of Finland. Work stress was operationalized according to the job strain model. Gender, age, education, occupational status, and marital status were recorded as socio-demographic factors. Occupational burnout was related to an increased prevalence of depressive and anxiety disorders and alcohol dependence among the men and women. Burnout was also related to musculoskeletal disorders among

8 ABSTRACT

the women and cardiovascular diseases among the men independently of socio-demographic factors, physical strenuousness of work, health behaviour, and depressive symptoms. Th e odds of having at least one long, medically-certiﬁ ed sickness absence were higher for employees with burnout than for their colleagues without burnout. For severe burnout, this association was independent of co-occurring common mental disorders and physical illnesses for both genders, as was also the case for mild burnout among the women. In a subgroup of the men with absences, severe burnout was related to a greater number of absence days than among the women with absences. High job strain was associated with a higher occurrence of burnout and depressive disorders than low job strain was. Of these, the association between job strain and burnout was stronger, and it persisted after control for socio-demographic factors, health behaviour, physical illnesses, and various indicators of mental health. In contrast, job strain was not related to depressive disorders after burnout was accounted for. Among the working population over 30 years of age, burnout was positively associated with age. Th ere was also a tendency towards higher levels of burnout among the women with low educational attainment and occupational status and among the unmarried men. In conclusion, a considerable overlap was found between burnout, mental disorders, and physical illnesses. Still, burnout did not seem to be totally redundant with respect to ill health. Burnout may be more strongly related to stressful work characteristics than depressive disorders are. In addition, burnout seems to be an independent risk factor for work disability, and it could possibly be used as a marker of health-impairing work stress. However, burnout may represent a diﬀ erent kind of health risk for men and women, and this possibility needs to be taken into account in the promotion of occupational health.

9 TIIVISTELMÄ

Tässä väitöskirjassa tarkasteltiin työuupumusta työterveyspsykologian näkökulmasta. Työuupumuksella tarkoitetaan uupumusasteisen väsy- myksen, kyynistyneisyyden ja alentuneen ammatillisen itsetunnon muo- dostamaa oireyhtymää, joka voi kehittyä pitkään jatkuneen työstressin seurauksena. Tutkimuksen päätavoitteena oli selvittää työuupumuksen samanaikaista esiintymistä mielenterveyden häiriöiden, tuki- ja liikunta- elinsairauksien ja sydän- ja verenkiertoelinten sairauksien kanssa, koska nämä sairausryhmät ovat Suomessa yleisimmin työkyvyttömyyseläkkeen perusteena. Lisäksi tutkittiin, onko työuupumus yhteydessä työkuormi- tukseen samalla tavalla kuin masennushäiriöt ja onko työuupumuksella yhteyttä pitkiin sairauspoissaoloihin. Myös sosiodemograﬁ sten taus- tatekijöiden yhteyksiä työuupumukseen kartoitettiin työssäkäyvässä väestössä. Tutkimuksessa käytettiin kansallisesti edustavaa työikäisistä (30–64- vuotiaista) suomalaisista (n = 3151–3424) muodostettua Terveys 2000 -aineistoa. Työuupumusta mitattiin Maslachin yleisellä työuupumuksen arviointimenetelmällä (MBI-GS). Mielenterveyden häiriöiden diagnoosit perustuivat standardoituun mielenterveyshaastatteluun (Composite In- ternational Diagnostic Interview) ja somaattisten sairauksien diagnoosit lääkärintarkastukseen. Tiedot lääkärin määräämistä yli 9 päivää kestä- neistä sairauspoissaoloista poimittiin Kansaneläkelaitoksen rekisteristä. Työkuormitus määriteltiin työn vaatimusten ja hallintamahdollisuuksien avulla. Sosiodemograﬁ sina taustatekijöinä käytettiin sukupuolta, ikää, siviilisäätyä, koulutusta ja ammattiasemaa. Työuupumus oli yhteydessä masennus- ja ahdistushäiriöihin sekä alkoholiriippuvuuteen miehillä ja naisilla. Lisäksi se oli yhteydessä tuki- ja liikuntaelinsairauksiin naisilla ja sydän- ja verenkiertoelinten

10 TIIVISTELMÄ

sairauksiin miehillä näiden sairausryhmien tavallisten riskitekijöiden ohella. Vakava-asteinen työuupumus oli itsenäisesti yhteydessä pitkiin sairauspoissaoloihin, vaikka samanaikaisesti esiintyneet työssäkäyvillä yleiset mielenterveyden häiriöt ja somaattiset sairaudet otettiin huomioon. Miehillä, joilla oli ollut vähintään yksi pitkä sairauspoissaolojak- so, vakavaan työuupumukseen liittyi enemmän poissaolopäiviä kahden vuoden aikana kuin vastaavassa tilanteessa olleilla naisilla. Naisilla myös lievä työuupumus oli yhteydessä pitkiin sairauspoissaoloihin. Kuormittavaa työtä tekevillä oli useammin työuupumusta tai masen- nushäiriö kuin vähän kuormittavaa työtä tekevillä. Merkitsevä yhteys työkuormituksen ja työuupumuksen välillä säilyi, vaikka työntekijöiden masennusoireilu ja mielenterveyden häiriöt otettiin huomioon. Sen sijaan työkuormituksen ja masennushäiriön välillä ei ollut suoraa yhteyttä, jos samanaikainen työuupumus huomioitiin. Työuupumus oli 30 vuotta täyttäneiden työssäkäyvien joukossa yleisempää ikääntyneillä työntekijöillä nuorempiin verrattuna. Lisäksi vaikutti siltä, että työuupumus olisi muihin verrattuna hieman yleisempää erityisesti vähän koulutetuilla ja matalassa ammattiasemassa työskente- levillä naisilla sekä naimattomilla miehillä. Vaikka työuupumus oli selvästi yhteydessä mielenterveyden häiriöihin ja somaattisiin sairauksiin, se ei ollut aivan samalla tavalla yhteydessä työkuormitukseen eikä työkyvyttömyyteen kuin nämä sairaudet. Työ- uupumus näyttää liittyvän vahvemmin työkuormitukseen kuin masen- nushäiriöt ja vaikuttaa olevan samanaikaisista yleisistä sairauksista huo- limatta itsenäinen riskitekijä pitkille sairauspoissaoloille. Työuupumusta voitaisiinkin pitää terveyttä vaarantavan työstressin hälytysmerkkinä. Se näyttää muodostavan osin erilaisen riskin miesten ja naisten terveydelle, mikä on tarpeen ottaa huomioon työterveyttä edistettäessä.

11 LIST OF ORIGINAL PUBLICATIONS

Th is review is based on the following original publications, which are referred to with Roman numerals I–VI. Th e original articles have been re-published in this report with the permission of Springer-Verlag Ger- many (I), Lippincott, Williams & Wilkins (II), Elsevier (III, V, VI), and Blackwell Publishing (IV).

I. Ahola, K., Honkonen, T., Isometsä, E., Kalimo, R., Nykyri, E., Koskinen, S., Aromaa, A. & Lönnqvist, J. (2006) Burnout in the general population. Results from the Finnish Health 2000 Study. Social Psychiatry and Psychiatric Epidemiology 41, 11–17.

II. Ahola, K., Honkonen, T., Kivimäki, M., Virtanen, M., Isometsä, E., Aromaa, A. & Lönnqvist, J. (2006) Contribution of burnout to the association between job strain and depression: the Health 2000 Study. Journal of Occupational and Environmental Medicine 48, 1023–1030.

III. Ahola, K., Honkonen, T., Isometsä, E., Kalimo, R., Nykyri, E., Aromaa, A. & Lönnqvist, J. (2005) Th e relationship between job- related burnout and depressive disorders – results from the Finnish Health 2000 Study. Journal of Aﬀ ective Disorders 88, 55–62.

IV. Ahola, K., Honkonen, T., Pirkola, S., Isometsä, E., Kalimo, R., Nykyri, E., Aromaa, A. & Lönnqvist, J. (2006) Alcohol dependence in relation to burnout among the Finnish working population. Addiction 101, 1438–1443.

12 LIST OF ORIGINAL PUBLICATIONS

V. Honkonen, T., Ahola, K., Pertovaara, M., Isometsä, E., Kalimo, R., Nykyri, E., Aromaa, A. & Lönnqvist, J. (2006) Th e association between burnout and physical illness in the general population – results from the Finnish Health 2000 Study. Journal of Psycho- somatic Research 61, 59–66.

VI. Ahola, K., Kivimäki, M., Honkonen, T., Virtanen, M., Koskinen, S., Vahtera, J. & Lönnqvist, J. (in press) Occupational burnout and medically certiﬁ ed sickness absence: the population-based sample of Finnish employees. Journal of Psychosomatic Research.

In addition, some unpublished data have been included in this thesis.

13 1. INTRODUCTION

Th e psychosocial characteristics of work have changed during the past few decades in response to trends in the global economy and increasing demands in worklife (Maslach & Leiter 1997, Allvin & Aronson 2001, Landbergis 2003, Allvin et al. 2006, Sennett 2006). Although both positive and negative changes have occurred in the quality of work (Lehto & Sutela 2005, Green 2006), concerns have been raised. For example, it has been argued that competitive pressure in the manufacturing industry, growing demands of consumers in the service industry, and the fast development of the high technology industry increasingly tax employees' resources via demands for excessive work hours and fl exibility, the responsibility for work processes, the need for continuous learning and re-orientation, job insecurity, and the blurring of the line separ- ating work and private life (Allvin & Aronson 2001, Sparks et al. 2001, Shirom 2003). In the context of occupational health, the problems of individual employees related to work stress are often encountered as burnout. One of the Oxford English language dictionaries (Hornby 1982) defi nes burnout as "to ruin one's health by overwork". Scientifi c interest in occupational burnout started approximately three decades ago, after Herbert Freudenberger (1974), a psychiatrist, described a negative phe- nomenon among dedicated volunteers working in a clinic for drug ad- dicts. "Burn-out" meant that a staff member became gradually exhausted from excessive demands on energy, strength, or resources about a year after he or she began work and showed various physical, behavioural, and mental symptoms. At the same time, Christina Maslach (1976), a researcher in social psychology, reported how professionals in health and social services can lose all their emotional feelings and concern for their

14 1. INTRODUCTION

clients after months of listening to their problems and how this change correlates with other damaging indices of stress, for example, alcoholism, mental illness, and suicide. Th ese observations on burnout were made in human service work in which contacts with other people constitute the major part of the task and can become a source of stress (Maslach & Jackson 1981, Zapf 2002, Schaufeli 2006). Later it was discovered that burnout can evolve in a wide range of occupations (Leiter & Schaufeli 1996, Demerouti et al. 2001, Toppinen-Tanner et al. 2002) and can be conceptualized generally as a crisis in one's relationship with work instead of with clients (Schaufeli et al. 1996). Th e fact that occupational burnout tends to show considerable sta- bility over time indicates the chronic nature of the problem (McKnight & Glass 1995, Bakker et al. 2000b, Taris et al. 2005). Th e estimated prevalence of severe burnout has been around 6–7% in Finnish and Swedish working populations (Kalimo & Toppinen 1997, Hallsten et al. 2002). However, because burnout has mainly been studied in non- random vocational and organizational samples, population-based evidence on the distribution of burnout by socio-demographic factors is scarce. Compared with the large quantity of research on the developmental process of burnout, much less is known of the consequences of burnout for individuals and society. Observed correlations between burnout and depressive symptoms have raised questions about the conceptual redundancy of burnout and depression, but it is unclear to what extent burnout is related to mental disorders and physical illnesses and whether it has any independent status in relation to ill health.

1.1 Concept of occupational burnout

Occupational burnout refers to a negative consequence of chronic work stress (Shirom 1989, Schaufeli & Enzmann 1998, Maslach et al. 2001). Th ere are several theoretical models on the origins of burnout, ranging from individual and interpersonal explanations to organizational and societal approaches. Many of these models share a basic assumption of some kind of a chronic discrepancy between the expectations of a moti- vated employee and the reality in unfavourable work conditions, which develops towards burnout via dysfunctional ways of coping (Schaufeli &

15 1. INTRODUCTION

Enzmann 1998). Regarding unfavourable work conditions, occupational burnout has been presented to result especially from a combination of high demands and low resources at work (Demerouti et al. 2001, Schaufeli & Bakker 2004). Job demands refer to the physical, psychological, social, and organizational aspects of the job that require sustained eff ort and are therefore associated with physiological and psychological costs for the individual. Job resources refer to those corresponding aspects of the job that are functional in achieving work goals, reducing the associated costs, and stimulating learning and professional development. It has been proposed that individual factors concerning personality, for example, alexithymia, low sense of coherence, un-hardiness, and neu- roticism, infl uence vulnerability to burnout (Nowack 1986, McCranie & Brandsma 1988, Zellars et al. 2000, Kalimo et al. 2003, Mattila et al. 2007), which may then interact with the situational factors that are conductive to the development of burnout (Maslach & Leiter 1997, Schaufeli & Enzmann 1998, Shirom 2003, Hakanen 2004). Based on the utilization of the concept in peer-reviewed publications, burnout is generally (e.g. Schaufeli & Enzmann 1998) defi ned as a three-dimensional psychological syndrome of emotional exhaustion, depersonalization, and diminished personal accomplishment that may occur in human service work (Maslach & Jackson 1996). In a general form, not restricted to any particular kind of work, the dimensions of burnout are labelled exhaustion, cynicism, and diminished professional effi cacy (Schaufeli et al. 1996). Exhaustion is generic and refers to feelings of overstrain, tiredness, and fatigue. Cynicism (depersonalization) refl ects an indiff erent and distant attitude towards work (clients), disengagement from it, and a lack of enthusiasm. Professional effi cacy (personal accomplishment) consists of feelings of competence, successful achievement, and accomplishment in one's work, which are assumed to diminish as burnout develops. Exhaustion is considered to result from long-term involvement in an over-demanding work situation (Shirom 1989, Maslach & Jackson 1996, Schaufeli et al. 1996) and to lead to changes in attitudes towards work and oneself as a worker. Cynicism develops as a dysfunctional way of coping in exhausting situations, which further reduce the possibilities to fi nd creative solutions at work and build professional effi cacy (Leiter & Schaufeli 1996, Schaufeli et al. 1996). Th is developmental sequence

16 1. INTRODUCTION

between the dimensions of burnout has gained the most consistent support (Leiter & Maslach 1988, Bakker et al. 2000b, Toppinen-Tanner et al. 2002, Taris et al. 2005), although other sequences have also been presented (Golembiewski et al. 1983, 1996, Lee & Ashforth 1993, Bakker et al. 2000b, Demerouti et al. 2001, van Dierendonck et al. 2001). Some researchers have proposed that exhaustion and cynicism would be the primary dimensions of burnout, while diminished professional effi cacy would be a separate but related entity (Leiter 1993, Lee & Ashforth 1993, Halbesleben & Demerouti 2005). Supporting this suggestion, professional effi cacy has loaded on a diff erent factor than exhaustion and cynicism in structural equation modelling (Schaufeli & Bakker 2004). However, this occurrence could also refl ect a statistical artefact because, in contrast to the positively worded items of the exhaustion and cynicism dimensions, the items of professional effi cacy are worded negatively (Schaufeli & Taris 2005). Because the associations between the three burnout dimensions and between the dimensions and the characteristics of employees and work have been found to be complex, the Maslach Burnout Inventory Manual recommends the use of separate scores for the dimensions rather than the construction of a single burnout score (Maslach & Jackson 1996), a recommendation also supported by statistical arguments (Taris et al. 1999). On the other hand, because burnout is, by defi nition, a single construct, a specifi c work-related syndrome, the use of a single score has been considered appropriate when the main interest is in the burnout syndrome and when the study design is complex (Taris et al. 1999, Bren- ninkmeijer & van Yperen 2003). Single scores for burnout have been constructed relative to the distribution of the dimensional scores in a population (Schaufeli et al. 2001, Brenninkmeijer & van Yperen 2003, Roelofs et al. 2005) or as a sum score of the dimensional scores (Buunk et al. 2001, Kalimo et al. 2006). Th e phase model for burnout (Golem- biewski et al. 1983, 1996) can also be regarded as means of forming a single burnout score. In this model, the burnout process is divided into eight phases that are defi ned in terms of the diff erent combinations of the dimensional scores above or below the median. Th e dimensional defi nition of burnout, which is based on observations on burnt out employees, has been criticized especially for lacking theoretical arguments for grouping diff erent concepts together in a

17 1. INTRODUCTION

cluster of heterogeneous symptoms (Shirom 2003, Kristensen et al. 2005). Th erefore, in the alternative defi nitions of burnout, the exhaustion dimension alone is considered to constitute the burnout syndrome (Shirom 1989, Hallsten 1993, Pines 1993, Kristensen et al. 2005). Burnout is defi ned as a state of physical fatigue and mental exhaustion (Pines 1993, Kristensen et al. 2005), and of cognitive weariness (Shirom 1989). Exhaustion has also been complemented with performance-based self-esteem to take into account another line of criticism concerning the inclusion of persons without high initial motivation (Hallsten et al. 2005). In these alternative defi nitions, exhaustion is either considered generic or is attributed to specifi c domains, for example, the work context. At present, there is no universal consensus on the defi nition of burnout (Cox et al. 2005). In this dissertation, the term “burnout” is used to refer to the dimensional burnout concept, and the term “exhaustion” refers to the exhaustion-based burnout concept.

1.2 Socio-demographic factors and occupational burnout

Socio-demographic associations can be the most reliably studied in representative population-based samples that are free from any selection in the sample formation. However, population studies on burnout are scarce. One Swedish study (Lindblom et al. 2006) and one Finnish study (Kalimo & Toppinen 1997, Kalimo 2000) have examined burnout, and another Swedish study has examined exhaustion (Hallsten et al. 2002). Th e main ﬁ ndings from these studies are summarized in Table 1. Burnout and exhaustion were commoner among women in all three population studies than among men (Kalimo & Toppinen 1997, Hallsten et al. 2002, Lindblom et al. 2006). Among the working population of one Swedish county, employees with a high level of burnout were older, more often over 50 years of age, than those with a moderate level of burnout but not older than those with a low level of burnout (Lindblom et al. 2006). Among the Finnish working population, the level of burnout was higher among employees over 55 years of age than among the younger employees (Kalimo & Toppinen 1997, Kalimo 2000). However, in an earlier study among the Swedish working population, an opposite trend

18 1. INTRODUCTION A high level of burnout was commoner among the women and the over-50- employees than year-old among the other groups. Marital status and educa- to not related tion were burnout. Main results Exhaustion was commoner among the women, employees younger than 50 years of age, upper grade non-manual workers, and the married versus other groups. Burnout was commoner among the women and those over 55 years of age. burnout Severe was commonest among those with little vocational education. syndrome (MBI-GS) syndrome measure achievement-based self-esteem (BM and a self-construed scale) syndrome (MBI-GS) syndrome 1812 Three-dimensional 3502 Exhaustion with 2300 Three-dimensional Working Working population of one county Sample N Burnout Working population Working Working population survey (61%) Study design rate) (response survey (71%) survey (66%) Sweden Cross-sectional Country Sweden Cross-sectional Finland Cross-sectional , a a Kalimo & 1997 Toppinen Lindblom et al. 2006 Hallsten et al. 2002 Table 1. Review of the population-based studies on socio-demographic factors and burnout Table Authors and date Kalimo 2000 not peer-reviewed, MBI-GS = Maslach Burnout Inventory - General Survey, BM = Burnout Measure MBI-GS = Maslach Burnout Inventory - General Survey, not peer-reviewed, a

19 1. INTRODUCTION

was detected for exhaustion when it was combined with performance- based self-esteem; it was lower among employees older than 50 years of age than among those younger (Hallsten et al. 2002). Th e level of exhaustion has been found to be higher among upper-grade non-manual workers (Hallsten et al. 2002), while the results regarding the association between burnout and education were mixed (Kalimo & Toppinen 1997, Lindblom et al. 2006). Results on the relationship between burnout and marital status are not consistent either. A recent Swedish population-based study failed to establish an association between marital status and burnout (Lindblom et al. 2006), although, in a previous study, exhaustion was found to be higher among those not married than among those married when it was combined with performance-based self-esteem (Hallsten et al. 2002). Most of the knowledge concerning the relative prevalence of burnout in various population subgroups is based on indirect evidence, because very few studies have been primarily set up to study the relationship between socio-demographic factors and burnout. Usually these associations have been reported as a part of the sample description and then adjusted for. In the few published representative studies, burnout was related to gender, age, education, and occupational and marital status (Table 1), but the diﬀ erences according to these factors were small and, to some extent, inconsistent. Th e only peer-reviewed population study (Lindblom et al. 2006) was not nationally representative and did not include gender-stratiﬁ ed analyses on the associations of burnout.

1.3 Job strain and occupational burnout

Th e job strain model is probably the most widely used conceptualization of health-impairing work stress. It has successfully predicted cardiovascular diseases, mental disorders, and musculoskeletal problems among employees (Karasek 1979, Karasek & Th eorell 1990, Bongers et al. 1993, Th eorell & Karasek 1996, van der Doef & Maes 1999, Ariëns et al. 2001, Feveile et al. 2002, de Lange et al. 2003, Belkic et al. 2004, Kivimäki et al. 2006b, Standsfeld & Candy 2006). Th e job strain model characterizes jobs according to two main aspects, the psychological demands of the work situation and the control

20 1. INTRODUCTION

available to the worker facing these demands in making decisions on his or her work activity and to use his or her skills (Karasek 1979). Th e combination of high demands and low control, referred to as high job strain, is hypothesized to produce the highest risk of health impairment, while active work involving high demands and high control is proposed to promote learning, growth, and satisfaction and to predict only an average amount of psychological strain (Karasek & Th eorell 1990). Th e job strain model of work stress resembles the work demands and resources model of burnout (Schaufeli & Bakker 2004), which includes a general representation of demands and resources at work. Job strain has been found to be associated with burnout and exhaustion (Table 2). High job demands and low control showed a positive interaction eff ect on burnout among German software professionals (Sonnentag et al. 1994) and on exhaustion among Dutch human service workers (de Jonge et al. 2000). In these studies, job control moderated the associations between job demands and burnout or exhaustion, as originally proposed by Karasek (1979). High job strain compared with low job strain was positively associated with exhaustion in a cross-sectional population-based study among Finnish wage earners (Kauppinen- Toropainen et al. 1983) and in several human service samples (Raff erty 1987, Landsbergis 1988, Melamed et al. 1991, Bourbonnais et al. 1998, 1999). Active and passive work has also to some extent been shown to be associated with an increased level of exhaustion (Kauppinen-Toropainen et al. 1983, Landsbergis 1988, Bourbonnais et al. 1998, 1999). In addition, most studies that have explored the correlates of burnout have supported independent positive associations between high work load and low autonomy at work on one hand and a high level of burnout (Lee & Ashforth 1996, Schreurs & Taris 1998, Kalimo et al. 2003, Schaufeli & Bakker 2004, Lindblom et al. 2006) or exhaustion (Borritz et al. 2005) on the other. Similar associations of job strain with burnout and exhaustion and with problems of mental health have raised the question of whether burnout is conceptually redundant with respect to low mental health and especially to depression. As a diff erence between the two, it has been proposed that burnout would be work-related, whereas depression is expected to be context-free or multi-factorial in origin (Warr 1987, Maslach et al. 2001). Th is assumption has not, however, been

21 1. INTRODUCTION c ed by control ed by control Main results with High job strain compared low job strain was positively associated with emotional exhaustion. The joint effect of time pressure of time pressure The joint effect and low self-determination was work-related to increased related exhaustion. Passive jobs were to exhaustion among related women. Emotional exhaustion was higher in high strain work and passive work than in low strain work. was found under conditions of low demands, high perceived and high social support. control, The effect of stressors at work on of stressors The effect burnout was modifi at work. High job strain, active work, and positively asso- passive work were ciated with emotional exhaustion. High job strain, active work, and positively asso- passive work were ciated with emotional exhaustion. Focussed job demands and showed an interaction control on exhaustion in specifi effect occupational groups. measure (MBI) syndrome exhaustion (self-construed scale) syndrome (MBI) syndrome syndrome syndrome (self-construed scale) (MBI) (MBI) (MBI) 188 Three-dimensional 289 Tthree-dimensional 267 Exhaustion (BM) The lowest level of exhaustion 180 Two-dimensional 5471 Work-related 1891 Emotional exhaustion 1378 Emotional exhaustion 2485 Emotional exhaustion ve Sample N Burnout Wage earning Wage population Human service workers Nursing staff in Nursing staff several units Female social workers in one organization Software Software professionals in several organizations Nurses in one district Nurses in one district Human service workers in fi sectors (response rate) (response sectional survey sectional survey (93%) sectional survey (38%) sectional survey (76%) sectional survey (90%) sectional survey (62%) sectional survey (79%) Cross- sectional survey Country Study design USA Cross- Finland Cross- USA Cross- Israel Cross- Germany Cross- Canada Cross- Canada Cross- Nether- lands a Table 2. Review of the literature on job strain and burnout 2. Review of the literature Table Authors and date Kauppinen-To- et al. ropainen 1983 Rafferty 1987 Rafferty Landsbergis 1988 Melamed et al. 1991 Sonnentag et al. 1994 Bourbonnais et al. 1998 Bourbonnais et al. 1999 de Jonge et al. 2000 not peer-reviewed; MBI = Maslach Burnout Inventory; BM = Burnout Measure not peer-reviewed; a

22 1. INTRODUCTION

defi nitely confi rmed. Only one study has explored burnout and depressive symptoms in relation to diff erent contexts (Bakker et al. 2000a). In this cross-sectional study among Dutch teachers, a lack of reciprocity in work context was associated with a high level of burnout, whereas a lack of reciprocity in private life was associated with depressive symptoms indicating support for the work-relatedness of burnout and the absence of such relatedness concerning depression. However, the evidence was based on a rather small sample of one vocational group, and it covered only a single aspect of the psychosocial work environment, which is not included in the leading model of work-related health (Karasek & Th eorell 1990). Th e associations between job strain, burnout, and depression have not been previously analysed in the same study.

1.4 Occupational burnout and health problems

Th ere is strong evidence indicating that continuous work stress, when operationalized by demanding psychosocial work characteristics and especially by job strain, is detrimental to health (Leino & Hänninen 1995, National Research Council and the Institute of Medicine 2001, Hagen et al. 2002, Kivimäki et al. 2006b, Stansfeldt & Candy 2006). Th erefore, it is highly plausible that occupational burnout, defi ned as a consequence of chronic work stress, would also be associated with ill health. Th ere are several conceivable mechanisms to account for the adverse health eff ects of long-term work stress (McEwen & Stellar 1993, Maier & Watkins 1998, Kiecolt-Glaser et al. 2002, Siegrist 2002, Epel et al. 2004, 2006, Chandola et al. 2006). Th e autonomic nervous system and the hypothalamus-pituitary-adrenal cortex axis, which are partly responsible for an individual's potential to adapt to stressful challenges, can, if overly activated, also result in harmful allostatic load (Frankenhauser 1989, Brunner 1997, McEwen 1998b). Adverse stress processes may become pronounced in some persons because innate temperamental aspects of personality (Cloninger et al. 1993) have been shown to aff ect a person's emotional experiences and physiological activation in challenging situations (Keltikangas-Järvinen & Heponiemi 2004, Ravaja et al. 2006, Puttonen et al. 2005, Tyrka et al. 2006). Th ese kinds of mechanisms

23 1. INTRODUCTION

may also account for the possible associations between burnout and ill health (Shirom et al. 2005, Toker et al. 2005, Melamed et al. 2006). In addition to direct eff ects, work stress, as well as burnout, may also infl uence health through health behaviour (i.e., health-related habits and behavioural decisions of employees) (Siegrist & Rödel 2006, Kouvonen et al. 2007). However, the connection between burnout and health could also operate in a reverse direction. If suff ering from a disease results in the lowering of work ability, it could also hinder the attainment of the work goals, which, in turn, could activate the process of burnout (Hallman et al. 2003, de Lange et al. 2005, Donders et al. 2007). Among the Finnish work force, the three most prominent disease groups related to work disability are mental disorders, musculoskeletal disorders, and cardiovascular diseases (Th e Social Insurance Institution of Finland 2005). In 2000, these diseases accounted for 75% of all causes of disability pensions and for 62% of the causes of compensated sickness absences (Th e Social Insurance Institution of Finland 2000). Together these diseases cover 43% of the global disease burden in Euro- pean countries, indicated by disability-adjusted life years (World Health Organization 2002).

Mental health problems Psychiatric classifi cations are categorical systems that divide mental disorders into types based on sets of descriptive criteria. Th e classifi cations currently in use are the International Statistical Classifi cation of Diseases and Related Health Problems (ICD-10) (World Health Organization 1992) and the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (American Psychiatric Association 1994). Among the work force, depressive disorders, alcohol use disorders, and anxiety disorders are common mental disorders and relevant in terms of work disability (Raitasalo & Maaniemi 2006, Sanderson & Andrews 2006, Honkonen et al. 2007, Raitasalo & Maaniemi 2007). In the classifi cation systems, burnout can be coded as a factor that infl uences health status (Z73.0 burnout) (World Health Organization 1992) or as a condition that may require clinical attention (V62.2 occupational problem) (American Psychiatric Association 1994).

24 1. INTRODUCTION

Summaries of the existing publications on burnout and mental health problems are presented in Table 3 (depression), Table 4 (anxiety), and Table 5 (alcohol problems) in chronological order. Most of the research conducted on the relationship between burnout and mental health problems concerns depression. Only two studies have investigated depressive disorders, but they did not use structured psychiatric diagnostic methods. Among Finnish participants in occupational rehabilitation, a depressive disorder, based on a medical examination in the beginning of the rehabilitation programme, was almost always (90%) present among those who suﬀ ered from severe burnout (Kinnunen et al. 2004). Self-reported physician-diagnosed depression was more prevalent also among burnt out teachers than among their colleagues without burnout in a cross-sectional North American survey (Belcastro & Hayes 1984). Most of these teachers with depression reported that they had developed depression after becoming a teacher. Four studies have investigated burnout and depressive symptoms in a longitudinal design. In a 1-year follow-up among Canadian teachers, high scores on the exhaustion and depersonalization dimensions of burnout predicted depressive symptoms among women, while high scores on exhaustion and diminished personal accomplishment predicted depressive symptoms among men (Greenglass & Burke 1990). Among North American medical residents, patient-related but not job-related exhaustion predicted mood disturbance during 1-year of follow-up (Hillhouse et al. 2000). On the other hand, depressive symptoms at baseline correlated positively with exhaustion 20 years later among North American physicians (McCranie & Brandsma 1988). However, a 2-year longitudinal study among North American nurses failed to provide support for a temporal relation between burnout and depression (McKnight & Glass 1995). In cross-sectional designs, burnout and depressive symptoms have been found to correlate positively in various study samples: among the Swedish working population (Lindblom et al. 2006), in human service work (Meier 1984, Firth et al. 1986, Landsbergis 1988, Seidman & Zager 1991, Glass et al. 1993, McKnight & Glass 1995, Baba et al. 1999, Sears et al. 2000, Korkeila et al. 2003), and among patients (Roe- lofs et al. 2005). Corresponding correlations were found also between exhaustion and depressive symptoms in human service work (Firth et

25 1. INTRODUCTION

al. 1987, Jayaratne et al. 1986, Lemkau et al. 1988, Martin et al. 1997, Tselebnis et al. 2001) and heterogeneous occupations (Toker et al. 2005, Middeldorp et al. 2006). Among Greek hospital staff , the lower the scores on personal accomplishment, the closer the relationship between the attitudes towards work and the emotional tone of the participants (Iacovides et al. 1999). Despite a signifi cant positive association between burnout and depressive symptoms, burnout and depression have been statistically diff erenti- ated from each other. Confi rmatory factor analyses on cross-sectional data from four diff erent studies on human service work showed that the items of a burnout inventory and a depression scale did not load on the same factor. Instead, a model with two second-order factors of burnout and depression was preferred over one second-order factor of negative aff ectivity (Glass et al. 1993, Leiter & Durup 1994, Iacovides et al. 1999, Bakker et al. 2000a). When the measures of burnout and depression were completed with a self-constructed scale of professional depression, many items of the work-related scales (emotional exhaustion and professional depression) loaded on the same factor in contrast to the non-contextual scale of depressive symptoms (Firth et al. 1987). Diff erences have also have been noted in the associations between burnout and depression. In a cross-sectional study among Dutch teachers, a reduced sense of superiority was more characteristic of depression than of burnout (Brenninkmeijer et al. 2001); depressive symptoms were related to burnout when the teachers experienced low superiority. In another cross-sectional Dutch study among teachers, burnout was related to a lack of reciprocity with students, while depression was related to a lack of reciprocity with one's spouse (Bakker et al. 2000a). In addition, burnout and depression were related diff erently to some demographic and work characteristics in a random sample of French health care professionals (Martin et al. 1997). Burnout was commoner among women with unconventional work hours and employees who were young or had instrumental work motivation, while depression was positively related to seniority at work and a lack of social support. Diff erences were also found between exhaustion and depression at the neurobiological level; they related diff erently to infl ammation biomarkers among Israeli employees and the relationships were dependent on gender (Toker et al. 2005). Among women, a high level of exhaustion, unlike depression,

26 1. INTRODUCTION ed Beck Depression ed Beck Depression Main results with burnout more Teachers physician-diag- often reported than teachers nosed depression with no burnout. Burnout positively correlated (Costello-Com- with depression Scale, MMPI, Depression rey self-rating). The burnout dimensions correlated positively with depression (BDI). Employees high on emotional exhaustion and low on personal accomplishment had higher (6 items). levels of depression Several items of exhaustion and scales depression professional (modiﬁ Inventory) loaded on the same Exhaustion correlated factor. positively with professional depression. The burnout dimensions correlated positively with depression (Job Content Survey). Emotional exhaustion correlated positively with psychotic depression (Millon Clinical Multiaxial Inventory). (MMPI at time Depression positively with 1) correlated exhaustion (time 2) over two decades. ed MBI) measure syndrome (MBI) syndrome (MBI) syndrome (MBI) syndrome (modi- syndrome ﬁ tion (MBI) (MBI) syndrome (MBI) syndrome exhaustion (Tedium scale) 265 Three-dimensional 320 dimensional Three 200 Three-dimensional 75 Three-dimensional 200 Emotional exhaus- 289 Three-dimensional 67 Three-dimensional 440 of Syndrome Sample N Burnout Teachers in one Teachers district faculty Teaching members of one university Nurses in one district Female welfare workers in Nursing staff several hospitals in Nursing staff several units Medical residents Physicians from eight classes in one medical college (response rate) (response survey (58%) survey (61%) survey (41%) survey (85%) survey (40%) survey (38%) survey (94%) nal survey (72% at time 2) Country Study design USA Cross-sectional USA Cross-sectional USA Cross-sectional UK Cross-sectional USA Cross-sectional USA Cross-sectional USA 20-year longitudi- Authors and date Belcatsro & Belcatsro Hays 1984 Meier 1984 USA Cross-sectional Firth et al. 1986 Jayaratne et al. 1986 Firth et al. 1987 Landsbergis 1988 Lemkau et al. 1988 McCranie & Brandsma 1988 Table 3. continues... Table Table 3. Review of the literature on burnout 3. Review of the literature and depression Table

27 1. INTRODUCTION Emotional exhaustion and depersonalization predicted (Hopkins Symptom depression Checklist) among women. Emo- tional exhaustion and diminished personal accomplishment among depression predicted men. Burnout subscales correlated (self- positively with depression report). The burnout dimensions correlated positively with depression (BDI). The items of the burnout scales loaded on and depression factors. different A model with two primary factors (burnout and depression; over a model BDI) was preferred with one primary factor (negative affectivity). The burnout dimensions correlated positively with depression (BDI) at time 1 and 2. The exhaustion and depression changed concurrently. scores Emotional exhaustion and de- to related personalization were (CES-D). Burnout depression and to differently related depression the demographic factors and work characteristics. Burnout positively correlated to depres- with and was related sion (CES-D). syndrome (MBI) syndrome sponding (Teacher sponding (Teacher Burnout Scale) (MBI) syndrome (MBI) syndrome (MBI) syndrome (MBI) syndrome (MBI) syndrome 361 Three-dimensional 365 Pattern of re- 162 Three-dimensional 307 Three-dimensional 536 Three-dimensional 119 Three-dimensional Teachers and Teachers principals on one school board Teachers in one Teachers district Random sample of nurses in one hospital Health care workers in one hospital Nurses 100Random sample of health care Three-dimensional in professionals one hospital Nurses from several units nal survey (57% at time 2) survey survey (28%) survey nal survey (85% at time 1, attrition 20%) survey (86%) Cross-sectional survey (48%) Canada 1-year longitudi- USA Cross-sectional USA Cross-sectional Canada Cross-sectional USA 2-year longitudi- France Cross-sectional Carib- bean Greenglass Greenglass & Burke 1990 Seidman & Zager 1991 Glass et al. 1993 Leiter & Durup 1994 McKnight & Glass 1995 Martin et al. 1997 Baba et al. 1999 Table 3. continues... Table Table 3. continues... Table

28 1. INTRODUCTION Not all of the employees with had the burnout syndrome (Zung Self-Rating depression Scale). Items of Depression the burnout and depression The separately. scales clustered lower the personal accomplish- the symptoms ment, the more depression. resembled in the A lack of reciprocity with students was relationship to burnout related but not (CES-D). A lack to depression in the relation- of reciprocity ship with one's spouse was but not to to depression related burnout. exhaustion Patient-related mood at time 3 predicted disturbance (POMS) at time 4. exhaustion was not Job-related to mood disturbance in related time. The burnout dimensions correlated positively with depression (CES-D). sense of superiority A reduced characteristic for de- was more (CES-D) than for burn- pression to was related out. Depression burnout when low superiority was experienced. Emotional exhaustion and diminished personal accomplish- positively with ment correlated (BDI). depression ed three-di- syndrome (MBI) syndrome haustion (SBS-HP) (MBI) syndrome (MBI) syndrome mensional syndrome mensional syndrome (MBI) (MBI-ES) syndrome 154 Three-dimensional 46 of ex- Syndrome 264 Three-dimensional 79 Three-dimensional 368 Modiﬁ Teachers in sev- Teachers eral schools Medical residents Extension agents in one organization Teachers 190 Three-dimensional Nurses in one hospital Nursing staff in Nursing staff one hospital Cross-sectional Cross-sectional survey (83%) nal survey (85%) survey (88%) Cross-sectional survey survey survey (100%) Nether- lands USA 1-year longitudi- USA Cross-sectional Nether- lands Greece Cross-sectional Greece Cross-sectional Bakker et al. 2000a Hillhouse et al. 2000 Sears et al. 2000 Brennink- meijer et al. 2001 Tselebis et al. 2001 Iacovides et al. 1999 Table 3. continues... Table 3. continues... Table

29 1. INTRODUCTION ammation emiologic Studies c environmental factors. c environmental brinogen). Burnout positively correlated depression. with self-reported Depressive disorder was com- disorder Depressive moner among the rehabilitants burnout with severe than among the others. The burnout dimensions correlated positively with depression (Symptom Checklist SCL-90) among those with burnout and among those with no burnout. with Exhaustion correlated (Personal Health depression Exhaustion and Questionnaire). as- differently were depression sociated with the infl protein, biomarkers (C-reactive fi The burnout dimensions cor- positively with depres- related sion (the Hospital Anxiety and Scale). A high level Depression to de- of burnout was related pression. Exhaustion correlated positively Exhaustion correlated (Young with anxious depression Adult Self Report). The association between exhaustion and was explained by depression genetic and individual- shared specifi syndrome (MBI-GS) syndrome syndrome (MBI-GS) syndrome haustion (SMBM) syndrome (MBI-GS) syndrome Exhaustion (MBI- GS) syndrome (MBI) syndrome le of Mood States, SMBM = Shirom-Melamed Burnout le of Mood States, SMBM = Shirom-Melamed Measure 218 Three-dimensional 154 Three-dimensional 95+73 Three-dimensional 1563 of ex- Syndrome 1812 Three-dimensional 4309 +1008 Random sample of physicians in psychiatry Participants in occupational rehabilitation Consecutive patients with clinical burnout and with another diagnosis Employed clients of one medical centre Working popu- Working lation of one county Twins and their Twins siblings medical examination (94%) Cross-sectional Cross-sectional survey (100%) medical examination (91%) survey (61%) Cross-sectional Cross-sectional survey survey (74%) MBI = Maslach Burnout Inventory; MBI-GS = Maslach Burnout Inventory - General Survey; MBI-ES = Maslach Burnout Inventory Finland Cross-sectional Finland Cross-sectional Nether- lands Israel Cross-sectional Sweden Cross-sectional Nether- lands a Kinnunen et al. 2004 Korkeila et al. 2003 Roelofs et al. 2005 Toker et al. Toker 2005 Lindblom et al. 2006 Middel- dorp et al. 2006 not peer-reviewed; Table 3. continues... Table a Depression Scale, SBS-HP = Staff BurnoutPOMS = Profi Scale, SBS-HP = Staff Scale for Health Professionals, Depression - Educators Survey, MMPI= Minnesota Multiphasic Personality Inventory, BDI = Beck Depression Inventory, CES-D = Center for Epid Inventory, BDI = Beck Depression MMPI= Minnesota Multiphasic Personality Inventory, - Educators Survey,

30 1. INTRODUCTION -Melamed Main results Employees high on emotional exhaustion and low on personal accomplishment had a higher level of anxiety (4 items). Emotional exhaustion positively with correlated anxiety (Millon Clinical Multiaxial Inventory). The burnout dimensions to positively related were trait anxiety (STAI). Burnout correlated positively with trait anxiety (STAI). The burnout dimensions positively with correlated trait and state anxiety (STAI). The burnout dimensions positively with correlated anxiety (Symptom Check- list-90) among patients with and without burnout. Exhaustion correlated positively with anxiety (adapted questionnaire). The burnout dimensions positively with correlated anxiety (the Hospital Anxi- Scale). ety and Depression ed MBI) measure sional syndrome (modiﬁ sional syndrome (MBI) sional syndrome (MBI) sional syndrome (MBI) sional syndrome (MBI) sional syndrome (MBI-GS) haustion (SMBM) sional syndrome (MBI-GS) 47 Three-dimen- 212 Three-dimen- 117 Three-dimen- 1563 of ex- Syndrome 1812 Three-dimen- 95+73 Three-dimen- Sample workersFemale welfare 75 Three-dimen- Medical residents N Burnout workers in Health care one hospital 67 Three-dimen- in one Nursing staff hospital Female nurses in one hospital Consecutive patients with clinical burnout and with another diagnosis Employed clients of one medical centre population of Working one county (response rate) (response survey (85%) survey (94%) survey (28%) survey (83%) survey (85%) Cross-sectional survey (100%) medical examination (91%) survey (61%) USA Cross-sectional USA Cross-sectional Norway Cross-sectional USA Cross-sectional USA Cross-sectional Nether- lands Israel Cross-sectional Sweden Cross-sectional Country Study design Jayaratne et al. 1986 Lemkau et al. 1988 Richardsen et al. 1992 Corrigan et al. 1995 Turnipseed 1998 Roelofs et al. 2005 Toker et al. 2005 Lindblom et al. 2006 Table 4. Review of the literature on burnout 4. Review of the literature and anxiety Table Authors and date Burnout Measure MBI=Maslach Burnout Inventory; MBI-GS=Maslach Burnout Inventory-General Survey; STAI=State-Trait Anxiety Inventory; SMBM=Shirom MBI=Maslach Burnout Inventory; MBI-GS=Maslach Burnout Inventory-General Survey; STAI=State-Trait

31 1. INTRODUCTION c shoulder h risk for Patients with pain had a higher level of burnout than patients without pain. Burnout was the for correlate strongest non-speciﬁ pain. Main results Main results Exhaustion was associated with an elevated risk for alcohol dependence (CAGE). Work-related exhaus- Work-related positively tion correlated with the risk for alcohol dependence (AUDIT). of tension, anxiety, withdrawal and isola- tion (self- construed scale) mensional syndrome (MBI-GS) measure measure exhaustion (MBI) ed exhaustion (CBI) 3378 Three-di- 993 emotional 312 work-relat- 838+135 Syndrome 30 to 64-year-old 30 to 64-year-old working population SamplePatients with physician-diagnosed pain and without self- pain reported N Burnout Sample N Burnout Transit operators in Transit one organization Random sample of dentists cal examination (69%) (response rate) (response (85% and 68%) (response rate) (response cal examination (78%) (64%) Finland medi- Cross-sectional Sweden survey Cross-sectional Country Study design Country Study design USA medi- Cross-sectional Australia survey Cross-sectional Table 5. Review of the literature on burnout 5. Review of the literature and alcohol problems Table Authors and date instrument to identify those with a hig CAGE = A screening CBI = Copenhagen Burnout Inventory, MBI = Maslach Burnout Inventory, Indicator Test Eye-opener), AUDIT = Alcohol Use Disorder alcohol dependence (Cut-down, Annoyed, Guilty, Cunradi et al. 2003 Winwood et al. 2003 Miranda et al. 2005 Soares & Soares Jablonska 2004 Table 6. Review of the literature on burnout 6. Review of the literature and musculoskeletal problems Table Authors and date MBI-GS = Maslach Burnout Inventory-General Survey

32 1. INTRODUCTION

was associated with increased concentrations of C-reactive protein and ﬁ brinogen. Compared with depression, anxiety and alcohol problems have been studied much less in relation to burnout. In cross-sectional designs, signiﬁ cant positive correlations were found between burnout and state and trait anxiety among a working population (Lindblom et al. 2006), human service workers (Richardsen et al. 1992, Corrigan et al. 1995, Turnipseed 1998), and patients (Roelofs et al. 2005), and also between exhaustion and anxiety in human service work (Jayaratne et al. 1986, Lemkau et al. 1988) and heterogeneous occupations (Toker et al. 2005). Th ere was also cross-sectional evidence of a positive association between exhaustion and a risk for alcohol dependence in a random sample of Australian dentists (Winwood et al. 2003) and among North American transit operators (Cunradi et al. 2003).

Musculoskeletal problems Compared with mental health, physical health has been studied to a much less degree in relation to burnout. Two studies have explored the relationship between burnout and musculoskeletal pain (Table 6). In the cross-sectional population-based Finnish Health 2000 Study, a positive association was found between burnout and non-speciﬁ c shoulder pain (Miranda et al. 2005). Among Swedish pain patients, a positive association emerged between burnout and heterogeneous musculoskeletal pain (Soares & Jablonska 2004).

Cardiovascular diseases Th e three research publications on burnout and cardiovascular diseases are reviewed in Table 7. In a longitudinal setting among male employees who took part in a medical examination, self-reported burnout predicted physician-diagnosed myocardial infarctions in a 4-year follow-up (Appels & Shouten 1991). In a cross-sectional study among North American teachers, burnout was associated with self-reported physician-diagnosed cardiovascular disease (Belcastro 1982). None of the teachers reported having a cardiovascular disease before they had begun teaching. In another cross-sectional study, women with coronary heart disease reported

33 1. INTRODUCTION Main results High scores on the High scores burnout dimensions were to self-reported related physician-diagnosed cardiovascular disorders. Burnout was related to Burnout was related risk of myo- an increased infarction. cardial Women with CHD Women higher level of reported exhaustion than matched healthy controls. measure syndrome (MBI) syndrome (self-report, single- (self-report, item) construed index) 3877 Uni-dimensional 97+97 Exhaustion (self- Sample N Burnout Teachers 181 Three-dimensional Male employees aged 39-45 years or 54-65 years Consecutive female patients with CHD and healthy controls (response rate) (response survey (79%) 4-year longitudinal medical examination survey (100%) Country Study design USA Cross-sectional Nether- lands Sweden Cross-sectional Authors and date Belcastro Belcastro 1982 Appels & Shouten 1991 Hallman et al. 2003 Table 7. Review of the literature on burnoutdiseases 7. Review of the literature and cardiovascular Table disease CHD = cardiovascular MBI = Maslach Burnout Inventory,

34 1. INTRODUCTION

higher levels of exhaustion than their matched healthy controls (Hallman et al. 2003).

Summary and limitations of previous research All studies published on burnout and mental health problems support a positive association between burnout or exhaustion on one hand and depression, anxiety, or alcohol problems on the other. In addition to signifi cant positive correlations, qualitative diff erences have been found between burnout and depression. Burnout and depressive symptoms have diff ered according to statistical considerations, on a neurobiological level, and to their associations with other factors. Th e studies on burnout and mental health problems have relied on self-reports or screening instruments in the assessment of mental health. For assessing mental health, this approach may be inadequate because the duration and clinical validity of the symptoms remain un- known. Self-report measures, especially in cross-sectional designs, may also exaggerate the observed relationships between burnout and mental health problems due to common method variance (i.e., shared variance in measurement that is attributed to instrumentation rather than to the association between the constructs) (Lindell & Whitney 2001). Th ere are no studies on burnout and mental health in which validated diagnostic methods have been used. Th erefore, it is not known how occupational burnout is related to depressive and other mental disorders fulfi lling the standardized diagnostic criteria. Th e few studies conducted on burnout and musculoskeletal problems or cardiovascular diseases support a positive association also between burnout and physical ill health. However, some of the studies concerning cardiovascular diseases were conducted only among men or women. Very few of the studies on burnout and health have analysed this relationship according to gender. Because the work-related correlates of health have been found to vary to some degree between the genders due to, for example, gender segregation in worklife, diff erences in work conditions, diff erential vulnerability to stressors, and diff erent help-seeking behaviour among men and women (Punnet & Herbert 2000, Taylor et al. 2000, Väänänen et al. 2003, Oliver et al. 2005, Rugulies et al. 2006), a gender perspective may be justifi ed when the associations between burnout and health are studied.

35 1. INTRODUCTION

From an epidemiological point of view, there are serious limitations in the studies on burnout and health problems. Most studies have used vocational or organizational samples, primarily of human service work, which were rarely randomly sampled. Th e scarcity of population-based studies is problematic regarding health, because employed samples generally include persons with a good level of functioning. Th is so-called "healthy worker eff ect" (i.e., those employed are healthier than those who are not employed) may dilute any association with health in occupational samples. Furthermore, most studies have had a modest sample size, and some of the surveys have been vulnerable to self-selection bias due to a rather low response rate. Th erefore, currently it is not reliably known to what extent people suff ering from burnout fulfi l the diagnostic criteria for mental disorders, musculoskeletal disorders, or cardiovascular diseases, and whether the associations between burnout and health are similar for men and women.

1.5 Occupational burnout and sickness absence

Medically certifi ed sickness absences are hypothesized to serve as a global measure of health that is tuned to the requirements in each employee's occupational setting (Marmot et al. 1995, Kivimäki et al. 2003). Th ey are often distinguished from short and self-certifi ed absences that can also be viewed as a coping strategy in a diffi cult situation or as an indicator of low motivation (Marmot et al. 1995, Borritz et al. 2006). Especially long sickness absences have been found to predict future disability pensioning (Kivimäki et al. 2004, Lund et al. 2007). Th erefore, they can also be used as a proxy measure for work disability. A summary of the research publications on the relationship between burnout and sickness absence is presented in Table 8. Burnout predicted sickness absences in two longitudinal studies. A higher level of burnout was related to a greater number of medically certifi ed spells of absence during a 3-year follow-up in a multi-national forest industry organization (Toppinen-Tanner et al. 2005) and to longer company-registered absence duration during a 1-year follow-up in a nutrition production organization (Bakker et al. 2003). In the former study, burnout increased

36 1. INTRODUCTION ed Work-related exhaustion Work-related was positively associated with self-reported sickness absence days and spells in a cross-sectional and a prospective setting. Burnout was positively associated with future medically certiﬁ absences. Burnout was positively to the duration related of company-registered sickness absences. Exhaustion was posit- to self- ively related short-term reported absences among male employees. Main results Burnout was positively to the amount related long of self-reported sickness stress-related absences during the past year. Work-related ex- Work-related haustion (CBI) syndrome (MBI-GS) syndrome syndrome (MBI-GS) syndrome tion (MBI) measure achievement-based self-esteem (BM and a self-construed scale) 214 Three-dimensional 3895 Three-dimensional 1446 Emotional exhaus- 3502 Exhaustion with 824 (time 2) 1561 (time 1), Human service workers in several organizations Employees in one multina- tional organization Production em- Production ployees in one organization Transit op- Transit erators in one organization Sample N Burnout Working popu- Working lation 3-year longitudinal survey (80% at time 1, 75% at time 2). tudinal survey (64%) linked to absence registers 1-year longitudinal survey (65%) linked to absence registers medical examination (73%) (response rate) (response survey (71%) ; MBI-GS = Maslach Burnout Inventory-General Survey, BM = Burnout MBI = Maslach Burnout Measure, Inventory; ; MBI-GS = Maslach Burnout Inventory-General Survey, Den- mark Finland 3-year longi- Nether- lands USA Cross-sectional Sweden Cross-sectional Country Study design a Borritz et al. 2006 Hallsten et al. 2002 Toppinen- et al. Tanner 2005 Bakker et al. 2003 Cunradi et al. 2005 Author and date not peer-reviewed Table 8. Review of the literature on burnout 8. Review of the literature and sickness absence Table a CBI = Copenhagen Burnout Inventory

37 1. INTRODUCTION

future absences granted for mental disorders and diseases of the musculoskeletal and circulatory systems (Toppinen-Tanner et al. 2005). A prospective association was also registered between a high level of work- related exhaustion and the number of self-certifi ed sickness absences and days during a 3-year follow-up among Danish human service workers (Borriz et al. 2006). In a cross-sectional setting, long self-reported stress-related absences were commoner among those who had high levels of exhaustion combined with performance-based self-esteem in the Swedish working population (Hallsten et al. 2002). A cross-sectional association between work-related exhaustion and self-reported absences was also found among Danish human service workers (Borritz et al. 2006) and North American transit operators (Cunradi et al. 2005). Population-based evidence on the association between burnout and sickness absence is based on exhaustion and self-reported absence (Hallsten et al. 2002). Prospective evidence concerning the relationship between burnout and medically certifi ed sickness absence covers only one occupational branch (Toppinen-Tanner et al. 2005). On the basis of this evidence, it can be concluded that burnout is related to sickness absence, and this relationship further supports the association between burnout and health problems. However, because burnout and exhaustion are associated with health problems (Shirom et al. 2005), the observed relationship between burnout and sickness absence could be fully explained by the co-existing disorders and illnesses. None of the previous studies on burnout and sickness absence have controlled for co-occurring mental disorders or physical illnesses. Th erefore, it is not known whether burnout has any independent contribution to sickness absence or whether the observed associations mainly refl ect the relationship between burnout and ill health.

38 2. PRESENT STUDY

2.1 Framework of the study

Th is study was conducted in the framework of occupational health psychology, a specialty in psychology that aims at promoting healthy workplaces (Quick 1999). A workplace is assumed to be healthy when it supports eﬃ cient and high-quality work production as well as the motivation, job satisfaction, and well-being of the employees. Research in occupational health psychology focusses on psychologically mediated processes between the psychosocial work environment and health (Sauter et al. 1999). Although health is widely understood as complete physical, mental, and social well-being (World Health Organization 1946), a traditional medical disease model of ill-health has mostly been applied in research to date (Schaufeli 2004). A discrepancy of some kind between the individual and his or her environment is a common moderator of outcomes in psychology (Tin- sley 2000). Relevant external or internal demands that are estimated to exceed the resources available to cope with them are assumed to activate the stress process in an individual (Lazarus & Folkman 1986, Lazarus 1991). Work-related and social aspects of the perceived environment are related to employees' physiological, psychological, and behavioural processes (Caplan et al. 1975). Th e perception of the environment is thought to always include individual appraisal (Lazarus 1991). Individual processes may, when prolonged or extremely strong, accumulate as strain and aﬀ ect the health status of an employee. Th e interactions between the environment and health are complex (see Figure 1). Every association can be interpreted in both directions, from the environment to health and from health to the environment. In addition

39 2. PRESENT STUDY

to the characteristics of the work environment, also individual factors aff ect the stress process (Caplan et al. 1975, Lindström et al. 2005). Th e theoretical stress process can be viewed as resembling the generalized adaptation syndrome (GAS) which is presented in psychophysical terms (Selye 1956). In GAS, excessive demands fi rst induce an alarm reaction and the mobilization of energetic resources, then resistance, and fi nally exhaustion as the capacity for adjustment is depleted, if resistance fails. Th e core elements in the theories of psychosocial stress (i.e., the personal meaning of the demands, the discrepancy between the individual and his or her environment, and the coping strategies used) are included also in the meta-theory of occupational burnout (Schaufeli & Enzmann 1998), but the status of burnout in the arena of occupational health psychology is not clearly defi ned. Even though burnout is more or less unanimously defi ned as a consequence of chronic work-related stress, it has sometimes been considered a process causing illnesses and sometimes more of an illness or adverse outcome of stress in itself (Maslach 2001, Maslach et al. 2001). Due to the variability of theoretical models and practical operationalizations in psychology, the empirical results are always to some extent dependent on the chosen definitions and measures. In the present study, burnout has been regarded as a three-dimensional work-related syndrome of exhaustion, cynicism, and diminished professional effi cacy (Schaufeli et al. 1996, Maslach et al. 2001). Th is psychological defi nition encompasses the process of energy depletion at work instead of reducing burnout to a state of fatigue (Schaufeli & Taris 2005), and it therefore suits the purposes of the present study in the occupational health psychology framework. Due to its wide usage, the qualities and shortcomings of the dimensional burnout concept are well-known, and the results are also comparable with many other national and international burnout studies. Because the exhaustion dimension of burnout is regarded as the core of the syndrome in the dimensional models (Brenninkmeijer & van Yperen 2003, Roelofs et al. 2005, Schaufeli & Taris 2005), it is unlikely that exhaustion-based and dimensional operationalizations of burnout would yield essentially diff erent results.

40 2. PRESENT STUDY

2.2 Aims of the study

Th e principal aim of the present study was to clarify the status of burnout in the arena of occupational health psychology. With the use of a nationally representative sample, it is possible to avoid the shortcomings of previous studies in which non-random samples were used. More speciﬁ cally, the objective was to determine the extent to which burnout co-occurs with common health problems, that is mental disorders, musculoskeletal disorders, and cardiovascular diseases. Th e examination of health in this study was restricted to these diseases due to their vast impact on work disability. Furthermore, whether burnout has any independent status in relation to ill health was explored with respect to its associations with work characteristics and work disability. An additional aim was to establish the role of some individual factors in the level and prevalence of burnout in an adult working population aged 30 years or over. Most of the analyses were stratiﬁ ed by gender because many health-related

Individual factors • gender • age • education • occupational status • marital status

Perceived psychosocial Physiological, psychological, Health and illnesses work environment and behavioural processes: • depressive disorders • job strain transient and accumulated • anxiety disorders • occupational burnout • alcohol dependence • musculoskeletal disorders • cardiovascular diseases • sickness absence

Figure 1. The stress process (modiﬁ ed from Caplan et al. 1975 and Lindström et al. 2005). Factors examined in the present study are printed in italics (the arrows indicate an associative relationship).

41 2. PRESENT STUDY

aspects of work have been recently found to be gender-related. Th e variables examined in this study are presented in the framework of the hypothesized associations between the psychosocial work environment and health in occupational health psychology in Figure 1.

Th e speciﬁ c study questions were: 1. How is burnout related to gender, age, education, occupational grade, and marital status in a working population (I, II)? 2. How is job strain related to burnout and depressive disorders (II)? 3. How is burnout related to depressive disorders, anxiety disorders, alcohol dependence, musculoskeletal disorders, and cardiovascular diseases among men and women, and what is the overall co-occurrence of burnout with mental disorders and physical illnesses ( III, IV, V)? 4. How does burnout contribute to long medically certiﬁ ed sickness absence among men and women (VI)?

42 3. METHODS

3.1 Procedure

A multidisciplinary epidemiological health study, the Health 2000 Study, was carried out in Finland during August 2000 and June 2001 to obtain up-to-date information on the most important national public health problems, including their causes and treatment, as well as the functional capacity and work ability of the population (Aromaa & Koskinen 2004). Due to a fi nancial imperative to set priorities, this health-oriented study focussed on persons over 30 years of age, among whom illnesses are, on average, commoner. Th e two-stage stratifi ed cluster sample represented the population living on the Finnish mainland and included 8028 persons aged 30 years or over. Th e fi ve Finnish university hospital districts were used for stratifi cation and sampling, each serving about one million inhabitants and diff ering in geography, economic structure, health services, and the socio-demographic characteristics of the population. From each of the fi ve strata, 16 health care districts were sampled as clusters, adding up to 80 districts in the whole country. Firstly, the 15 largest health care districts were included with a probability of one. Next, the other 65 health care districts were included in the sample with a probability proportional to the population size. Finally, from each of these 80 areas, a random sample was drawn from the National Population Register so that the total number of persons drawn from each stratum was proportional to the population size. (Aromaa & Koskinen 2004) Th e participants were fi rst interviewed at home by trained interviewers of Statistics Finland, the Finnish National Bureau for Statistics. Th e structured interview lasted about 90 minutes and covered information

43 3. METHODS

on socio-demographic factors, health and illnesses, use of medication and health services, living habits including smoking, type of work, work capacity, and the need for health services. Th en the participants were given a questionnaire that covered information on their functional capacity, leisure-time activities, physical activity, alcohol consumption, job strain, burnout, and depressive symptoms. Th e questionnaire was to be returned at the time of the clinical health examination, which included a structured interview on mental health, approximately 4 weeks later. Information on sickness absence was extracted from a register of Th e Social Insurance Institute of Finland. It was linked to the other data by means of each participant's personal identifi cation number, which is given to all Finns at birth and used for all contacts in health care. During the fi rst interview, the respondents received an information leafl et on the study and gave their written consent to participate. Th e Health 2000 Study was approved in 2000 by the ethics committee for epidemiology and public health in the hospital district of Helsinki and Uusimaa in Finland.

3.2 Participants

Of the total sample (n=8028), 7419 persons participated in at least one phase of the study. Th e participants accounted for 93% of the 7977 persons alive on the day the study began. Of the 558 non-participants, 416 refused, 110 were not located, and 32 were abroad. Of the total sample, 5871 persons were of working age (30 to 64 years). Of this base population for studies I–V, 88% was interviewed, 84% returned the questionnaire, 82% participated in the clinical health examination, and 80% participated in the mental health interview. At the time of the ﬁ rst interview, 95 participants were on sick leave, and 34 were on leave of absence. In study VI, the 30- to 60-year-olds were used as the base population (n=5380). Th e participants were excluded if they were not currently employed according to their main activity, were on maternity or parental leave, had more than one missing value per dimension on the burnout inventory, or had missing data on the relevant health variables. A maximum of one missing value per dimension in the burnout inventory was replaced

44 3. METHODS ed sick- Logistic regression analysis, analysis of variance Medically certiﬁ ness absence Sum score, Sum score, as categorized Cross-tabulations, logistic regression analysis Physical illnesses Sum score, as Sum score, categorized and continuous Logistic regression analysis Alcohol dependence Sum score, Sum score, as continuous a disorders Sum score, Sum score, as categorized Cross-tabulations, risk ratios Job strain Depressive Sum score, Sum score, as dichotomized Logistic regression analysis graphic factors scores, as con- scores, tinuous analyses of variance and covariance CharacteristicBase population (n)Participation (%): Interview Questionnaire Health examination Study I Mental health interview 5871 Sickness absence registerStudy population (n): Study II participants Working 5871 Final study sample - - 84Study design: - 88 Study III to Burnout in relation 3637 5871 Study IV Socio-demo- 3424 80 84 - 88 - Study V 3387 5871 3270 Study VI 80 84 88 3387 5871 - - 3270 80 5380 3387 84 88 3251 - - 3473 84 - 88 3368 82 3307 - 3151 84 80 87 83 100 Results by gender Yes No Partly Yes Partly Yes Age of the participants Burnout variable 30 - 64 years 30 - 64 years 30 - 64 years Dimensional 30 - 64 years 30 - 64 years 30 - 60 years Statistical analysis Multivariate In the summary of the results, logistic regression analysis was used to calculate the odds ratios logistic regression In the summary of results, Table 9. Descriptive characteristics of the studies I-VI Table a

45 3. METHODS

by the mean of the existing values of the respondent on the particular dimension. Th e size of the fi nal study population ranged from 3151 to 3424 persons in the studies I–VI (Table 9). Th e weighted gender and age distribution of the participants was crudely similar to the corresponding distribution of the total Finnish work force (Statistics Finland 2000). Th e weighted percentage of women was 48%, and the mean age of the participants was 44 years. Of the participants, 78% was married or cohabiting, and 30% was manual workers. Th e participants represented diff erent occupational branches (Ahola et al. 2004). Table 10 contains a detailed description of the participants by gender. Th e men and women diff ered statistically signifi cantly with respect to most of the socio-demographic and clinical characteristics examined. Compared with the men, the women were a little older, had a higher level of education, were more often unmarried, worked more often part time, and were more often in lower-level non-manual work, in passive or high-strain work, and in physically non-strenuous work. Furthermore, the women had less risky health behaviour than men. However, there was no gender diff erence in the level of burnout. Instead, the women had a higher prevalence of depressive and anxiety disorders and depressive symptoms, but a lower prevalence of alcohol dependence, and more long sickness absences than the men.

3.3 Measures

Occupational burnout Occupational burnout was measured with the Maslach Burnout Inven- tory – General Survey (MBI-GS) (Schaufeli et al. 1996, Kalimo et al. 2006). Th e MBI-GS consists of the following three subscales: exhaustion (fi ve items, Cronbach's = 0.91), cynicism (fi ve items, = 0.79), and (diminished) professional effi cacy (six items, = 0.82). Th e items were scored on a 7-point frequency rating scale ranging from 0 (never) to 6 (daily). High scores on exhaustion and cynicism and low scores on professional effi cacy were indicative of burnout. Th e items of professional effi cacy were reversed (diminished professional effi cacy).

46 3. METHODS (2)= 3.60, p= 0.166 (3)= 10.88, p= 0.013 (2)= 56.26, p< 0.001 (3)= 92.12, p< 0.001 (3)= 261.2, p< 0.001 All Men Women Difference n (weighted %) n (weighted %) n (weighted %) (N=3424) No Mild SevereSocio-demographic characteristics Age 30-34 years 35-44 years 45-54 years 55-64 yearsBasic education < 9 years 2475 (72.2) 9-11 years 867 (25.4) 12 years 82 ( 2.4) education Vocational 1249 (72.8) None or a course 546 (15.6) 1165 (33.5) School 434 (25.3) 1288 (38.3) Institute 33 ( 1.9) Higher education 425 (12.7) 1226 (71.6) 287 (16.6)Occupational status 601 (34.9) Upper non-manual 433 (25.5) 626 (36.8) 1206 (35.9) Lower non-manual 1108 (32.3) 49 ( 2.9) 202 (11.7) Manual 259 (14.4) 851 (25.2) 1108 (31.8) Self-employed 564 (31.8) Married 688 (40.2) 662 (39.9) 576 (33.6) 517 (15.0) 223 (13.9) 445 (26.0) 1202 (35.3) 452 (26.2) 943 (27.4) 854 (24.5) 518 (31.2) 940 (26.9) 532 (30.9) 241 (14.0) 709 (41.4) 406 (24.3) 656 (37.9) 458 (26.6) 321 (18.7) 519 (15.4) 267 (15.6) 1018 (30.2) 276 (16.1) 493 (28.7) 485 (28.2) 2670 (78.0) 533 (30.9) 333 (19.5) 655 (38.3) 673 (39.4) 1379 (80.3) 186 (11.0) 363 (21.4) 1291 (75.5) (1)= 12.91, p< 0.001 Burnout (N=3424) Table 10. Socio-demographic and clinical characteristics of the study population by gender Table Table 10. continues... Table

47 3. METHODS (3)= 36.08, p< 0.001 (2)= 20.23, p< 0.001 F(1,3365)= 45.79, p< 0.001 F(1,3276)= 403.3, p<0.001 , b , a (N=3368) (N=3270) : mean; SE 26.4; 0.07 26.9; 0.10 25.9; 0.11 2 kg/m drinks/month: mean; SE 26.3; 0.73 38.2; 1.17 13.0; 0.57 Alcohol consumption Sickness absence (N=3211) 654 (20.3) 263 (16.4) 391 (24.6) (1)= 36.90, p< 0.001 Clinical characteristics, mental-related disorderDepressive Anxiety disorderAlcohol dependenceLifetime mental disorder symptomsDepressive Job strain Low-strain work Active work 204 ( 6.1) Passive work High-strain work 306 ( 9.3) 136 ( 4.3) 125 ( 3.8)Full-time work Clinical characteristics, 616 (18.7) 64 ( 3.9)physical-related 122 ( 7.4)Musculoskeletal disorder 112 ( 6.9) 46 ( 2.8) diseaseCardiovascular 838 (26.5) 229 (14.0) of workPhysical strenuousness 140 ( 8.5) Sedentary work 700 (22.1) 184 (11.4) Some walking 732 (23.0) (1)= 23.52, p< 0.001 910 (28.4) 24 ( 1.5) 472 (29.5) ( 4.9) 79 A lot of walking (1)= 15.03, p< 0.001 387 (23.9) 1003 (30.2) 3034 (92.9) (1)= 56.22, p< 0.001 strenuous Very (1)= 11.00, p< 0.001 Daily smoking 389 (24.3) (1)= 53.17, p< 0.001 331 (20.6) 525 (15.9) 409 (25.5)Physical activity 366 (23.1) 1577 (96.3) < 2 times a week 503 (30.1) 2-3 times a week 1336 (40.0) 311 (19.7) > 3 times a week 275 (16.5) 401 (25.6) 501 (31.6) Body mass index 1457 (89.2) 882 (26.2) 782 (23.5) 500 (30.2) (1)= 64.25, p< 0.001 660 (39.6) 332 (10.2) (1)= 0.004, p= 0.947 250 (15.4) 382 (22.9) 846 (25.3) 1474 (44.1) 378 (22.8) (1)= 0.64, p= 0.423 1181 (35.1) 245 (14.8) 676 (40.6) 706 (20.9) 481 (28.7) 784 (46.7) 500 (29.9) 404 (24.3) 590 (35.0) 87 ( 5.2) 308 (18.3) 365 (21.5) 690 (41.2) (3)= 87.29, p< 0.001 591 (35.1) (1)= 19.46, p< 0.001 398 (23.7) Data missing for 90 persons; Data missing for 1 person; a b Table 10. continues... Table

48 3. METHODS

For the assessment of the level of burnout, a weighted sum score of the dimensional sum scores was calculated (II–VI) (Kalimo et al. 2006). Exhaustion, cynicism, and diminished professional effi cacy had diff erent weights in the syndrome (Kalimo et al. 2003). Th is syndrome indicator had been constructed with the help of a discriminant function analysis, in which various health-related indicators were used as dependent variables (Kalimo & Toppinen 1997). Th e coeffi cients were formed by weighting each dimension so that the scores corresponded to the original response scale (0.4 x exhaustion + 0.3 x cynicism + 0.3 x diminished professional effi cacy). Burnout and the dimensional scores were categorized as fol- lows (Kalimo et al. 2003, 2006): no symptoms (sum score 0–1.49), mild symptoms (sum score 1.5–3.49), and severe symptoms (sum score 3.5–6). According to this categorization, symptoms that were experienced approximately daily or weekly were severe, they occurred monthly when mild, and they were experienced only a few times a year or never in cases of no burnout (III, VI). In study II, burnout was dichotomized as no burnout versus burnout (mild or severe).

Mental disorders Th e assessment of the 12-month prevalence of mental disorders was based on the Composite International Diagnostic Interview (CIDI, version 2.1) (Andrews & Peters 1998, Wittchen et al. 1998), which is a fully standardized diagnostic interview for the assessment of mental disorders for research purposes. A Finnish translation of the German, computerized version of the CIDI was used (M-CIDI). Th e CIDI questions were designed to elucidate symptoms and behaviour that map on each diagnostic criterion for mental disorders according to the Diagnostic and Statistical Manual of Mental Disorders – Fourth Edition (DSM-IV) (American Psychiatric Association 1994) in addition to information about their onset, recency, and associated impairments (Andrews & Peters 1998). In approximately 75 minutes, the CIDI allows the estimation of DSM-IV diagnoses for major mental disorders (i.e., mental disorders due to general medical conditions, substance-related, psychotic, mood, anxiety, somatoform and dissociative, and eating disorders), while the assessment of personality disorders was not included (Wittchen et al. 1998). However, in the CIDI interview in the Health 2000 Study,

49 3. METHODS

mental disorders due to general medical conditions, special phobias, obsessive-compulsive and post-traumatic stress disorders, somatoform and dissociative disorders, and eating disorders were not assessed because the time allocated for the mental health interview was set at 30 minutes (Aromaa & Koskinen 2004). Th e 21 interviewers were health care professionals trained for 3–4 days in the use of CIDI interview by physicians who had themselves been trained by an authorized trainer of the World Health Organization. In a separate analysis, the CIDI interviewers examined, pairwise and independently, a consecutive series of 49 visitors to occupational health services to test the test-retest reliability of the CIDI depressive disorders section. Th e Kappa values for the two interviews were 0.88 (95% CI 0.64–1.00, observer agreement 94%) for major depressive disorder and 0.88 (95% CI 0.64–1.00, observer agreement 98%) for dysthymic disorder (Pirkola et al. 2005). Depressive disorders included major depressive disorder and dysthymic disorder in studies II and VI. In study III, also minor depressive disorder was included in the depressive disorders. Minor depressive disorder comprises current sub-clinical depression (i.e., 2–4 symptoms of depression). Th e anxiety disorders (unpublished data) included panic disorder with or without agoraphobia, generalized anxiety disorder, social phobia, phobia not otherwise specifi ed, and agoraphobia without panic disorder. Th e alcohol use disorders included alcohol dependence and alcohol abuse. A participant was identifi ed as having a mental disorder (VI) if he or she fulfi lled the criteria for a depressive (major depressive and dysthymic disorder), anxiety, or alcohol use disorder. Lifetime mental disorders (II) were assessed with a single-item question in the home interview asking whether a physician had ever confi rmed a psychiatric illness or a mental disorder for the participant.

Depressive symptoms Th e original Beck Depression Inventory (Beck et al. 1961, 1988) was used to assess depressive symptoms as an indicator of mental health (II, V). Th e inventory consisted of 21 items scored from 0 to 3. An acceptable answer was expected for at least 14 items. Missing values (7 at the most) were replaced by the mean of the existing values of the respondent. A sum

50 3. METHODS

score was then calculated for the depressive symptoms (II, V). Depres- sive symptoms were dichotomized (II) as no depressive symptoms (0–9 points) or depressive symptoms (10–63 points) (Beck et al. 1988).

Physical illnesses Th e clinical health examination began with a symptom interview covering musculoskeletal, cardiovascular and respiratory symptoms, atopy, and allergies. Th en the participants were given an additional questionnaire about infections and vaccinations to be fi lled out during the health examination. Next, the research physician took a medical history and performed a standard 30-minute clinical examination including tests related to the functioning and movements of the joints. Th e health examination also included the measurement of height and weight, body circumference, electrocardiogram, blood pressure, spirometry, bioimped- ance, heel bone density, and orthopantomography. Th e diagnostic criteria of the physical illnesses were based on current clinical practice. Th e separate diagnoses were divided into the following four main groups: musculoskeletal disorders (V), cardiovascular diseases (V), respiratory diseases, and other physical illnesses including, for example, diabetes, metabolic disorders, skin disorders, and allergies. Th e participant was identifi ed as having a physical illness (VI) if he or she fulfi lled any diagnostic criteria of a musculoskeletal disorder, cardiovascular disease, respiratory disease, or other physical illness.

Sickness absence In Finland, the national sickness insurance scheme covers the entire population and reimburses the loss of income due to a sickness absence on the basis of a medical certiﬁ cate and a period of more than 9 consecutive work days. Th e number of compensated sickness absence days in 2000 and 2001 was extracted from the register of Th e Social Insurance Institution of Finland. Variables of at least one long sickness absence (no/yes) and the number of compensated days were formed for those with at least one absence during the 2-year period (VI).

51 3. METHODS

Job strain Job strain (II) was measured with the Job Content Questionnaire (Karasek et al. 1998). Th e scale of job demands comprised ﬁ ve items (Cronbach's = 0.79), and the scale for job control comprised nine items ( = 0.85). Responses were given on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). To create an indicator of job strain, the job demand and job control scales were dichotomized at their median (Landbergis et al. 1994), and the following four subgroups were formed for work (Karasek 1979, Karasek & Th eorell 1990): low-strain work (low demands with high control), active work (high demands with high control), passive work (low demands with low control), and high- strain work (high demands with low control).

Socio-demographic factors Information on the socio-demographic factors was collected during the ﬁ rst interview at home. Age was categorized in diﬀ erent ways in studies I–VI. Marital status was dichotomized as married or co-habiting (married) and the rest (unmarried). Basic education was categorized as less than comprehensive schooling (< 9 years), comprehensive schooling (9–11 years), and secondary schooling (12 years). Vocational education was categorized as a course or less, school level, institute level, or higher vocational education. Occupational status was formed on the basis of occupation and the type of business: upper grade non-manual, lower grade non-manual, manual, and self-employment (Statistics Finland 1999). In study VI, the groups of upper and lower grade non-manual work were combined as non-manual work.

Confounding factors In addition to socio-demographic factors, health behaviour and the physical strenuousness of the work were used as confounding factors in the analyses concerning physical illnesses, and work hours and health behaviour were used in the analyses for job strain. Th e physical strenuousness of work was assessed with a questionnaire using a single question with a following 4-point scale: mostly sedentary work, work including a

52 3. METHODS

fair amount of walking, work including a lot of walking, climbing stairs or carrying heavy loads, or very strenuous work. Daily smoking (no/yes) was inquired about in the home interview. Habitual alcohol consumption and health-enhancing physical activity were assessed with the questionnaire. Th e participant's report of the frequency and amount of consumed beer, wine, and spirits was transformed into drinks per month. Health-enhancing physical activity included exercise causing at least slight shortness of breath and sweating for a minimum of 30 minutes at a time. It was classiﬁ ed as once a week or less, two to three times a week, and at least four times a week. Body mass index (kg/m2) was calculated on the basis of the participants' height and weight, which were measured in the clinical health examination. Work hours were inquired about in the home interview and classiﬁ ed as full-time or part-time.

3.4 Statistical analyses

Cross-tabulations and chi-square tests were used to describe the participants according to gender and to analyse the association between burnout and depressive disorders (III), musculoskeletal disorders (V), and cardiovascular diseases (V). In addition, cross-tabulations were used to summarize the overall co-occurrence of burnout, mental disorders, and physical illnesses (unpublished data). A 2x5 multivariate analysis of variance was used to analyse the relation of gender and age to burnout using continuous variables of exhaustion, cynicism, and lack of professional effi cacy as dependent variables at the same time to take into account the multi-dimensional nature of burnout (I). When the multivariate eff ect was signifi cant, univariate analyses of variance were conducted for each burnout dimension separately. Th e focus was on the variables that had a statistically signifi cant eff ect on every burnout dimension. Th e diff erences between the categories of the independent variables were analysed with 95% confi dence intervals of the means. Th e relation of basic and vocational education, occupational grade, and marital status to the continuous burnout variable was analysed with multivariate analyses of covariance separately for the men and women using the categorical factors as independent variables one at a time and age as a continuous variable as a covariate.

53 3. METHODS

Logistic regression analysis was used to investigate the relationship between the socio-demographic factors and burnout (II), between job strain and burnout and depressive disorders (II), and between burnout and depressive disorders (III), alcohol dependence (IV), anxiety disorders (unpublished data), musculoskeletal disorders (V), cardiovascular diseases (V), and sickness absence (VI). Th e models, except in study III, were adjusted for potential confounding factors. In the models of job strain, an interaction term was applied to test whether the association was dependent on gender (II). In studies III–VI, the analyses were conducted according to gender. Th e analyses of sickness absence were repeated for the employees with mental or physical disorders to examine whether burnout increased the odds for sickness absence also in an unhealthy population. A univariate analysis of variance was used for the employees with sickness absences to calculate the mean diff erence and its signifi cance in the logarithmically transformed number of sickness absence days separately according to burnout, mental disorders, and physical illness (VI). Th ese analyses were adjusted for socio-demographic factors, and the analyses for burnout were adjusted further for mental disorders and physical illnesses. Sampling parameters and weighting adjustment were used in all of the analyses to account for the survey design complexities, including clustering in a stratifi ed sample, and the loss of participants (Lehtonen et al. 2003, Aromaa & Koskinen 2004). Th e data were analysed using the SAS (I) and SUDAAN (II–VI) statistical program packages. Th e SUDAAN has been specifi cally designed to analyse cluster-correlated data in complex sample surveys.

54 4. RESULTS

Th e results are presented in accordance with study questions 1–4. Firstly, the associations between socio-demographic factors (i.e., gender, age, marital status, basic and vocational education, and occupational grade) and burnout are presented (I, II). Secondly, the relationship of job strain to burnout and to depressive disorders is compared (II). Th irdly, the associations between burnout and mental disorders (i.e., depressive disorders, anxiety disorders, and alcohol dependence), musculoskeletal disorders, and cardiovascular diseases are shown (III–V, and unpublished data). Finally, the contribution of burnout to long medically certiﬁ ed sickness absences is investigated (VI).

4.1 Association between socio-demographic factors and occupational burnout

Gender, age and burnout Th e odds of having burnout did not diff er by gender when burnout comprised both mild and severe symptoms (Table 11) (II). When burnout was analysed as a continuous variable, gender did have a signifi cant multivariate main eff ect on burnout (F=24.7, df=3, p< 0.001). However, the univariate eff ects of gender were signifi cant only for the dimensions of exhaustion and cynicism. Th e level of exhaustion was higher among the women than among the men. Th e diff erence in the means of cynicism for the men and women was in the opposite direction, but it was not statistically signifi cant (Table 12) (I). Having burnout was more probable among the 55- to 64-year-old employees than in the younger age groups (Table 11) (II). Age had a

55 4. RESULTS

Table 11. Odds ratios for burnout by gender and age, and gender- and age-adjusted odds ratios for burnout by occupational grade and marital status

Burnout Factor N Cases Odds ratio (95% CIa) Gender Men 1637 441 1.00 Women 1633 459 1.07 (0.92-1.25) Age (years) 30-34 517 130 1.00 35-44 1116 265 0.91 (0.73-1.14) 45-54 1236 347 1.15 (0.92-1.45) 55-64 401 158 1.89 (1.44-2.49) Burnout Factor N Cases Adjustedb odds ratio (95% CIa) Occupational grade Upper non-manual 922 230 1.00 Lower non-manual 896 200 0.84 (0.68-1.04) Manual 958 335 1.69 (1.36-2.10) Self-employed 490 135 1.14 (0.89-1.46) Marital status Married 2558 670 1.00 Unmarried 712 230 1.33 (1.10-1.61) a CI indicates confi dence interval b Adjusted for gender and age signifi cant multivariate main eff ect on burnout when it was treated as a continuous variable (F=6.4, df=12, p< 0.001), and this eff ect was signifi cant for all of the burnout dimensions. Th e level of exhaustion was higher in the age group of 53–64 years than among the 30- to 41- year-olds, and the level of cynicism was higher in the age group of 53–64 years than among the 36- to 52-year-olds. In addition, professional effi cacy was lower among the 53- to 64-year-olds than among the 42- to 47-year-olds, among whom it was lower than among the 30- to 35-year- olds. Furthermore, professional effi cacy was lower in the age group of 48–52 years than in the age group of 36–41 years. Th e interaction eff ect between gender and age was not statistically signifi cant.

56 4. RESULTS a N Exhaustion Cynicism Lack of PE a dence intervals dence Men Women cacy Factor N Exhaustion Cynicism Lack of PE AllAge (years) 30-35 36-41 42-47 48-52 53-64 1716Basic education 343 1.00 (0.95-1.05) 0.98 (0.88-1.08) 1.12 (1.07-1.18) < 9 years 350 1.19 (1.06-1.33) 1.25 (1.20-1.31) 0.90 (0.78-1.02) 1.10 (1.00-1.21) 9–11 years 1.00 (0.88-1.13) 373 1708 1.10 (0.98-1.21) 1.05 (0.94-1.16) 1.20 (1.15-1.26) 12 years 332 1.15 (1.02-1.28) 300 1.03 (0.98-1.09) 1.00 (0.87-1.14) 1.02 (0.91-1.13) 1.29 (1.18-1.40) education Vocational 1.22 (1.17-1.28) 340 318 1.05 (0.91-1.18) 0.96 (0.84-1.08) 1.08 (0.97-1.20) 1.07 (0.95-1.20) 1.38 (1.25-1.50) 1.01 (0.91-1.12) None or a course 0.90 (0.79-1.01) 1.24 (1.11-1.36) 373 688 1.13 (1.00-1.25) 1.15 (1.03-1.26) 1.02 (0.93-1.11) 1.42 (1.27-1.56) 576 School 344 0.99 (0.91-1.07) 1.14 (1.06-1.22) 0.96 (0.86-1.06) 1.21 (1.09-1.34) 1.09 (1.00-1.19) 1.20 (1.10-1.31) 1.45 (1.35-1.56) Institute 445 352 1.02 (0.90-1.13) 1.18 (1.09-1.27) 1.03 (0.92-1.15) 452 1.50 (1.35-1.65) 1.22 (1.10-1.34) Higher education 0.98 (0.89-1.07) 1.21 (1.10-1.33) 518 1.29 (1.17-1.42) 1.14 (1.03-1.24) 1.49 (1.36-1.62) 1.35 (1.23-1.47) 532 1.49 (1.35-1.64) Occupational grade 1.03 (0.95-1.11) 1.03 (0.95-1.12) 1.14 (1.04-1.24) 1.48 (1.36-1.60) 0.92 (0.84-1.00) Upper non-manual 241 406 0.95 (0.84-1.07) 1.14 (1.04-1.23) 1.35 (1.21-1.50) 656 Lower non-manual 1.11 (0.96-1.25) 709 1.22 (1.13-1.31) 1.19 (1.06-1.31) 1.01 (0.93-1.09) 458 321 1.04 (0.93-1.15) 1.55 (1.41-1.70) 1.04 (0.96-1.12) Manual 1.00 (0.91-1.09) 1.10 (1.02-1.19) 0.96 (0.84-1.08) 1.09 (1.01-1.17) 267 1.09 (0.98-1.19) 1.06 (0.93-1.19) 1.30 (1.20-1.39) Self-employed 0.87 (0.76-0.97) 276 0.99 (0.91-1.08) 0.99 (0.89-1.09) 1.34 (1.20-1.48) 1.04 (0.90-1.18)Marital status 493 1.17 (1.05-1.30) 1.13 (0.99-1.27) 1.11 (1.02-1.20) 485 533 1.13 (1.02-1.25) Married 1.25 (1.15-1.35) 1.11 (1.01-1.20) 0.92 (0.82-1.01) 673 1.23 (1.12-1.33) 0.99 (0.90-1.08) 0.97 (0.88-1.07) 333 Unmarried 1.08 (0.99-1.16) 1.08 (0.99-1.16) 1.01 (0.92-1.10) 1.05 (0.94-1.17) 0.97 (0.89-1.05) 655 1.07 (0.95-1.19) 1.13 (1.04-1.21) 1.03 (0.94-1.13) 1.15 (1.01-1.29) 1.22 (1.12-1.31) 1.52 (1.42-1.62) 186 1379 1.11 (0.96-1.27) 363 0.97 (0.91-1.03) 337 1.01 (0.84-1.17) 1.43 (1.28-1.57) 1.07 (1.01-1.13) 1.12 (0.99-1.25) 1.29 (1.07-1.50) 1.35 (1.20-1.50) 1.22 (1.09-1.36) 1.22 (1.16-1.28) 1.57 (1.41-1.72) 1.39 (1.25-1.52) 1291 1.19 (1.12-1.26) 417 0.99 (0.93-1.05) 1.25 (1.13-1.37) 1.20 (1.13-1.26) 1.16 (1.05-1.28) 1.31 (1.19-1.43) PE indicates professional efﬁ PE indicates professional Table 12. Level of the burnout to the socio-demographic factors among men and dimensions according Table women: weighted means and their 95% conﬁ a

57 4. RESULTS

Education, occupational grade and burnout Both basic and vocational education had a signifi cant multivariate main eff ect on burnout when age was adjusted for (F=3.5, df=9, p< 0.001, and F=4.5, df=11, p< 0.001 for the women; F=3.9, df=9, p< 0.001, and F=4.0, df=11, p< 0.001 for the men), but the univariate eff ects of these variables were statistically signifi cant only for the women. Th e level of burnout was a little higher among the women who had not fi nished comprehensive school than among those who had (Table 12) (I). Th e women who had little vocational education scored somewhat higher on exhaustion and diminished professional effi cacy than the women with school- and institute-level education and higher on cynicism than the women with a school-level education. Among the men, education had a statistically signifi cant univariate eff ect only on diminished professional effi cacy (Table 12) (I). Mild-to-severe burnout was more prevalent among the manual workers than among the non-manual workers or the self-employed when gender and age were adjusted for (Table 11) (II). Occupational grade also had a signifi cant multivariate eff ect on burnout when age was adjusted for (F=5.1, df=11, p< 0.001 for the women; F=7.1, df=11, p< 0.001 for the men), but the univariate eff ects of occupational grade were signifi cant only for the women. Th e female manual workers scored higher on exhaustion than the self-employed and the lower-level non-manual female workers and higher on cynicism and diminished professional effi cacy than the non-manual female workers (Table 12) (I). As for the men, occupational grade had a statistically signifi cant univariate eff ect on exhaustion and lack of professional effi cacy.

Marital status and burnout As a dichotomy, burnout was more prevalent among those unmarried than among those married when gender and age were adjusted for (Table 11) (II). When the analyses for the continuous burnout variable were stratifi ed by gender, marital status had a signifi cant multivariate main eff ect on burnout only for the men, when age was adjusted for (F=5.0, df=5, p< 0.001). Th e univariate eff ect of marital status among the men was signifi cant for every burnout dimension. Th e level of burnout was

58 4. RESULTS

higher among the unmarried men, but the diff erence was statistically signifi cant only for the cynicism dimension of burnout (Table 12) (I). Th e eff ect of marital status on burnout was not signifi cant among the women.

Brief summary As a summary of the results concerning the ﬁ rst study question on the associations between socio-demographic factors and burnout, it can be concluded that the level of burnout was unrelated to gender. Instead, burnout was more likely among employees who were over 55 years of age, in manual work or unmarried. Th e associations of burnout partly diﬀ ered between the genders. Th e results indicated a negative relationship between burnout and the level of education and occupational status especially among the women and an association between burnout and being unmarried especially among the men.

4.2 Job strain in relation to occupational burnout and depressive disorders

Th e odds ratios of job strain for burnout are presented in Table 13 (II). Th e employees with high-strain work, passive work, and active work had higher odds ratios for burnout than their counterparts with low-strain work. Th e association was the strongest for high-strain work. Th e associations persisted after adjustment for socio-demographic factors, health behaviour, physical illnesses, and the indicators of mental health. Th ere was no signiﬁ cant attenuation in the strength of the association between job strain and burnout after adjustment for depressive symptoms and depressive disorders. Th e odds ratios for depressive disorders in relation to job strain are presented in Table 14 (II). High-strain work and active work were associated with depressive disorders before, but not after, adjustment for burnout. No statistically signiﬁ cant interaction between job strain and gender was evident in the models for burnout and depressive disorders (p> 0.07).

59 4. RESULTS

Table 13. Odds ratios for burnout in relation to job strain

Burnout Adjusteda Adjustedb Adjustedcodds odds ratio odds ratio ratio Work N Cases (95% CId) (95% CId) (95% CId) Low strain 820 95 1.00 1.00 1.00 Active 688 161 2.34 (1.75-3.12) 2.31 (1.72-3.11) 1.90 (1.39-2.60) Passive 880 240 2.94 (2.24-3.86) 3.16 (2.38-4.19) 3.05 (2.25-4.13) High strain 702 342 7.36 (5.62-9.65) 7.86 (5.99-10.3) 6.69 (5.06-8.82) a Adjusted for gender and age b Adjusted for gender, age, marital status, occupational grade, work hours, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, physical illnesses, and lifetime mental disorders c Adjusted for gender, age, marital status, occupational grade, work hours, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, physical illnesses, lifetime mental disorders, depressive symptoms, and depressive disorders d CI indicates conﬁ dence interval

Table 14. Odds ratios for depressive disorders in relation to job strain

Depressive disorders Adjusteda Adjustedb Adjustedc odds ratio odds ratio odds ratio Work N Cases (95% CId) (95% CId) (95% CId) Low strain 820 37 1.00 (ref.) 1.00 (ref.) 1.00 (ref.) Active 688 49 1.57 (1.03-2.38) 1.58 (1.01-2.49) 1.20 (0.76-1.91) Passive 880 45 1.06 (0.69-1.63) 1.03 (0.63-1.66) 0.68 (0.41-1.13) High strain 702 55 1.69 (1.11-2.58) 1.66 (1.03-2.66) 0.83 (0.52-1.34) a Adjusted for gender and age b Adjusted for gender, age, marital status, occupational grade, work hours, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, physical illnesses, and lifetime mental disorders c Adjusted for gender, age, marital status, occupational grade, work hours, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, physical illnesses, lifetime mental disorders, and burnout d CI indicates conﬁ dence interval

60 4. RESULTS

In conclusion, high job strain was related to burnout and to depressive disorders, but the association was stronger for burnout.

4.3 Association between occupational burnout and disorders and illnesses

Burnout and mental disorders Of the employees with severe burnout, 45% had major depressive disorder or dysthymic disorder. Th e gender- and age-adjusted odds of having major depressive disorder or dysthymic disorder accompanied with mild-to-severe burnout was 5.0-fold (95% CI 3.9–6.6; II). When the depressive disorders also included minor depressive disorder, a depressive disorder was present in 53% of the participants with severe burnout, in 20% of the participants with mild burnout, and in 7% of the participants with no burnout. Th e weighted prevalence and odds ratios for depressive disorders are presented according to the level of burnout in Table 15 (III). Compared with the persons without burnout, those with a depressive disorder had a crude odds ratio that was 14-fold when severe burnout was present and 3.2-fold when mild burnout was present. Burnout was related to major depressive disorder among both the men and the women, as was also severe burnout to dysthymic disorder. Mild but not severe burnout was related to minor depressive disorder among the men and women. Severe burnout was related to a substantial probability for major depressive disorder when compared with the situation with no burnout. For the men, the odds of severe burnout in relation to major depressive disorder were 44-fold, and for the women they were 18-fold. However, the gender diﬀ erence was not statistically signiﬁ cant. Moreover among the men, mild burnout was also related to dysthymic disorder. Also the odds of mild burnout versus the situation of no burnout in relation to any depressive disorder were higher among the men than among the women (Table 15) (III). Th e weighted prevalence and odds ratios for depressive disorders in relation to the separate dimensions of burnout are presented in Table 16 (III). A depressive disorder was diagnosed for 35% of those who had severe exhaustion, for 33% of those who had severe cynicism, and for

61 4. RESULTS c (187 cases) a Minor DD sive disorder prevalence OR (95% CI) c b (40 cases) prevalence OR (95% CI) c (173 cases) Dysthymic disorder a prevalence OR (95% CI) c 1.001.00 2.7 1.00 1.2 1.00 0.6 1.00 0.2 1.00 4.2 1.00 3.0 1.00 (391 cases) Major DD dence interval; in the original publication (study III), risk ratios were used dence interval; in the original publication (study III), risk ratios were a Any DD nout 2370nout 7.4 1196 4.3 Gender N prevalence OR (95% CI) Men and Women No bur Mild burnout burnout Severe 822Men 78 No bur 20.3 Mild burnout 52.9 burnout Severe 3.20 (2.55-4.03) 409Women 14.1 (9.22-21.7) 32 No burnout 9.0 17.9 Mild burnout 40.2 1174 46.1 burnout Severe 3.57 (2.59-4.92) 4.83 (3.34-6.98) 413 24.3 (15.2-39.1) 19.0 (9.03-40.1) 46 10.8 2.1 6.9 23.0 8.9 34.1 1.00 58.0 3.28 (1.68-6.42) 6.34 (3.29-12.2) 2.46 (1.86-3.26) 15.4 (6.04-39.3) 44.2 (18.9-104) 11.4 (6.21-20.9) 11.2 9.5 1.9 7.6 12.6 44.8 2.74 (1.87-4.01) 2.39 (1.79-3.21) 12.4 (3.05-50.2) 1.88 (0.82-4.32) 90.3 (15.6-522) 17.6 (9.29-33.3) 4.4 2.2 9.3 1.00 5.8 6.1 1.89 (0.82-4.33) 3.32 (2.08-5.29) 2.00 (0.47-8.47) 5.62 (1.58-20.1) 9.8 9.0 1.84 (1.28-2.65) 1.2 1.67 (0.58-4.84) 1.00 5.6 1.00 Of the participants with dysthymic disorder, 9 had a concurrent major depressive disorder and 10 had a concurrent minor depres and 10 had a concurrent disorder major depressive 9 had a concurrent Of the participants with dysthymic disorder, DD indicates depressive disorder DD indicates depressive OR indicates odds ratio; CI conﬁ Table 15. Weighted prevalence (%) and odds ratios for depressive disorders in relation to the level of burnout in relation disorders (%) and odds ratios for depressive prevalence 15. Weighted Table to gender according a b c

62 4. RESULTS c

a Minor DD (187 cases) order Prevalence OR (95% CI) c

b (40 cases) Dysthymic disorder Prevalence OR (95% CI) c

a Major DD (173 cases) Prevalence OR (95% CI) c

a dence interval; in the original publication, risk ratios were used dence interval; in the original publication, risk ratios were Any DD (391 cases) cacy 922204 16.7 23.6 2.17 (1.75-2.69) 3.35 (2.33-4.80) 6.9 13.4 1.97 (1.41-2.76) 4.08 (2.60-6.41) 1.7 2.4 2.08 (1.13-3.83) 2.85 (1.06-7.70) 8.2 8.8 2.04 (1.51-2.77) 2.21 (1.36-3.59) d d 2150 8.5 1.00 3.6 1.00 0.9 1.00 4.2 1.00 d Level of dimension N Prevalence OR (95% CI) Mild lack of PE Severe lack of PE Severe No exhaustionMild exhaustion 2500 exhaustionSevere 602 174 8.0 20.5No cynicism 35.1Mild cynicism 1.00 2.96 (2.32-3.78) cynicismSevere 6.21 (4.43-8.70) 2414 692 170 9.9 22.8No lack of PE 8.3 18.4 32.9 3.82 (2.71-5.40) 10.3 (6.84-15.4) 1.00 2.48 (1.94-3.17) 2.8 5.42 (3.86-7.62) 2.3 5.3 1.00 8.0 21.1 3.50 (1.72-7.13) 8.54 (3.69-19.7) 2.61 (1.90-3.59) 8.07 (5.33-12.2 3.2 8.8 8.1 1.3 5.3 1.00 1.95 (1.43-2.65) 1.79 (1.01-3.16) 0.7 1.45 (0.63-3.34) 6.38 (2.76-14.5) 1.00 9.2 8.1 2.20 (1.60-3.05) 1.93 (1.08-3.45) 0.9 1.00 4.7 1.00 4.4 1.00 Of participants with dysthymic disorder, 9 had a concurrent major depressive disorder and 10 a concurrent minor depressive dis minor depressive and 10 a concurrent disorder major depressive 9 had a concurrent Of participants with dysthymic disorder, efﬁ PE indicates professional DD indicates depressive disorder DD indicates depressive OR indicates odds ratio; CI conﬁ Table 16. Weighted prevalence (%) and odds ratios for depressive disorders in relation to the level of burnout in relation disorders (%) and odds ratios for depressive prevalence 16. Weighted Table dimensions a b c d

63 4. RESULTS

24% of those who had severly diminished professional effi cacy. All of the dimensions of burnout, at both severe and mild levels, were signifi cantly related to the occurrence of a depressive disorder. Concerning diff erent depressive disorders, only the association between mild cynicism and dysthymic disorder was not statistically signifi cant. An anxiety disorder was present among 21% of the employees with severe burnout, among 8% with mild burnout, and among 2% with no burnout. Th e odds ratios for burnout in relation to anxiety disorders according to gender are presented in Table 17 (unpublished data). Burnout was a signifi cant correlate of anxiety disorders for both genders. For the anxiety disorders, each 1-point increase in the burnout sum score was associated with a 129% increase in the odds for anxiety disorders among the men and a 105% increase among the women. Th ese associations did not attenuate after adjustment for socio-demographic factors.

Table 17. Odds ratios of burnout for anxiety disorders according to gender

Anxiety disorders Odds ratio Adjusteda odds ratio Gender N Cases (95% CIb) (95% CIb) Men 1608 46 2.29 (1.83-2.85) 2.30 (1.81-2.93) Women 1601 79 2.05 (1.71-2.47) 2.08 (1.72-2.51) a Adjusted for age, marital status, and basic education b CI indicates conﬁ dence interval

Table 18. Odds ratios of burnout for alcohol dependence according to gender

Alcohol dependence Odds ratio Adjusteda odds ratio Gender N Cases (95% CIb) (95% CIb) Men 1622 112 1.51 (1.27-1.78) 1.51 (1.28-1.79) Women 1629 25 1.80 (1.35-2.40) 2.06 (1.52-2.81) a Adjusted for age, marital status, and basic education b CI indicates conﬁ dence interval

64 4. RESULTS

Alcohol dependence was present among 10% of all the employees with severe burnout, among 7% of those with mild burnout, and among 3% of those with no burnout. Burnout was a signiﬁ cant correlate of alcohol dependence for both genders. Th e odds ratios for burnout in relation to alcohol dependence are presented in Table 18 (IV) according to gender. For alcohol dependence, each 1-point increase in the burnout sum score was associated with an 80% increase in the odds for alcohol dependence among the women and a 51% increase among the men. Th e associations remained signiﬁ cant after adjustment for socio- demographic factors.

Burnout and musculoskeletal disorders Of the persons with severe burnout, 47% had a musculoskeletal disorder compared with 36% of those with mild burnout and with 28% of those with no burnout. Th e weighted prevalence of musculoskeletal disorders increased with the level of burnout for both genders (Table 19) (V). Th e prevalence of musculoskeletal disorders increased also with the level of all three dimensions of burnout in the whole study population (Table 20) (V). Th e odds ratios for musculoskeletal disorders in relation to burnout are presented in Table 21 (V) according to gender. Among the women, burnout was signiﬁ cantly related to musculoskeletal disorders even when socio-demographic factors, the physical strenuousness of work, health behaviour, and depressive symptoms were adjusted for. After these ad-

Table 19. Weighted prevalence (%) of musculoskeletal disorders according to the level of burnout

All Men Women Burnout N Prevalence N Prevalence N Preva- lence No 2438 27.6 1228 28.3 1210 26.7 Mild 849 36.1 422 33.9 427 38.4 Severe 81 46.7 33 51.1 48 45.1

(df), p 32.5 (2), <.001 11.3 (2), <.01 24.5 (2), <.001

65 4. RESULTS

Table 20. Weighted prevalence (%) of musculoskeletal disorders according to the level of the burnout dimensions

Exhaustion Cynicism Lack of PEa Level N Prevalence N Prevalence N Prevalence No 2571 27.3 2481 28.1 2209 28.1 Mild 617 38.6 714 34.3 948 33.9 Severe 180 43.2 173 42.1 211 34.9

(df), p 42.7 (2), <.001 22.6 (2), <.001 16.4 (2), <.01 a PE indicates professional efﬁ cacy

Table 21. Odds ratios of burnout for musculoskeletal disorders according to gender

Odds Adjusteda Adjustedb Adjustedc ratio odds ratio odds ratio odds ratio Gender N (95% CId) (95% CId) (95% CId) (95% CId) Men 1628 1.25 (1.11-1.40) 1.21 (1.08-1.36) 1.20 (1.06-1.35) 1.08 (0.93-1.25) Women 1596 1.39 (1.25-1.56) 1.29 (1.16-1.45) 1.31 (1.17-1.47) 1.22 (1.07-1.38) a Adjusted for age, marital status, basic education, occupational grade, and physical strenuousness of work b Adjusted for age, marital status, basic education, occupational grade, physical strenuousness of work, daily smoking, alcohol consumption, health-enhancing physical activity, and body mass index c Adjusted for age, marital status, basic education, occupational grade, physical strenuousness of work, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, and depressive symptoms d CI indicates conﬁ dence interval

justments, each 1-point increase in burnout score was related to a 22% increase in the odds for musculoskeletal disorders. Among the men, the association between burnout and musculoskeletal disorders was not signiﬁ cant after depressive symptoms were adjusted for.

66 4. RESULTS

Burnout and cardiovascular diseases Of the persons with severe burnout, 28% had a cardiovascular disease compared with 20% of those with mild burnout and with 14% of those without burnout. Th e weighted prevalence of cardiovascular diseases increased statistically signiﬁ cantly with the level of burnout among the men (Table 22) (V). Th e prevalence of cardiovascular diseases increased with the severity of all three dimensions of burnout in the whole study population (Table 23) (V). Th e odds ratios for cardiovascular diseases in relation to burnout are presented in Table 24 (V). Among the men, burnout was signiﬁ cantly

Table 22. Weighted prevalence (%) of cardiovascular diseases according to the level of burnout

All Men Women Burnout N Prevalence N Prevalence N Prevalence No 2438 14.3 1228 14.3 1210 14.3 Mild 849 19.5 422 21.2 427 17.7 Severe 81 27.6 33 36.4 48 21.1

(df), p 23.1 (2), <.001 24.0 (2), <.001 4.00 (2), n.s.

Table 23. Weighted prevalence (%) of cardiovascular diseases according to the level of the burnout dimensions

Exhaustion Cynicism Lack of PEa Level N Prevalence N Prevalence N Prevalence No 2571 14.5 2481 14.8 2209 14.1 Mild 617 19.7 714 17.1 948 18.8 Severe 180 23.3 173 27.8 211 22.8

(df), p 20.3 (2), <.001 21.6 (2), <.001 18.7 (2), <.001 a PE indicates professional efﬁ cacy

67 4. RESULTS

Table 24. Odds ratios of burnout for cardiovascular diseases according to gender

Odds Adjusted Adjusted Adjusted ratio odds ratioa odds ratiob odds ratioc Gender N (95% CI)d (95% CI)d (95% CI)d (95% CI)d Men 1628 1.38 (1.22-1.56) 1.35 (1.17-1.54) 1.34 (1.16-1.54) 1.35 (1.13-1.61) Women 1598 1.23 (1.07-1.41) 1.03 (0.88-1.21) 1.05 (0.90-1.23) 0.96 (0.81-1.13) a Adjusted for age, marital status, basic education, occupational grade, and physical strenuousness of work b Adjusted for age, marital status, basic education, occupational grade, physical strenuousness of work, daily smoking, alcohol consumption, health-enhancing physical activity, and body mass index c Adjusted for age, marital status, basic education, occupational grade, physical strenuousness of work, daily smoking, alcohol consumption, health-enhancing physical activity, body mass index, and depressive symptoms d CI indicates conﬁ dence interval related to cardiovascular diseases, even when socio-demographic factors, the physical strenuousness of work, health behaviour, and depressive symptoms were adjusted for. After these adjustments, each 1-point increase in the burnout score was related to a 35% increase in the odds for cardiovascular disease among the men. Among the women, the association between burnout and cardiovascular diseases was not signiﬁ cant after the socio-demographic factors and the physical strenuousness of work were adjusted for.

Co-occurrence of burnout with mental disorders and physical illnesses With the dichotomy of “no burnout” versus “mild-to-severe burnout”, 27.2% of the participants had burnout in this study. Th e co-occurrence of mild-to-severe burnout with mental disorders and physical illness in the total sample is presented in Figure 2 (unpublished data). Mental disorders included depressive disorders (i.e., major depressive disorder and dyshymic disorder), anxiety disorders, and alcohol use disorders, and physical illnesses included cardiovascular diseases, musculoskeletal disorders, respiratory illnesses, and a group of other physical illnesses. Participants were included in every combination of burnout and dis-

68 4. RESULTS

Mild-to-severe burnout Mental disorders 3.1% 6.8% 3.6% 4.6% 12.7% 4.6%

34.2%

Physical illnesses 30.4% No burnout or disorders

Figure 2. Co-occurrence of mild-to-severe burnout with mental disorders and physical illnesses in the total sample (weighted prevalences, adding up to 100% of N = 3211)

Severe burnout Mental disorders 0.5% 0.2% 6.2% 0.9% 0.7% 8.3%

46.2%

Physical illnesses 36.9% No severe burnout or disorders

Figure 3. Co-occurrence of severe burnout with mental disorders and physical illnesses in the total sample (weighted prevalences, adding up to 100% of N = 3211) orders. Mild-to-severe burnout co-occurred with physical illnesses in 17.3% of the cases and with mental disorders in 7.7% of the cases in the total sample. A severe level of burnout was present in 2.3% of the participants. Th e co-occurrence of severe burnout with mental disorders and physical

69 4. RESULTS

Table 25. Summary of the weighted prevalence (%) of common mental disorders, musculoskeletal disorders, and cardiovascular diseases according to the level of burnout

Level of Depressive Anxiety Alcohol burnout disorders disorders dependence No 3.2 2.0 3.3 Mild 10.8 7.6 6.7 Severe 45.3 21.0 10.0

Level of Musculoskeletal Cardiovascular No disorder burnout disorders diseases or illness No 27.6 14.3 41.2 Mild 36.1 19.5 25.8 Severe 46.7 27.6 9.6

illness among all the participants is presented in Figure 3 (unpublished data). Again, participants were included in every combination of burnout and disorders. Severe burnout co-occurred with physical illnesses in 1.6% of all the cases and with mental disorders in 1.4% of all the cases in this study. Th e weighted prevalence of depressive disorders, anxiety disorders, alcohol dependence, musculoskeletal disorders, and cardiovascular diseases is summarized in Table 25 according to the level of burnout. Of those with severe burnout, 47% had a musculoskeletal disorder, and 45% had a depressive disorder. Ten per cent of those with severe burnout did not have any co-occurring disorder or illness.

Brief summary Burnout co-occurred to a high extent with mental disorders and physical illnesses. Th ere was a signiﬁ cant association between burnout and depressive disorders, anxiety disorders, and alcohol dependence for both genders, between burnout and musculoskeletal disorders among the women, and between burnout and cardiovascular diseases among the men.

70 4. RESULTS

4.4 Contribution of occupational burnout to long sickness absences

Th e association between burnout and the existence of at least one long sickness absence during a 2-year period is presented in Table 26 (VI). Burnout was positively associated with sickness absence for both genders after adjustment for the socio-demographic factors and mental disorders. After additional adjustment for physical illnesses, severe burnout was still positively related to sickness absence for both genders. Th e adjusted odds of having a long absence with severe burnout were 6.9-fold for the men and 2.1-fold for the women when compared with the situation with no burnout. However, the gender diff erence was not statistically signifi cant. In contrast to men’s results, the women’s association between mild burnout and sickness absence was also statistically signifi cant after adjustment for physical illnesses.

Table 26. Odds ratios for having a long medically certiﬁ ed sickness absence during 2 years in relation to the level of burnout according to gender

Sickness absence Number Adjusteda Adjustedb Adjustedc of odds ratio odds ratio odds ratio Gender N absences (95% CId) (95% CId) (95% CId) Men No burnout 1154 162 1.00 1.00 1.00 Mild burnout 393 79 1.42 (1.05 - 1.92) 1.37 (1.00 - 1.86) 1.26 (0.93 - 1.71) Severe burnout 29 18 8.00 (3.09 - 20.7) 7.21 (2.69 - 19.4) 6.87 (2.65 - 17.8) Women No burnout 1138 239 1.00 1.00 1.00 Mild burnout 389 127 1.71 (1.31 - 2.22) 1.60 (1.23 - 2.09) 1.55 (1.18 - 2.03) Severe burnout 44 21 2.82 (1.49 - 5.37) 2.15 (1.12 - 4.11) 2.10 (1.10 - 4.01) a Adjusted for age, marital status, and occupational status b Adjusted for age, marital status, occupational status, and mental disorders c Adjusted for age, marital status, occupational status, mental disorders, and physical illnesses d CI indicates conﬁ dence interval

71 4. RESULTS b p-value a days Difference of absence Mean number N b p-value a Men with absence with absence Women days Difference of absence Mean number N Burnout No Mild SevereMental disorder No 162 YesPhysical illness 79 18 No 33.8 Yes 47.7 196 93.2 63 0.00 40.8 68 12.3 46.2 191 55.2 22.5 <0.001 49.1 0.407 0.00 7.97 127 21 239 0.00 22.6 0.105 38.2 83.0 33.2 78 309 0.007 0.32 271 41.4 0.00 47.3 34.5 116 0.661 41.2 0.135 27.5 11.7 0.00 11.4 0.094 0.00 0.001 Number of days was log-transformed before the analyses Number of days was log-transformed before Adjusted for age, marital status, and occupational status Table 27. Difference in the mean number of compensated sickness absence days during a 2-year period 27. Difference Table to burnout, and physical ill- among employees with at least one long absence according mental disorders, ness a b

72 4. RESULTS

Among the men with co-occurring disorders, severe burnout was related to 7.7-fold odds (95% CI 3.2–18.4) for sickness absence compared with the situation of no burnout (OR for mild burnout 1.3, 95% CI 0.9–1.8). Among the women with co-occurring disorders, the respective probability of severe burnout for sickness absence was 2.6-fold (95% CI 1.3–5.1), and that for mild burnout was 1.5-fold (95% CI 1.1–2.0). Th e number of compensated sickness absence days in relation to burnout among those who had at least one long absence is presented in Table 27 (VI). Among the participants with absences, the number of sickness absence days was signifi cantly associated with burnout among the men but not among the women. Th e men with severe burnout had 55 excess sickness absence days over the 2-year period when compared with the men without burnout after socio-demographic factors were adjusted for. In comparison, having a physical illness was related to 23 excess days among the men. Th e association between sickness absence days and severe burnout remained statistically signifi cant among the men even after adjustment for co-occurring common mental disorders and physical illnesses. In this adjusted model, the men with severe burnout had 52 excess sickness absence days (p=0.002) over the 2-year period. Th e women with severe burnout had 41 excess sickness absence days after adjustment for socio-demographic factors, but this association did not reach statistical signifi cance. Concerning the results for the fourth study question, the contribution of burnout to work disability, it can be concluded that burnout was related to long medically certifi ed sickness absences for both genders. Th e association between severe burnout and sickness absence was independent of co-occurring common mental disorders and physical illnesses.

73 5. DISCUSSION

5.1. Synopsis of the main ﬁ ndings

In this representative sample of the Finnish working population aged 30 years or over, the prevalence of severe burnout was 2.3%, while some symptoms of burnout, from mild to severe, were present in 27% of the cases. Employees aged 55 years or over had higher odds of having symptoms of burnout than the younger ones did. Manual workers and those unmarried had higher odds of mild-to-severe burnout than the other groups of participants. Especially among the women, those who had not fi nished comprehensive school had a higher level of burnout than the women who had comprehensive schooling. Moreover, the women in manual work scored higher on the exhaustion dimension of burnout than the women in lower-level non-manual work or self-employment did. Th ey also scored higher on cynicism and lower on the professional effi cacy dimension than the women in non-manual work did. Especially among the men, the level of cynicism was higher among those unmarried than among those married. High job strain was associated with higher odds of having mild-to- severe burnout and a depressive disorder than low job strain was. Of these, the association with burnout was stronger, and it remained after adjustment for the indicators of mental health. In contrast, the associations between job strain and depressive disorders disappeared after adjustment for burnout. Unlike depressive disorders, burnout was also related to active and passive work after adjustments. Burnout and all of its three sub-dimensions were associated with depressive disorders for both genders. Forty-fi ve per cent of the employees with severe burnout also had a depressive disorder; when minor depressive disorder was also included among the depressive disorders, the

74 5. DISCUSSION

corresponding percentage was 53%. Both severe and mild burnout was related to major depressive disorder and dysthymic disorder, while mild burnout was also related to minor depressive disorder. Th e odds for any depressive disorder among the employees with mild burnout were higher among the men than among the women. Of the employees with severe burnout, 21% suff ered from an anxiety disorder, and 10% had alcohol dependence. Th e associations between burnout and anxiety disorders and alcohol dependence remained for both genders after adjustment for socio-demographic factors. Burnout and all of its three sub-dimensions were related to musculoskeletal disorders and cardiovascular diseases in the total sample. Among the participants with severe burnout, a musculoskeletal disorder was present in 47% of the cases, and a cardiovascular disease occurred in 28% of the cases. After adjustment for socio-demographic factors, the physical strenuousness of work, health behaviour, and depressive symptoms, burnout was associated with cardiovascular diseases only among the men and with musculoskeletal disorders only among the women. Ten per cent of those with severe burnout had no co-occurring disorder or illness. Th e odds of having at least one long medically certifi ed sickness absence during a 2-year period were higher for the employees with burnout than for their colleagues free of burnout. For the men who had at least one long absence in 2 years, severe burnout was associated with 55 compensated sickness absence days in addition to 9 days' qualifying period. For the women, the corresponding number was 41 days, but it did not reach statistical signifi cance. For both genders, severe burnout was associated with sickness absence even after adjustment for co-occurring common mental disorders and physical illnesses. Among the women, mild burnout was also independently associated with sickness absence.

5.2 Occupational burnout in the arena of occupational health psychology

Burnout and ill health

Depressive and anxiety disorders Th e observed positive association between burnout and depressive disorders in this population-based sample supports the concept that burnout

75 5. DISCUSSION

and depression are associated; they seem to co-occur at a high rate. Th is fi nding is in agreement with the numerous positive associations found earlier between burnout and depressive symptoms in diff erent kinds of samples (Meier 1984, Firth et al. 1986, Landsbergis 1988, Seidman & Zager 1991, Glass et al. 1993, McKnight & Glass 1995, Baba et al. 1999, Sears et al. 2000, Roelofs et al. 2005, Lindblom et al. 2006). Furthermore, the present results are in line with the fi ndings of two earlier studies that explored the associations between burnout and depressive disorders in selected samples (Belcastro & Hayes 1984, Kinnunen et al. 2004). Mild and severe levels of the three dimensions of burnout (i.e., exhaustion, cynicism, and diminished professional effi cacy) were all associated with an increased probability of depressive disorders and, therefore, supported the syndrome quality of burnout. Th ese associations are in line with the results of six earlier studies (Firth et al. 1986, Landsbergis 1988, McKnight & Glass 1995, Sears et al. 2000, Roelofs et al. 2005, Lindblom et al. 2006), but are contrary to the results of four studies concerning human service work in which only some of the burnout dimensions were related to depressive symptoms. Two small-scale studies found only the exhaustion dimension, which has usually shown the strongest associations with depression (Firth et al. 1986, Landbergis 1988, McKnight & Glass 1995, Sears et al. 2000, Lindblom et al. 2006), to be related to depressive symptoms (Lemkau et al. 1988, Tselenis et al. 2001). Another study, which used a modifi ed version of the Maslach Burnout Inventory and a mean split of the dimensional burnout scores, found high levels of exhaustion and diminished professional effi cacy to be associated with depressive symptoms (Jayaratne et al. 1986). Furthermore, in logistic regression analyses, only exhaustion and depersonalization, categorized with tertiles, showed a signifi cant association with depressive symptoms, as assessed with the Depression Scale by the Center for Epidemiologic Studies (Martin et al. 1997). Th erefore, at least when considered in relation to depressive disorders, burnout could be considered a single work-related construct, even though consisting of three dimensions. Th e present study confi rmed a positive association also between burnout and anxiety disorders. Such an association has been suggested earlier by fi ve studies showing a relationship between high levels of burnout syndrome or the three dimensions of burnout and a high level of state and trait anxiety among the Swedish working population (Lindblom et

76 5. DISCUSSION

al. 2006), in samples of human service work (Richardsen et al. 1992, Corrigan et al. 1995, Turnipseed 1998), and among patients with work- related problems (Roelofs et al. 2005). Depressive and anxiety disorders are assumed to result from complex interlinked genetic, temperamental, and past and present environmental infl uences (Kendler et al. 1993, 2002 & 2006, Merikangas 2005, Caspi & Moffi tt 2006). In the psychiatric literature, common risk factors for these disorders have included, for example, female gender, low socioeconomic position and security, the presence of physical illnesses, previous mental disorders, and a low level of social support (Salokangas & Pou- tanen 1998, Gruenberg & Goldstein 2003, Merikangas 2005, Rihmer & Angst 2005). Stressors, such as work-related life events, have also been found to precede the onset of depressive and anxiety disorders (Kessler 1997, Tennant 2001, Standsfeld 2002, Leskelä et al. 2004, Wilhelm et al. 2004, Standfeld & Candy 2006). It is possible that burnout is a part of the process leading to the development of depression or anxiety in situations in which the predisposing stressor is work-related (Greenglass & Burke 1990, Golembiewski et al. 1992, Leiter & Durup 1994, Iacovides et al. 1999, Bakker et al. 2000a, Hillhouse et al. 2000, Standsfeld 2002, Iacovides et al. 2003). Accord- ing to a recent hypothesis, adverse psychosocial work characteristics may aff ect mental health through a psychological pathway of injured self-esteem via erosion in the feelings of mastery in an adverse work situation (Standsfeld & Candy 2006). Burnout could partly indicate this erosion of mastery and therefore be associated with depressive or anxiety disorders. Erosion in the feelings of mastery over the work situation closely resembles the dimension of diminished professional effi cacy in the three- dimensional defi nition of burnout (Maslach & Jackson 1996). However, the relationship between burnout and depressive or anxiety disorders could also be reversed. Employees with mental health problems may face diffi culties in meeting their job demands or drift into jobs with low resources, or they may artifi cially perceive their work situation more negatively (de Lange et al. 2005) and thus show higher levels of burnout. Employees with various chronic illnesses, including depressive and anxiety disorders, reported higher levels of exhaustion in a cross-sectional study among university employees (Donders et al. 2007). Support for bi-directional associations between low mental health and negatively

77 5. DISCUSSION

perceived work characteristics have been found in prospective studies (de Jonge et al. 2001, de Lange et al. 2004). However, the pathway from health to work was less prominent than the one from work to health in these studies. Previous ﬁ ndings on the temporal order between burnout and depressive symptoms are mixed. One longitudinal study supported prospective associations from the burnout dimensions to depressive symptoms (Greenglass & Burke 1990), another from depressive symptoms to exhaustion (McCranie & Bransma 1988), while a third failed to ﬁ nd support for either of these sequences. In the last study (McKnight & Glass 1995), burnout and depression were suggested to develop "in tandem". In a recent prospective study, in which well-being at baseline was taken into account, both of the sequences, from burnout to depression and from depression to burnout, emerged (Ahola & Hakanen 2007), supporting a bi-directional association also between burnout and mental health.

Alcohol dependence Th e present results on mental disorders supported a positive association between burnout and alcohol dependence. Th is association is in agreement with two previous cross-sectional studies that found evidence indicating that work-related exhaustion is related to the screened risk for alcohol dependence (Cunradi et al. 2003, Winwood et al. 2003). Alcohol dependence is a mental disorder, which is assumed to develop as a consequence of a complex interplay between individual vulnerability, environmental risk factors, and pathological neural learning during repetitive use (Kaprio et al. 1987, Heath & Nelson 2002, Hyman 2005, Schuckit 2005). Individual vulnerability for alcohol problems is thought to include a familiar risk of the sensitized eﬀ ects of alcohol use, the positive expectancy of the eﬀ ects of alcohol, and the low personal and social resources for responding adaptively to stressors (Cooper et al. 1990 & 1992, Zimmermann et al. 2004). Could a stressful work environment be an underlying factor for the observed association between burnout and alcohol dependence? Prospect- ive associations have been found between adverse psychosocial work characteristics and an elevated risk for alcohol dependence, and these associations indicate that work stress is one possible environmental risk

78 5. DISCUSSION

factor for this disorder. In the North American Epidemiologic Catchment Area Program, high job strain was found to be a signifi cant predictor of alcohol use disorders among men (Crum et al. 1995). Th e Whitehall II study among British civil servants showed that psychosocial work characteristics predisposed both men and women to a risk of alcohol dependence; eff ort-reward imbalance was a crucial factor for the men and low decision latitude was important for the women (Head et al. 2004). Th ese results on work stress as an environmental risk factor for alcohol dependence are in accordance with the present result of a positive association between burnout, a possible consequence of chronic work stress, and alcohol dependence. However, as the development of burnout is related to a self-perpetuating process due to inadequate coping strategies (Schaufeli & Enzmann 1998), it is also possible that employees with low personal and social resources are vulnerable to both burnout and addictive behaviour in demanding work situations. Indeed, inadequate coping strategies have been suggested to be an essential element behind both burnout (Schaufeli & Enzmann 1998) and serious alcohol problems (Cooper et al. 1992). Th e association between burnout and alcohol dependence can also derive from a connection between stress, craving for stress relief, and addictive behaviour on a neurobiological level (Brady & Sonne 1999). It has been theoretically proposed that stressful experiences and addictive behaviour may share common neural pathways involving dopamine receptors in the limbic system. Stress from disappointed reward expectancies due to unfavourable social exchange has been suggested to trigger neuro-regulatory dysfunction in reward-sensitive brain areas and further mitigate compensatory addictive behaviour (Siegrist 2000 & 2002, Hyman 2005). Th is process may be especially pronounced among persons with a familiar vulnerability for stronger dampening eff ects of alcohol in stress situations (Zimmermann et al. 2004). In this way, socio-environmental and personal factors can interact in shaping the complex patterns behind alcohol dependence at the neural level (Heath & Nelson 2002).

Musculoskeletal disorders Th e robust positive association in the present study between burnout and musculoskeletal disorders independently of socio-demographic factors,

79 5. DISCUSSION

the physical strenuousness of work, and health behaviour is in line with the associations between burnout and musculoskeletal pain reported in two earlier studies (Soares & Jablonska 2004, Miranda et al. 2005). Musculoskeletal disorders are multi-factorial in nature; individual, environmental, and psychosocial factors aff ect their development (Hagberg et al. 1995). Exposure to work-related biomechanical and psychosocial factors has been found to be associated with musculoskeletal disorders (Bongers et al. 1993, Hagberg et al. 1995, Punnet & Herbert 2000, National Research Council and the Institute of Medicine 2001). In addition to direct pathways between work characteristics and musculoskeletal disorders, including repetitive work aff ecting muscles and joints and work stress increasing muscle tension, experienced strain has also been shown to mediate between work characteristics and the interpretation and reporting of musculoskeletal problems (Bongers et al. 1993, 2002, Ari ns et al. 2001, National Research Council and the Institute of Medi- cine 2001, Lundberg et al. 2002, Larsman et al. 2006). Th e direct and indirect links between work characteristics and musculoskeletal disorders may also explain the observed association between burnout and these disorders. On the other hand, the association may also be reversed so that a musculoskeletal disorder and a related decrease in one's work ability may have a negative eff ect on the capability of a worker to cope with his or her previous job demands and thus increase the risk of burnout (Hallman et al. 2003, Donders et al. 2007).

Cardiovascular diseases Th e positive unadjusted association between burnout and clinically determined cardiovascular diseases is in agreement with similar results on burnout and self-reported cardiovascular disease in previous studies (Belcastro 1982, Hallman et al. 2003). In addition, it is consistent with earlier studies reporting associations between exhaustion and coronary risk factors (Melamed et al. 2006), such as increased levels of total cholesterol, low-density lipoprotein cholesterol, triglycerides, and uric acid, as well as elevated diastolic blood pressure (Melamed et al. 1992) and inﬂ ammatory biomarkers (Grossi et al. 2003). Although chronic stress has also been associated with an increased secretion of the stress hormone cortisol at awakening, the results on burnout or exhaustion

80 5. DISCUSSION

and the levels of cortisol are mixed (Melamed et al. 1999, Pruessner et al. 1999, De Vente et al. 2003, Moch et al. 2003, Grossi et al. 2003 & 2005, Langelaan et al. 2006, Mommersteeg et al. 2006a, 2006b, 2006c). Th e associations between exhaustion and coronary risk factors indicate that burnout probably precedes cardiovascular disease rather than being a consequence of it. Th e observed association between burnout and cardiovascular diseases is in line with the preveious fi ndings showing a prospective association between work stress and cardiovascular disease. (For a meta-analysis of prospective cohort studies see Kivimäki et al. 2006b.) Th e adverse effect of sustained work stress on cardiovascular disease may be accounted for by several mechanisms (McEwen & Stellar 1993, Maier & Watkins 1998, Kiecolt-Glaser et al. 2002, Siegrist 2002, Epel et al. 2004, 2006, Chandola et al. 2006). Th e prolonged hyperactivity of the physiological stress mechanisms (i.e., the sympathetic-adrenergic-medullar system and the hypothalamic-pituitary-adrenal system), which slow the organism's basic reparatory functions in order to reallocate energy and prepare the body to meet the increased demands of the perceived situation (Frankenhauser 1989, Brunner 1997), may predispose a person to cardiovascular disease through the wear and tear of the organism (i.e., accumulation of the so-called allostatic load) (McEwen & Stellar 1993, McEwen 1998a & 1998b). Psychosocial work characteristics can also aff ect cardiovascular disease through negative emotions (Musselman et al. 1998, Kiecolt-Glaser et al. 2002, Belkic et al. 2004, Everson-Rose & Lewis 2005, Suls & Bunde 2005). In addition, chronic work stress may infl uence cardiovascular disease indirectly through an unhealthy life style, often associated with chronic strain, in particular low leisure- time physical activity, overweight, cigarette smoking, and heavy alcohol consumption, or their co-manifestation (Kouvonen et al. 2005, Siegrist & Rödel 2006, Kouvonen et al. 2007).

Summary It has previously been suggested that burnout is related to ill health. Th e present results from a representative sample of the Finnish working population over 30 years of age strongly support this hypothesis. Burnout showed a marked overlap with mental disorders and physical illnesses.

81 5. DISCUSSION

Because work stress has previously been shown to predispose people to depressive and anxiety disorders, alcohol dependence, musculoskeletal problems, and cardiovascular disease, it is plausible that burnout, as a consequence of chronic work stress, is related to these disorders. On the other hand, ill health has been shown to be related to exhaustion and a negative evaluation of work characteristics. Th erefore, it is also possible that health problems predispose a person to burnout.

Distinction between burnout and ill health

Association with work characteristics Th e consistent correlations between burnout and depressive symptoms (Meier 1984, Firth et al. 1986, Landsbergis 1988, Seidman & Zager 1991, Glass et al. 1993, McKnight & Glass 1995, Baba et al. 1999, Sears et al. 2000, Roelofs et al. 2005, Lindblom et al. 2006) have raised the question of whether burnout is conceptually redundant to depression. Th is study provided an opportunity to compare the work-related associations between burnout and depression and to evaluate the possible redundancy of these concepts in relation to one health-related aspect of the psychosocial work environment, the job strain. In this study, work stress was operationalized according to the job strain model, which has previously successfully predicted various manifestations of ill health (van der Doef & Maes 1999, de Lange et al. 2003, Kivimäki et al. 2006b, Standsfeld & Candy 2006). High job strain, when compared with low job strain, was related to burnout. Th is fi nding strengthens the previous evidence on the association between job strain and burnout, which has, to date, mainly been based on fi ndings concerning exhaustion (Raff erty 1987, Landsbergis 1988, Melamed et al. 1991, Bourbonnais et al. 1998, 1999). Th e association between high job strain and burnout was independent of the indicators of mental health. As a new contribution of the present study, the association between high job strain and depressive disorders was found to be fully dependent on burnout. In other words, after burnout was taken into account, there was no direct relationship between job strain and depressive disorders. Causal chains such as mediated eff ects cannot be inferred from cross- sectional observational epidemiological data. However, such data can be

82 5. DISCUSSION

used to test whether the observed associations are consistent with what one would expect if the path from job strain to burnout to depression were causal. In the present study, both of the operational criteria that would dem- onstrate a mediated eff ect of burnout between job strain and depressive disorders were met (Kenny 2005). Firstly, an association between burnout and depression was documented. Secondly, job strain was shown to be associated with depression, but only when burnout was not accounted for; the association between job strain and depressive disorders disappeared after adjustment for burnout. Th is kind of mediational eff ect of burnout between job strain and depressive symptoms was recently demonstrated also in a 3-year prospective study among dentists (Ahola & Hakanen 2007). Burnout has further been shown to mediate between work characteristics and health in studies in which health was indicated by sickness absence (Bakker et al. 2003) or psychosomatic complaints (Schaufeli & Bakker 2004). However, these elaborations need to be verifi ed with longitudinal data on depressive disorders among various occupations to establish the direction of the associations and true causality between work, burnout, and depression. Th e job strain model postulates that high job strain is a risk factor for mental strain and health problems, while active work with high demands and control is hypothesized to lead to average strain but also to favour- able consequences, for example, work motivation and the development of new behaviour patterns (Karasel & Th eorell 1990). When compared with low job strain in the present study, active work and passive work were related to an increased probablity for burnout after adjustment for mental health. Th ese associations are in accordance with the fi ndings of previous studies on burnout (Landbergis 1988) and exhaustion (Kaup- pinen-Toropainen et al. 1983, Bourbonnais et al. 1998, 1999). Th e observed probablities related to active and passive work in the present study were smaller than the one related to high-strain work. Correspond- ing associations were not found between active and passive work and depressive disorders. Th ese results strengthen the proposed diff erential associations with work characteristics for burnout and depression (Warr 1987, Maslach et al. 2001).

83 5. DISCUSSION

Association with work disability Th e observed positive association between burnout and long medically certifi ed sickness absence is in line with earlier results that have shown high work-related exhaustion to be associated with an increased prevalence of self-reported sickness absence in a Swedish working population (Hallsten et al. 2002), human service work (Borritz et al. 2006), and transportation work (Cunradi et al. 2005) and burnout to predict increased company-registered (Bakker et al. 2003) and medically certifi ed absences (Toppinen-Tanner et al. 2005). As an original contribution of the present study, it was shown that severe burnout had a positive association with long medically certifi ed sickness absence independently of co-existing common mental disorders and physical illnesses in a nationally representative sample. Th e controlled disorders and illnesses included depressive, anxiety, and alcohol use disorders, and musculoskeletal disorders, cardiovascular and respiratory diseases, and a heterogeneous group of other physical illnesses, among which are the leading causes of compensated sickness absences in Finland (Th e Social Insurance Institution of Finland 2005). Because long sickness absences have been found to predict future disability pensioning (Kivimäki et al. 2004, Lund et al. 2007), this result suggests that burnout may make a distinctive contribution to work disability when compared with ill health and therefore further supports the independent status of burnout in the arena of occupational health psychology.

Summary Despite the strong associations between burnout and mental disorders and physical illnesses, several fi ndings suggest that burnout may be partly distinct from depressive disorders and from ill health in general. Compared with depressive disorders, burnout was more strongly related to high job strain, and it was related to active and passive work after adjustments. Th e results were also consistent with the possibility that burnout would mediate the eff ects of job strain on depressive disorders. Furthermore, burnout had an independent contribution to work disability even when the eff ects of the co-occurring common mental disorders and physical illnesses were taken into account.

84 5. DISCUSSION

Role of individual factors in burnout

Gender and the level of burnout No gender diff erences were found for the level of burnout syndrome. Th is fi nding runs contrary to earlier results from population-based studies showing women to experience a slightly higher level of burnout than men (Kalimo & Toppinen 1997) or to be over-represented among employees with a high level of burnout (Lindblom et al. 2006). How- ever, women experienced a slightly higher level of exhaustion, and this fi nding replicated the results of previous studies in human service work (Maslach & Jackson 1981, Töyry 2005). It has been suggested that the possible gender diff erence in burnout may refl ect diff erences in roles and occupations (Schaufeli & Green- glass 2001). However, diff erential exposure has generally accounted only minimally for women's higher distress or lower health (Roxburgh 1996, Denton et al. 2004). In addition to having generally less control at work, women have been found to be more vulnerable to low job control than men (Rugulies et al. 2006); this fi nding supports the diff erential vulnerability to adverse conditions (Denton et al. 2004) and, especially, to low job control. In the present study, the women tended to show a slightly higher level of burnout when they had a low level of education or were in manual work than other women did. Parallel results of women with a low level of education showing worse health, indicated by self-reported general and mental health, fatigue, and musculoskeletal symptoms, than highly educated women was reported in a recent Swedish study (Dahlberg 2005). Vulnerability to low control may explain the association between a low level of education or occupational status and a higher level of burnout among women, because these socio-demographic factors are strong correlates of job control. An alternative explanation for the slightly higher level of burnout among women in manual work is a gender diff erence in the process of worklife exit among men and women. In Finland, women have reached the statutory retirement age more probably than men, who have retired more often earlier on disability or unemployment pensions (Hakola 2000a & 2000 b). Th e healthy worker eff ect has been shown to operate strongly among men in manual occupations (Bartley & Owen 1996). It

85 5. DISCUSSION

seems that men have to be healthy to remain employed in manual jobs. Th is assumption may apply to a less extent to women, who can continue to work with less selection and therefore be an unhealthier sub-population in their senior years than men are (Hakola 2000a). Th e present result that those married had less burnout than those not married recognizes social support as a general health resource (Broad- head et al. 1983, House et al. 1988). Signs of the protective eff ect of marital status have earlier been reported for exhaustion (Hallsten et al. 2002). However, when the analyses were segregated by gender, the positive eff ect of marital status on burnout applied only to the men; the married or co-habiting men showed less cynicism than the men who were single, divorced, or widowed. Compared with women, men have been reported to seek less social support when coping with job loss or depression (Bebbington 1987, Salokangas et al. 1988, Leana & Feldman 1991). As being married has been found to be benefi cial especially to men's health (House et al. 1988), it is possible that the family provides men an important source of social support and therefore protects them against burnout. In light of the present diff erent results among men and women, the earlier mixed results on socio-demographic factors and burnout may be due to unsegregated analyses by gender.

Gender differences in the associations between burnout and ill health Th e men with mild burnout in the present study had higher odds of having a depressive disorder than the women with a similar level of burnout. At least two issues may explain this ﬁ nding. Firstly, the importance of work-related problems in the aetiology of mental disorders may be greater among men than women. Work may be a more dominant arena for men, whereas, for women, the aetiological factors of mental health may be distributed across several spheres of life, including domestic factors and social relations. Supporting this assumption, the relationship between work stress and the use of antidepressant medication was found only among men in a Finnish population study (Virtanen et al. 2007). Th is assumption is also in line with studies showing unemployment to be a greater health risk for men than for women (Artazcoz et

86 5. DISCUSSION

al. 2004, Cooper et al. 2006). Furthermore, it was shown in a cross- sectional study among white-collar workers that men's general distress was mostly related to work conditions, while women's level of distress was determined by the interaction between conditions at work and at home (Krantz et al. 2005). Family structure and social relations were found to be more important determinants of women's than men's health in the large Canadian National Population Health Survey (Denton et al. 2004). Secondly, as elaborated by Hensing et al. (1996), the stigma and social consequences of having a psychiatric disorder may be worse at work for men than for women due to cultural role expectations. Because men are not supposed to show weakness, the consequences for the men suff ering from mild burnout may be more severe than for the women. In general, men have been found to use more alcohol and suff er more often from alcohol dependence than women (Grant et al. 2004, Pirkola et al. 2005). Th is fi nding has been explained by a gender diff erence in social and cultural behaviour regarding alcohol use (Holmila & Raitasalo 2005). In the present study, the association between burnout and alcohol dependence was similar and of about the same magnitude for both genders. Th is result may partly refl ect the changes in drinking habits among women during the past several years, including an increase in alcohol use (Grant et al. 2004, Raitasalo & Maaniemi 2007). Similar results for men and women in this study may also indicate that, in the relationship between burnout and serious alcohol problems, the signifi cance of individual factors outweigh the social and cultural behavioural models. As a new contribution of this study, burnout was found to be related to musculoskeletal disorders only among women when depression was taken into account. Musculoskeletal disorders and many of their risk factors have generally been more prevalent among women, and the prevalence suggests diff erential exposure (Punnet & Herbert 2000, de Zwart et al. 2001). Women may also have a special vulnerability to physical strain (Punnet & Herbert 2000, de Zwart et al. 2001). In spite of possible gender diff erences, previous analyses on work characteristics and musculoskeletal problems have seldom been stratifi ed by gender. Th e relationship between burnout and musculoskeletal disorders needs to be explored further in order to increase the understanding of why the association between burnout and musculoskeletal disorders may be mood-related only among men.

87 5. DISCUSSION

Th is study revealed that the association between burnout and cardiovascular diseases among women may depend on standard health risk factors such as older age, a low level of education, a low occupational grade, unmarried status (Hemingway & Marmot 1999, Kivimäki et al. 2002, Everson-Rose & Lewis 2005), and physical strain at work. In contrast, the association between burnout and cardiovascular diseases was independent of these factors among the men. Although the formal test of gender diff erence in the association between job strain and cardiovascular diseases failed to reach signifi cance in a meta-analysis (Kivimäki et al. 2006b), the evidence on the associations between job strain and cardiovascular diseases have generally been stronger and more consistent among men and more sparse and less consistent among women (Belkic et al. 2004). Further supporting the gender diff erence, the association between burnout and cardiovascular biomarkers has also diff ered to some extent between the genders (Shirom et al. 1997, Toker et al. 2005). To understand how stress aff ects the development of cardiovascular diseases among women, it may also be necessary to take into account domestic strain, because women may still be more responsible for domestic duties than men (Väänänen et al. 2004). Th e combined exposure to a high amount of both paid work hours and household work hours was signifi cantly related to a high level of distress among women, whereas men's distress was related to high paid work hours and not to household work hours in a cross-sectional study among white-collar workers (Krantz et al. 2005). Indeed, a double exposure to stress from work and from marriage was associated with the highest risk for coronary disease among women in the Stockholm Female Coronary Risk Study (Orth-Gomér & Leineweber 2005). Furthermore, low control at home predicted cardiovascular diseases among women but not among men in the prospective Whitehall II study among British civil servants (Chandola et al. 2004). Th erefore, uncontrolled domestic strain may have imputed the association between burnout and cardiovascular diseases among the women in the present study. Among the women, mild burnout was also independently related to sickness absence in contrast to the situation among the men. Instead, only for the men was severe burnout related to delayed return to work. Similar results showing that women have more absences and men have longer absences have been obtained previously concerning psychiatric

88 5. DISCUSSION

disorders and sickness absence in a Swedish population-based register study (Hensing et al. 1996). In the NEMESIS study, mental disorders were more important predictors of sickness absence among men than among women (Laitinen-Krispijn & Bijl 2000). A study with a large random British sample indicated that men tend to seek any form of help for stress or strain to a less degree than women (Oliver et al. 2005). A similar trend emerged concerning seeking professional help for psychiatric problems in four large-scale North American surveys (Kessler et al. 1981) and in a random Canadian sample (Bland et al. 1997). It seems that women interpret feelings of distress as signs of problems needing help more likely than men (Kessler et al. 1981). Th erefore, women seek help more probably in distress and may get it in an earlier phase of the problematic situation than men do. If men miss professional advice and also obtain less social support, which among women can buﬀ er mental health problems during work stress (Stansfeld et al. 1997, Väänänen et al. 2003), the problems may accumulate and a longer absence would then be needed for men to recover.

Age and the level of burnout In contrast to what has been presented in two previous burnout reviews (Schaufeli & Enzmann 1998, Maslach et al. 2001), the level of burnout was not lower in the older age groups in the present sample of the Fin- nish working population aged 30 years or over. Instead, the prevalence of burnout seemed to slightly increase with age, although the absolute diff erences between the age groups were small. Corresponding fi ndings have been obtained from Swedish and previous Finnish population-based samples (Kalimo & Toppinen 1997, Lindblom et al. 2006). Population studies catch all workers in diff erent age groups and also those who might have changed jobs or occupations because of burnout. Th erefore, population studies may provide a more reliable view on the relationship between age and burnout than studies using non-random occupational or organizational samples. Th ere are at least two conceivable explanations for an age-related increase in burnout. Firstly, the exposure time models of stress state that the incidence of ill health increases with the duration of the stressor exposure (Zapf et al. 1996). In line with these models, it has been

89 5. DISCUSSION

shown that prolonged exposure to poor psychosocial factors increases the probability of reduced well-being and the risk of disease (Kivimäki et al. 2002, Kalimo et al. 2003, Kivimäki et al. 2006a, Rugulies et al. 2006). It has also been emphasized that burnout takes time to develop (Schaufeli & Enzmann 1998, Maslach et al. 2001) and that the symptoms are persistent over time (McKnight & Glass 1995, Bakker et al. 2000b, Taris et al. 2005). Secondly, the change in worklife in Western countries has been rapid during the past decade. Earlier, age with accumulating work experience usually benefi tted the worker, raising his or her status and profi ciency. In today's continuously changing worklife, the growing demands for learning and fl exibility may become a burden particularly for ageing workers who, on average, have less education than younger workers. However, the situation of young employees needs to be investigated in more detail in future studies because employees under 30 years of age were not included in the medical examination of the Health 2000 Study, which determined the inclusion criteria in the present study. In a previous Finnish population study on burnout, which included employees from 24 years of age on, no diff erence in the prevalence of burnout was detected between the age groups of 24–34 years and 35–44 years (Kalimo & Toppinen 1997). However, it is also possible that age and accompanying work experience are benefi cial in some sectors of the labour market. Th e negative association between age and burnout reported in burnout reviews (Schaufeli & Enzmann 1998, Maslach et al. 2001) is based on research done in human service work (Maslach & Jackson 1981, Mor & Laliberte 1984, Birch et al. 1986, Huberty & Huebner 1988, Rogers & Dodson 1988, Poulin & Walter 1993, Vredenburgh et al. 1999) and was also found for a nationally representative sample of Finnish physicians (Töyry 2005). Th e experienced human service workers may have learned special professional and coping skills that protected them especially against burnout in demanding non-reciprocal interactions. Th is issue should be explored further in the future by comparing human service workers with employees in non-human service professions in population-based samples.

Summary Gender was not related to the level of burnout, but the associations between burnout and other socio-demographic factors and ill health

90 5. DISCUSSION

diff ered to some extent for the men and women. Especially among the men, burnout was related to being unmarried, having a cardiovascular disease, and being longer on sick leave. Especially among the women, burnout was related to low-quality jobs and having a musculoskeletal disorder. Th ese gender diff erences may partly explain earlier mixed results concerning burnout and these factors in samples combining men and women. Th e positive association between age and the level of burnout in a working population may refl ect the accumulated eff ects of prolonged work stress over time.

5.3 Evaluation of the study

Assessment of burnout Th e general version, the MBI-GS (Schaufeli et al. 1996), of the most widely used and most extensively validated burnout instrument, the MBI (Maslach & Jackson 1996), was chosen for the assessment of burnout in the Health 2000 Study because it is widely regarded as a gold standard for measuring burnout (Schaufeli & Enzmann 1998, Schaufeli & Taris 2005). Satisfactory reliability and validity has been confi rmed for the MBI-GS in diff erent occupations (Leiter & Schaufeli 1996, Taris et al. 1999, Schutte et al. 2000, Bakker et al. 2002). Th e nationally established procedure used in the present study to form a weighted sum score for the burnout syndrome and to categorize the burnout sum score in three levels on the basis of the frequency of the symptoms (Kalimo et al. 2006) has been published in a peer-reviewed scientifi c journal (Kalimo et al. 2003), but its clinical validity has not been confi rmed. Th e results of the present study are dependent on the three-dimensional conceptualization of burnout, which has been criticized for its vague theoretical basis and a circular argument between the defi nition of burnout and the related measure of burnout (Shirom 2003, Kristensen et al. 2005). Th e associations between the three-dimensional operation- alization of burnout and health may be stronger than the ones obtained between exhaustion and health, because the three-dimensional burnout syndrome has been proposed to be a possible consequence of prolonged exhaustion in the stress process. Th e available alternatives for assessing

91 5. DISCUSSION

burnout are not free of defi ciencies (Schaufeli & Taris 2005) and are much less evaluated and known. In the future, burnout research would benefi t from reaching a shared consensus defi nition for burnout and from validated multi-method practice guidelines for its assessment.

Major strengths Th is study employed a large population-based sample that represented the 30- to 64-year-old working population of Finland, including also employees who were on sick leave or leave of absence. Th is sample gives a unique quality to the present study as, to our knowledge, no previous fi ndings on burnout and its correlates from nationally representative samples have been published in international peer-reviewed journals. Th e participation rate in the data collection of the present study was high; at least 80% of the working-age participants were covered. Th e use of weighting adjustment and sampling parameters in the analyses enabled the complex sample survey design and diff erences in sample attrition between subgroups to be accounted for. Th erefore, the acquired results concerning burnout can be generalized to worklife, including all occupational branches in the age range of the participants. Another major strength of the present study was its assessment of the health and work disability of the participants. Th e assessment of mental disorders was based on a fully standardized diagnostic interview, the CIDI, which provided the estimates for the diagnoses of DSM-IV mental disorders (Andrews & Peters 1998, Wittchen et al. 1998). Th e CIDI has been shown to be a valid method for assessing common mental non-psychotic disorders among primary care attendees (Jordanova et al. 2004) but it has not been validated in general populations. In a community setting, the depression module of the CIDI has been found to over-estimate prevalence rates slightly (Kurdyak & Gnam 2005). Th is over-estimation may have led to a slight infl ation of the associations between burnout and depressive disorders and over-adjustment in the observed relationship between burnout and work disability. Th e assessment of physical illnesses was based on a comprehensive clinical health examination including a symptom interview, a history, and a medical examination by a research physician. Work disability was indicated by medically certifi ed sickness absences for over 9 consecutive work days (Kivimäki et al. 2004, Lund et al. 2007),

92 5. DISCUSSION

which were extracted from a register of Th e Social Insurance Institution of Finland covering all Finnish citizens. In the present study, the CIDI interview, the clinical health examination, and the use of absence records reduced the risks of common method bias, confounding by negative affectivity, selective record bias, and other problems characterizing research with only self-reported data.

Study limitations Th e main limitation of the present study was its cross-sectional design, which does not allow for the ascertainment of temporal order in the associations between burnout and ill health. Th ese associations are open to reversed causality (i.e., the direction of the associations remains un- known). In addition, due to the cross-sectional design, it was not possible to diff erentiate between the actual development of burnout during the work career and a cohort eff ect. Th e cross-sectional design was especially problematic in the examination of the associations between job strain and burnout, both of which were self-assessed and therefore had an association open to the eff ects of common method variance (Lindell & Whitney 2001). In a meta-analytic review, burnout has been shown to be related to negative aff ectivity (i.e., a predisposition to experience negative emotions) (Watson & Clark 1984, Th orensen et al. 2003). Negative aff ectivity can colour all perceptions (Brief et al. 1988) and therefore may have artifi cially strengthened the observed relationship between job strain and burnout in the present study. However, the relationships between work stressors and experienced strain have tended to remain signifi cant after adjustment for negative aff ectivity (Parkes 1990, Chen & Spector 1991, Moyle 1995, de Jonge et al. 2001). Job strain was more strongly related to burnout than to depression even when depression was also self-assessed in a longitudinal design (Ahola & Hakanen 2007); this fi nding suggests that the stronger association between job strain and burnout than the one between job strain and depression would not solely be an artifi cial result of common method variance. However, longitudinal studies on work, burnout, and health are called for to clarify the complex temporal associations in the process of stress in the work context.

93 5. DISCUSSION

Th e sampling excluded those 18 to 29 years of age from the health examination for fi nancial reasons; too, chronic illnesses are usually less common among the young. It is possibile that this exclusion may have slightly aff ected the observed association between burnout and illnesses. Th erefore, the results can be most reliably generalized to employees over 30 years of age. Of the mental disorders, the three prevalent and relevant groups of disorders regarding working populations (i.e., depressive disorders, anxiety disorders, and alcohol dependence) (Sanderson & Andrews 2006, Honkonen et al. 2007) were analysed in relation to burnout. Of the physical illnesses, musculoskeletal and cardiovascular disorders were investigated in detail because they are among the leading causes for work disability in Finland (Th e Social Insurance Institution of Finland 2005). Th ese disorders and illnesses belonged also to those that have been shown previously to be related to exhaustion (Shirom et al. 2005). When sickness absence was explored, also alcohol abuse was included in the mental disorders, and respiratory diseases and a mixed group of physical illnesses were added to the physical illnesses used as confounding factors. Not all of the possible disorders and illnesses that the subjects might have had were recorded. Th ey include adjustment disorders and neuras- thenia, which may be used as a diagnosis in cases of burnout (Schaufeli et al. 2001, Glass et al. 2004), and personality disorders. Th erefore, the present study may present an underestimation in the overall prevalence of ill health in relation to burnout, as well as regarding the adjustment for sickness absence. In addition, personality, temperament, and other individual factors, including recent life events, were not controlled for, and it is not known whether these variables would have modifi ed the relationships between burnout and ill health. Th ese possible associations remain as an important subject to be examined in future studies.

5.4 Conclusions

Th e present ﬁ ndings concerning the Finnish working population suggest that occupational burnout is strongly related to ill health. Burnout and common mental disorders seem to co-occur among both men and women; the most substantial overlap was observed with depressive dis-

94 5. DISCUSSION

orders. Th ere is also common co-existence with burnout and physical illnesses. After possible confounders were taken into account, burnout was related to musculoskeletal disorders among the women and to cardiovascular diseases among the men. Because the design of this study was cross-sectional, no inferences can be drawn about the true temporal order between burnout and the manifestations of ill health. Burnout was also related to long medically certifi ed sickness absences. Severe burnout has an independent contribution to sickness absence, which was not totally accounted for by common mental disorders and physical illnesses. Especially among the men, the absences due to severe burnout were lengthy, lasting twice as long as those for physical illnesses. Th e association between burnout and depressive disorders, musculoskeletal disorders, and cardiovascular diseases was evident for all three dimensions of burnout; this fi nding suggests that the associations between burnout and ill health are not explained solely by the exhaustion dimension of burnout. Signifi cant associations concerning all three dimensions of burnout may justify the use of the three-dimensional burnout concept and a total burnout score. Despite the strong associations between burnout and disorders and illnesses, burnout does not seem to be totally redundant to ill health. Burnout was more strongly work-related than depressive disorders and severe burnout was related to work disability independently of common disorders and illnesses. Th e present data suggest that there are no substantial diff erences in the level of burnout between subgroups of socio-demographic factors. However, the associations between burnout and other socio-demographic factors diff ered to some extent between the genders. Burnout is also partly a diff erent kind of health risk for men and women. Among the men, mild burnout was a higher risk in relation to depressive disorders, and severe burnout was associated with delayed return to work, while, among the women, mild burnout was more probably related to absence from work. Although cohort eff ects cannot be excluded, the possible positive association between age and burnout suggests that conditions increasing the risk of burnout either accumulate over time or are more prevalent among older employees. Th e present fi ndings suggest that burnout may capture accumulating psychological work-related strain.

95 5. DISCUSSION

5.5 Policy implications

In order to promote health and prevent an early exit from the labour market, the work situation of clients should always be assessed in health care when working-age persons are being delt with. Burnout can be di- rectly screened when work-related problems are encountered in medical check-ups and in occupational health services. Severe burnout can be regarded as a proxy for health-impairing work stress, and therefore can serve as a risk marker for work disability. Many of the work-related variables associated with a high level of psychological ill health are potentially amenable to change (Michie & Williams 2003, Bourbonnais et al. 2006). Continuous improvement of psychosocial work characteristics could probably help prevent some burnout, and it could be facilitated by regular workplace surveys. In order to elicit the need for individual and occupational interventions, it may be benefi cial to use coding for burnout in addition to making a diagnosis, as a factor that aff ects health status (World Health Organiza- tion 1992) or a condition that may require clinical attention (Ameri- can Psychiatric Association 1994), in all health care contacts. Besides tailored individual interventions, a change in the work conditions in which burnout has evolved may promote long-lasting recovery from this stress-related condition (Bernier 1998, Weber & Jaekel-Reinhard 2000, Glass et al. 2004). A higher probability of burnout among ageing workers is challenging when contrasted with the national aims to lengthen the approximate work career of the population set in face of the ageing work force and the tendency towards early retirement in Finland and other European countries (Järvisalo et al. 2005, Ilmarinen 2006a & 2006b). In addition to ageing workers, special attention should be paid to women with low-quality jobs and men with work-related problems, since some associations with burnout were highlighted in these groups. It is likely that, to be eff ective, interventions need to be tailored according to the special characteristics of the target group.

96 REFERENCES

Ahola, K., Honkonen, T., Isometsä, E., Kalimo, R., Nykyri, E., Koskinen, S., Aromaa, A. & Lönnqvist, J. (2004) Työuupumus Suomessa – Terveys 2000 -tutkimuksen tuloksia [Burn- out in Finland – Results from the Finnish Health 2000 Study]. Suomen Lääkärilehti 59, 4109–4113. Ahola, K. & Hakanen, J. (2007) Job strain, burnout and depressive symptoms: a prospective study among dentists. Journal of Aﬀ ective Disorders, Epub ahead of print April 18. D0I: 10.1016/j.jad.2007.03.004. Allvin, M. & Aronsson, G. (2001) Th e future of work environment reforms: does the concept of Work Environment apply within the New Economy? In A. Th örnqvist (ed.) Work Life, Work Environment and Work Safety in Transition: Historical and Sociological Perspec- tives on their Development in Sweden during the 20th Century, pp. 237–249. Stockholm: National Institute for Working Life. Allvin, M., Aronsson, G., Hagström, T., Johansson, G. & Lundberg, U. (2006) Gränslöst arbete. Socialpsykologiska perspektiv på det nya arbetslivet [Groudless work – social psychology perspective on the new work life]. Malmö: Liber. American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington: American Psychiatric Association. Andrews, G. & Peters, L. (1998) Th e psychometric properties of the Composite International Diagnostic Interview. Social Psychiatry and Psychiatric Epidemiology 33, 80–88. Appels, A. & Schouten, E. (1991) Burnout as a risk factor for coronary heart disease. Behavioral Medicine 17, 53–59. Ariëns, G. A. M., van Mechelen, W., Bongers, P. M., Bouter, L. M. & van der Wal, G. (2001) Psychosocial risk factors for neck pain: a systematic review. American Journal of Industrial Medicine 39, 180–193. Aromaa, A. & Koskinen, S. (2004) Health and functional capacity in Finland. Baseline results of the Health 2000 Health Examination Survey. Publications of the National Public Health Institute B12/2004. Helsinki: Hakapaino Oy. Artazcoz, L., Benach, J., Borrell, C. & Cortés, I. (2004) Unemployment and mental health: understanding the interactions among gender, family roles, and social class. American Journal of Public Health 94, 82–88.

97 REFERENCES

Baba, V. V., Galperin, B. L. & Lituchy, T. R. (1999) Occupational mental health: a study of work-related depression among nurses in the Caribbean. International Journal of Nursing Studies 36, 163–169. Bakker, A. B., Schaufeli, W. B., Demerouti, E., Janssen, P. P., van der Hulst, R. & Brouwer, J. (2000a) Using equity theory to examine the diff erence between burnout and depression. Anxiety, Stress, and Coping 13, 247–268. Bakker, A. B., Schaufeli, W. B., Sixma, H. J., Bosveld, W. & van Dierendonck, D. (2000b) Patient demands, lack of reciprocity, and burnout: a fi ve-year longitudinal study among general practitioners. Journal of Organizational Behavior 21, 425–441. Bakker, A. B., Demerouti, E. & Schaufeli, W. B. (2002) Validation of the Maslach Burnout Inventory – General Survey: an Internet Study. Anxiety, Stress, and Coping 15, 245–260. Bakker, A. B., Demerouti, E., de Boer, E. & Schaufeli, W. B. (2003) Job demands and job resources as predictors of absence duration and frequency. Journal of Vocational Behavior 62, 341–356. Bartley, M. & Owen, C. (1996) Relation between socioeconomic status, employment, and health during economic change, 1973–93. British Medical Journal 313, 445–449. Bebbington, P. (1987) Marital status and depression: a study of English national admission statistics. Acta Psychiatrica Scandinavica 75, 640–650. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. & Erbaugh, J. (1961) An inventory for measuring depression. Archives of General Psychiatry 4, 561–571. Beck, A. T., Steer, R. A. & Garbin, M. G. (1988) Psychometric properties of the Beck depression Inventory: twenty years of evaluation. Clinical Psychology Review 8, 77–100. Belcastro, P. A. (1982) Burnout and its relationship to teachers' somatic complaints and illnesses. Psychological Reports 50, 1045–1046. Belcastro, P. A. & Hayes, L. C. (1984) Ergophilia...ergophobia...ergo...burnout? Professional Psychology: Research and Practice 15, 260–270. Belkic, K. L., Landsbergis, A., Schnall, P. L. & Baker, D. (2004) Is job strain a major source of cardiovascular disease risk? Scandinavian Journal of Work, Environment & Health 30, 85–128. Bernier, D. (1998) A study of coping: successful recovery from severe burnout and other reac- tions to severe work-related stress. Work & Stress 12, 50–65. Birch, N., Marchant, M. P. & Smith, N. M. (1986) Perceived role confl ict, role ambiguity, and reference librarian burnout in public libraries. Library and Information Science Research 8, 53–65. Bland, R. C., Newman, S. C. & Orn, H. (1997) Help seeking for psychiatric disorders. Canadian Journal of Psychiatry 42, 935–942. Bongers, P. M., de Winter, C. R., Kompier, M. A. J. & Hildebrandt, V. H. (1993) Psychosocial factors at work and musculoskeletal disease. Scandinavian Journal of Work, Environment & Health 19, 297–312.

98 REFERENCES

Bongers, P. M., Kremer, A. M. & ter Laak, J. (2002) Are psychosocial factors risk factors for symptoms and signs of the shoulder, elbow, or hand/wrist? A review of the epidemiological literature. American Journal of Industrial Medicine 41, 315–342. Borritz, M., Bültmann, U., Rugulies, R., Christensen, K.B., Villadsen, E. & Kristensen, T. (2005) Psychosocial work characteristics as predictors for burnout: fi ndings from 3-year follow up of the PUMA study. Journal of Occupational and Environmental Medicine 47, 1015–1025. Borritz, M., Rugulies, R., Christensen, K. B., Villadsen, E. & Kristensen, T. (2006) Burnout as a predictor of self-reported sickness absence among human service workers: prospective fi ndings from three year follow up of the PUMA study. Occupational and Environmental Medicine 63, 98–106. Bourbonnais, R., Comeau, M., Vézina, M. & Dion, G. (1998) Job strain, psychological distress and burnout in nurses. American Journal of Industrial Medicine 34, 20–28. Bourbonnais, R., Comeau, M. & Vézina, M. (1999) Job strain and evolution of mental health among nurses. Journal of Occupational Health Psychology 4, 95–107. Bourbonnais, R., Brisson, C., Vinet, A., Vézina, M., Abdous, B. & Gaudet, M. (2006) Eff ective- ness of a participative intervention on psychosocial work factors to prevent mental health problems in a hospital setting. Occupational and Environmental Medicine 63, 335–342. Brady, K. T. & Sonne, S. C. (1999) Th e role of stress in alcohol use, alcoholism treatment, and relapse. Alcohol Research & Health 23, 263–271. Brenninkmeijer, V., van Yperen, N. W. & Buunk, B. P. (2001) Burnout and depression are not identical twins: is decline of superiority a distinguishing feature? Personality and Individual Diff erences 30, 873–880. Brenninkmeijer, V. & van Yperen, N. W. (2003) How to conduct research on burnout: advant- ages and disadvantages of a unidimensional approach in burnout research. Occupational and Environmental Medicine 60, i16-i20. Brief, A. P., Burke, M. J., Robinson, B. S., George, J. M. & Webster, J. (1988) Should negative aff ectivity remain an unmeasured variable in the study of job stress? Journal of Applied Psychology 73, 193–198. Broadhead, W. E., Kaplan, B. H., James, S. A., Wagner, E. H., Schoenbach, W. J., Grimson, R., Heyden, S., Tibblin, G. & Gehlbach, S. H. (1983) Th e epidemiologic evidence for a relationship between social support and health. American Journal of Epidemiology 117, 521–537. Brunner, E. (1997) Stress and the biology of inequality. Socioeconomic determinants of health. British Medical Journal 314, 1472–1476. Buunk, B. P., Ybema, J. F., Gibbons, F. X. & Ipenburg, M. (2001) Th e aff ective consequences of social comparison as related to professional burnout and social comparison orientation. European Journal of Social Psychology 31, 337–351. Caplan, R., Cobb, S., French Jr, J. R. P., Van Harrison, R. & Pinneau Jr., S. R. (1975) Job demands and worker health. Main eff ects and occupational diff erences. Washington: National Institute for Occupational Safety and Health.

99 REFERENCES

Caspi, A. & Moffi tt, T. (2006) Gene-environment interaction in psychiatry: joining forces with neuroscience. Nature Reviews. Neuroscience 7, 583–590. Chandola, T., Kuper, H., Singh-Manoux, A., Bartley, M. & Marmot, M. (2004) Th e eff ect of control at home on CHD events in the Whitehall II study: gender diff erences in psychosocial domestic pathways to social inequalities in CHD. Social Science & Medicine 58, 1501–1509. Chandola, T., Brunner, E. & Marmot, M. (2006) Chronic stress at work and the metabolic syndrome: prospective study. British Medical Journal 332, 521–525. Chen, P. Y. & Spector, P. E. (1991) Negative aff ectivity as the underlying cause of correlations between stressors and strains. Journal of Applied Psychology 76, 398–407. Cloninger, C. R., Przybeck, T. R. & Svrakic, D. M. (1993) A systematic psychobiological model of temperament and character. Archives of General Psychiatry 50, 975–990. Cooper, M. L., Russell, M. & Frone, M. R. (1990) Work stress and alcohol eff ects: a test of stress-induced drinking. Journal of Health and Social Behavior 31, 260–276. Cooper, L. M., Russell, M., Skinner, J. B., Frone, M. R. & Mudar, P. (1992) Stress and alcohol use: moderating eff ects of gender, coping, and alcohol expectancies. Journal of Abnormal Psychology 101, 139–152. Cooper, D., McCausland, W. D. & Th eodossiou, I. (2006) Th e health hazards of unemployment and poor education: the socioeconomic determinants of health duration in the European Union. Economics and Human Biology 4, 273–297. Corrigan, P. W., Holmes, E. P. & Luchins, D. (1995) Burnout and collegial support in state psychiatric hospital staff . Journal of Clinical Psychology 51, 703–710. Cox, T., Tisserand, M. & Taris, T. (2005) Th e conceptualization and measurement of burnout: Questions and directions. Work & Stress 19, 187–191. Crum, R. M., Muntraner, C., Eaton, W. W. & Anthony, J. C. (1995) Occupational stress and the risk for alcohol abuse and dependence. Alcoholism: Clinical and Experimental Research 19, 647–655. Cunradi, C. B., Greiner, B. A., Ragland, D. R. & Fisher, J. M. (2003) Burnout and alcohol problems among urban transit operators in San Francisco. Addictive Behaviors 28, 91–109. Cunradi, C. B., Greiner, B. A., Ragland, D. R. & Fisher, J. (2005) Alcohol, stress-related factors, and short-term absenteeism among urban transit operators. Journal of Urban Health: Bulletin of the New York Academy of Medicine 82, 43–57. Dahlberg, R. (2005) Health and working conditions among low educated women [Doctoral dissertation]. Work and Health 2005:9. Stockholm: National Institute for Working Life. de Jonge, J., Dollard, M. F., Dormann, C., Le Black, P. M. & Houtman, I. L. D. (2000) Th e demand-control model: specifi c demands, specifi c control, and well-defi ned groups. Inter- national Journal of Stress Management 7, 269–287.

100 REFERENCES

de Jonge, J., Dormann, C., Janssen, P. P. M., Dolard, M. F., Landeweerd, J. A. & Nijhuis, F. J. N. (2001) Testing reciprocal relationships between job characteristics and psychological well-being: a cross-lagged structural equation model. Journal of Occupational and Organ- izational Psychology 74, 29–46. de Lange, A. H., Taris, T. W., Houtman, I. L. D., Kompier, M. A. J. & Bongers, P. M. (2003) "Th e very best of the millenium": longitudinal research and the demand-control-(support) model. Journal of Occupational Health Psychology 8, 282–305. de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I. L. D. & Bongers, P. M. (2004) Th e relationships between work characteristics and mental health: examining normal, reversed and reciprocal relationships in a 4-wave study. Work & Stress 18, 149–166. de Lange, A. H., Taris, T., Kompier, M. A. J., Houtman, I. L. D. & Bongers, P. M. (2005) Dif- ferent mechanisms to explain the reversed eff ects of mental health on work characteristics. Scandinavian Journal of Work, Environment & Health 31, 3–14. De Vente, W., Van Amsterdam, J. G. C., Kamphuis, J. H. & Emmelkamp, P. M. G. (2003) Physiological diff erences between burnout patients and healthy controls: blood pressure, heart rate, and cortisol responses. Occupational and Environmental Medicine 60, i54-i61. Demerouti, E., Bakker, A. B., Nachreiner, F. & Schaufeli, W. B. (2001) Th e job demands-resources model of burnout. Journal of Applied Psychology 86, 499–512. Denton, M., Prus, S. & Walters, V. (2004) Gender diff erences in health: a Canadian study of the psychosocial, structural and behavioral determinants of health. Social Science & Medicine 58, 2585–2600. de Zwart, B. C. H., Frings-Dresen, M. H. W. & Kilbom, Å. (2001) Gender diff erence in upper extremity musculoskeletal complaints in the working population. International Archives of Occupational and Environmental Health 74, 21–30. Donders, N. C. G. M., Roskes, K. & van der Gulden, J. W. J. (2007) Fatigue, emotional exhaustion and perceived health complaints associated with work-related characteristics in employees with and without chronic diseases. International Archives of Occupational and Environmental Health 80, 557–587. Epel, E. S., Blackburn, E. H., Lin, J., Dhabhar, F. S., Adler, N. E., Morrow, J. D. & Cawthon, R. M. (2004) Accelerated telomere shortening in response to life stress. Proceedings of the National Academy of Sciences of the United States of America 101, 17312–17315. Epel, E., S., Lin, J., Wilhelm, F. H., Wolkowitz, O. M., Cawthon, R., Adler, N. E., Dolbier, C., Mendes, W. B. & Blackburn, E. H. (2006) Cell aging in relation to stress arousal and cardiovascular disease risk factors. Psychoneuroendochrinology 31, 277–287. Everson-Rose, S. A. & Lewis, T. T. (2005) Psychosocial factors and cardiovascular diseases. An- nual Review of Public Health 26, 469–500. Feveile, H., Jensen, C. & Burr, H. (2002) Risk factors for neck-shoulder and wrist-hand symptoms in a 5-year follow-up study of 3,990 employees in Denmark. International Archives of Occupational and Environmental Health 75, 243–251. Firth, H., McIntee, J., McKeown, P. & Britton, P. (1986) Burnout and professional depression: related concepts? Journal of Advanced Nursing 11, 633–641.

101 REFERENCES

Firth, H., KcKeown, P., McIntee, J. & Britton, P. (1987) Professional depression, 'burnout' and personality in longstay nursing. International Journal of Nursing Studies 24, 227–237. Frankenhaeuser, M. (1989) A biopsychosocial approach to work life issues. International Journal of Health Services 19, 747–758. Freudenberger, H. J. (1974) Staff burn-out. Journal of Social Issues 30, 159–165. Glass, D. C., McKnight, J. D. & Valdimarsdottir, H. (1993) Depression, burnout, and perceptions of control in hospital nurses. Journal of Consulting and Clinical Psychology 61, 147–155. Glass, L. S., Blais, B. R., Genovese, E., Goertz, M. N., Harris, J. S., Hoff man, H. E., Klees, J. E. & Rowlett, C. D. (2004) Occupational Medicine Practice Guidelines. Evaluation and Management of Common Health Problems and Functional Recovery of Workers, Second Edition. Beverly Farms: OEM Press. Golembiewski, R. T., Munzenrider, R. & Carter, D. (1983) Phases of progressive burnout and their work site covariants: critical issues in OD research and praxis. Th e Journal of Applied Behavioral Science 19, 461–481. Golembiewski, R. T., Munzenrider, R. F., Scherb, K. & Billingsley, W. (1992) Burnout and psychiatric "cases": Early evidence of an association. Anxiety, Stress, and Coping 5, 69–78. Golembieski, R. T., Boudreau, R. A., Munzenrider, R. F. & Luo, H. (1996) Global burnout: a worldwide pandemic explored by the phase model. Greenwich: JAI Press. Grant, B. F., Dawson, D. A., Stinson, F. S., Chou, S. P., Dufour, M. C. & Pickering, R. P. (2004) Th e 12-month prevalence and trends in DSM-IV alcohol abuse and dependence: United States, 1991–1992 and 2001–2002. Drug and Alcohol Dependence 74, 223–234. Green, F. (2006) Demanding work. Th e paradox of job quality in the affl uent economy. Princeton (NJ): Princeton University Press. Greenglass, E. R. & Burke, R. J. (1990) Burnout over time. Journal of Health and Human Resources Administration 13, 192–204. Grossi, G., Perski, A., Evengård, B., Blomkvist, V. & Orth-Gomér, K. (2003) Physiological correlates of burnout among women. Journal of Psychosomatic Research 55, 309–316. Grossi, G., Perski, A., Ekstedt, M., Johansson, T., Lindström, M. & Holm, K. (2005) Th e morning salivary cortisol response in burnout. Journal of Psychosomatic Research 59, 103–111. Gruenberg, A. M. & Goldstein, R. D. (2003) Mood disorders: depression. In A. J. Tasman, J. J. Kay & J. A. Lieberman (eds.) Psychiatry, Second Edition, pp. 1207–1213. New York: John Wiley. Hagberg, M., Silverstein, B., Wells, R., Smith, R., Carayon, P., Hendrich, H., Perusse, M., Kuorinka, I. & Forbier, L. (1995) Work-related musculoskeletal disorders (WMSDs): a reference book for prevention. London: Taylor & Francis. Hagen, K. B., Tambs, K. & Bjerkedal, T. (2002) A prospective cohort study of risk factors for disability retirement because of back pain in the general working population. Spine 27, 1790–1796.

102 REFERENCES

Hakanen, J. (2004) Työuupumuksesta työn imuun: työhyvinvointitutkimuksen ytimessä ja reuna-alueilla. [From Burnout to Work Engagement: in the Core and on the Fringes of Work-related Well-being] [Doctoral dissertation, University of Helsinki]. People and Work Research Reports 27. Helsinki: Finnish Institute of Occupational Health. Hakola, T. (2000a) Varhaiseen eläkkeelle siirtymiseen vaikuttavat tekijät [Main Causes of the Finnish Early Retirement]. Publications of the Ministry of Social Aff airs and Health 2000:11. Helsinki: Edita. Hakola, T. (2000b) Navigating through the Finnish Pension System. Helsinki: Government Insitute for Economic Research. Halbesleben, J. R. B. & Demerouti, E. (2005) Th e construct validity of an alternative measure of burnout: investigating the English translation of the Oldenburg Burnout Inventory. Work & Stress 19, 208–220. Hallman, T., Th omsson, H., Burell, G., Lisspers, J. & Setterlind, S. (2003) Stress, burnout and coping: diff erences between women with coronary heart disease and healthy matched women. Journal of Health Psychology 8, 433–445. Hallsten, L. (1993) Burning out: a framework. In W. B. Schaufeli, C. Maslach & T. Marek (eds.) Professional Burnout: Recent Developments in Th eory and Research, pp. 95–114. London: Taylor & Francis. Hallsten, L., Bellagh, K. & Gustafsson, K. (2002) Utbränning i Sverige – en populationsstudie [Burning-out in Sweden – a Population Study]. Stockholm: Institute for Working Life. Hallsten, L., Josephson, M. & Torgén, M. (2005) Performance-based Self-esteem. A Driving Force in Burnout Processes and Its Assessment. Work and Health 2005:4. Stockholm: National Institute for Working Life. Head, J., Stansfeld, S. A. & Siegrist, J. (2004) Th e psychosocial work environment and alcohol dependence: a prospective study. Occupational and Environmental Medicine 61, 219–224. Heath, A. C. & Nelson, E. C. (2002) Eff ects of the interaction between genotype and environment. Research into the genetic epidemiology of alcohol dependence. Alcohol Research & Health 26, 193–201. Hemingway, H. & Marmot, M. (1999) Psychosocial factors in the aetiology and prognosis of coronary heart disease: systematic review of prospective cohort studies. British Medical Journal 318, 1460–1467. Hensing, G., Alexanderson, K., Allebeck, P. & Bjurulf, P. (1996) Sick-leave due to psychiatric disorder: higher incidence among women and longer duration for men. British Journal of Psychiatry 169, 740–746. Hillhouse, J. J., Adler, C. M. & Walters, D. N. (2000) A simple model of stress, burnout and symptomalogy in medical residents: a longitudinal study. Psychology, Health & Medicine 5, 63–73. Holmila, M. & Raitasalo, K. (2005) Gender diff erences in drinking: why do they still exist? Addiction 100, 1763–1769.

103 REFERENCES

Honkonen, T., Virtanen, M., Ahola, K., Kivimäki, M., Pirkola, S., Isometsä, E., Aromaa, A. & Lönnqvist, J. (2007) Employment status, mental disorders and service use in the working age population. Scandinavian Journal of Work, Environment & Health 33, 29–36. Hornby, A. S. (1982) Oxford Advanced Learner's Dictionary of Current English. Oxford: Oxford University Press. House, J. S., Landis, K. R. & Umberson, D. (1988) Social relationships and health. Science 241, 540–545. Huberty, T. J. & Huebner, E. S. (1988) A national survey of burnout among school psychologists. Psychology in Schools 25, 54–61. Hyman, S. E. (2005) Addiction: a disease of learning and memory. American Journal of Psy- chiatry 162, 1414–1422. Iacovides, A., Fountoulakis, K. N., Moysidou, C. & Ierodiakonou, C. (1999) Burnout in the nursing staff . Is there a relationship between depression and burnout? International Journal of Psychiatry in Medicine 29, 421–433. Iacovides, A., Fountoulakis, K. N., Kaprinis, S. & Kaprinis, G. (2003) Th e relationship between job stress, burnout and clinical depression. Journal of Aff ective Disorders 75, 209–221. Ilmarinen, J. (2006a) Onko Suomella varaa vanheta? [Can Finland aff ord to age?] Peole and Work 20, 3–4. Ilmarinen, J. (2006b) Towards a longer worklife: ageing and the quality of worklife in the Eu- ropean Union. Helsinki: Finnish Institute of Occupational Health and Ministry of Social Aff airs and Health. Jayaratne, S., Chess, W. A. & Kunkel, D. A. (1986) Burnout: its impact on child welfare workers and their spouses. Social Work 31, 53–59. Jordanova, V., Wickramesinghe, C., Gerada, C. & Prince, M. (2004) Validation of two survey diagnostic interviews among primary care attendees: a comparison of CIS-R and CIDI with SCAN ICD-10 diagnostic categories. Psychological Medicine 34, 1013–1024. Järvisalo, J., Raitasalo, R., Salminen, J. K., Klaukka, T. & Kinnunen, E. (2005) Depression and other mental disorders, sickness absenteeism and work disability pensions in Finland. In J. Järvisalo, B. Anderson, W. Doedeker, I. Houtman (eds.) Mental Disorders as a Major Challenge in Prevention of Work Disability: Experiences in Finland, Germany, the Nether- lands and Sweden, pp. 27–59. Social Security and Health Reports 66. Helsinki: Th e Social Insurance Institution of Finland. Kalimo, R. & Toppinen, S. (1997) Työuupumus Suomen työikäisellä väestöllä [Burnout among Finnish working population]. Helsinki: Finnish Institute of Occupational Health. Kalimo, R. (2000) Th e challenge of changing work and stress for human resources. Th e case of Finland. Journal of Tokyo Medical University 58, 349–356. Kalimo, R., Pahkin, K., Mutanen, P. & Toppinen-Tanner, S. (2003) Staying well or burning out at work: work characteristics and personal resources as long-term predictors. Work & Stress 17, 109–122.

104 REFERENCES

Kalimo, R., Hakanen, J. & Toppinen-Tanner, S. (2006) Maslachin yleinen työuupumuksen arviointimenetelmä MBI-GS [Th e Finnish version of the Maslach Burnout Inventory – General Survey]. Helsinki: Finnish Institute of Occupational Health. Kaprio, J., Koskenvuo, M., Langinvainio, H., Romanov, K., Sarna, S. & Rose, R. J. (1987) Genetic infl uence on use and abuse of alcohol: a study of 5638 adult Finnish twin brothers. Alcoholism: Clinical and Experimental Research 11, 349–56. Karasek, R. A. (1979) Job demands, job decision latitude and mental strain: implications for job redesign. Administrative Science Quarterly 24, 285–308. Karasek, R. & Th eorell, T. (1990) Stress, Productivity and Reconstruction of Working Life. New York: Basic Books. Karasek, R., Brisson, C., Kawakami, N., Houtman, I., Bongers, P. & Amick, B. (1998) Th e job content questionnaire (JCQ): an instrument for internationally comparative assess- ments of psychosocial work characteristics. Journal of Occupational Health Psychology 3, 322–355. Kauppinen-Toropainen, K., Kandolin, I. & Mutanen, P. (1983) Job dissatisfaction and work-related exhaustion in male and female work. Journal of Occupational Behaviour 4, 193–207. Keltikangas-Järvinen, L. & Heponiemi, T. (2004) Vital exhaustion, temperament, and cardiac reactivity in task-induced stress. Biological Psychology 65, 121–135. Kendler, K. S., Kessler, R. C., Neale, M., Heath, A. C. & Eaves, L. J. (1993) Th e prediction of major depression in women: toward an integrated etiologic model. American Journal of Psychiatry 150, 1139–1148. Kendler, K. S., Gardner, C. O. & Prescott, C. A. (2002) Toward a comprehensive developmental model for major depression in women. American Journal of Psychiatry 159, 1133–1145. Kendler, K. S., Gardner, C. O. & Prescott, C. A. (2006) Toward a comprehensive developmental model for major depression in men. American Journal of Psychiatry 163, 115–124. Kenny, D. A. (2005) Mediation. In B. S. Everitt & D. C. Howell (eds.) Encyclopedia of Statistics in Behavioral Science, Volume 3, pp. 1194–1198. Chichester: John Wiley. Kessler, R. C. (1997) Th e eff ects of stressful life events on depression. Annual Review of Psy- chology 48, 191–214. Kessler, R. C., Brown, R. L. & Roman, C. L. (1981) Gender diff erences in psychiatric help- seeking – evidence from four large-scale surveys. Journal of Health and Social Behavior 22, 49–64. Kiecolt-Glaser, J. K., McGuire, L., Robles, T. F. & Glaser, R. (2002) Emotions, morbidity, and mortality: new perspectives from psychoneuroimmunology. Annual Review of Psychology 53, 83–107. Kinnunen, U., Hätinen, M., Toskala, A., Männikkö, K., Pekkonen, M., Sörensén, L., Mauno, S., Aro, A. & Alén, M. (2004) Työuupumus: arviointi, kehittyminen ja intervention vaikut- tavuus [Burnout: Assessement, Development and Evaluation of an Invervention]. Reports from the Department of Psychology 350. Jyväskylä: University of Jyväskylä.

105 REFERENCES

Kivimäki, M., Leino-Arjas, P., Luukkonen, R., Riihimäki, H., Vahtera, J. & Kirjonen, J. (2002) Work stress and risk of cardiovascular mortality: prospective cohort study of industrial employees. British Medical Journal 325, 857. Kivimäki, M., Head, J., Ferrie, J. E., Shipley, M. J., Vahtera, J. & Marmot, M. G. (2003) Sick- ness absence as a global measure of health: evidence from mortality in the Whitehall II prospective cohort study. British Medical Journal 327, 364. Kivimäki, M., Forma, P., Wikström, J., Halmeenmäki, T., Pentti, J., Elovainio, M. & Vahtera, J. (2004) Sickness absence as a risk marker of future disability pensions: the 10-town study. Journal of Epidemiology and Community Health 58, 710–711. Kivimäki, M., Head, J., Ferrie, J. E., Brunner, E., Marmot, M., Vahtera, J. & Shipley, M. J. (2006a) Why is evidence on job strain and coronary heart disease mixed? An illustra- tion of measurement challenges in the Whitehall II study. Psychosomatic Medicine 68, 398–401. Kivimäki, M., Virtanen, M., Elovainio, M., Kouvonen, A., Väänänen, A. & Vahtera, J. (2006b) Work stress in the etiology of coronary heart disease – a meta-analysis. Scandinavian Journal of Work, Environment & Health 32, 431–442. Korkeila, J. A., Töyry, S., Kumpulainen, K., Toivola, J.-M., Räsänen, K. & Kalimo, R. (2003) Burnout and self-perceived health among Finnish psychiatrists and child psychiatrists: a national survey. Scandinavian Journal of Public Health 31, 85–91. Kouvonen, A., Kivimäki, M., Elovainio, M., Virtanen, M., Linna, A. & Vahtera, J. (2005) Job strain and leisure-time physical activity in female and male public sector employees. Preventive Medicine 41, 532–539. Kouvonen, A., Kivimäki, M., Väänänen, A., Heponiemi, T., Elovainio, M., Ala-Mursula. L., Virtanen, M., Pentti, J., Linna, A. & Vahtera, J. (2007) Job strain and adverse health behaviors: the Finnish public sector study. Journal of Occupational and Environmental Medicine 49, 68–74. Krantz, G., Berntsson, L. & Lundberg, U. (2005) Total workload, work stress and perceived symptoms in Swedish male and female white-collar employees. European Journal of Public Health 15, 209–214. Kristensen, T. S., Borritz, M., Villadsen, E. & Christensen, K. B. (2005) Th e Copenhagen Burn- out Inventory: a new tool for the assessment of burnout. Work & Stress 19, 192–207. Kurdyak, P. A. & Gnam, W. H. (2005) Small signal, big noise: performance of the CIDI depression module. Canadian Journal of Psychiatry 50, 851–856. Laitinen-Krispijn, S. & Bijl, R. V. (2000) Mental disorders and employee sickness absence: the NEMESIS study. Social Psychiatry and Psychiatric Epidemiology 35, 71–77. Landsbergis, P. A. (1988) Occupational stress among health care workers: a test of the job demands-control model. Journal of Organizational Behavior 9, 217–239. Landsbergis, P. A., Schnall, P. L., Warren, K., Pickering, T. G. & Schwartz, J. E. (1994) As- sociation between ambulatory blood pressure and alternative formulations of job strain. Scandinavian Journal of Work, Environment & Health 20, 349–363.

106 REFERENCES

Landsbergis, P. A. (2003) Th e changing organization of work and the safety and health of working people: a commentary. Journal of Occupational and Environmental Medicine 45, 61–72. Langelaan, S., Bakker, A. B., Schaufeli, W. B., van Rhenen, W. & van Doornen, L. J. P. (2006) Do burned-out and work-engaged employees diff er in the functioning of the hypothalamic-pituitary-adrenal axis? Scandinavian Journal of Work, Environment & Health 32, 339–348. Larsman, P., Sandsjö, L., Klipstein, A., Vollenbroen-Hutten, M. & Christensen, H. (2006) Perceived work demands, felt stress, and musculoskeletal neck-shoulder symptoms among elderly female computer users. Th e NEW Study. European Journal of Applied Physiology 96, 127–135. Lazarus, R. S. (1991) Psychological stress in the workplace. Journal of Social Behaviour and Personality 6, 1–13. Lazarus, R. S. & Folkman, S. (1986) Cognitive theories of stress and the issue of circularity. In M. Appley & R. Trumbull (eds.) Dynamics of Stress. New York: Plenum Press. Leana, C. R. & Feldman, D. C. (1991) Gender diff erences in responses to unemployment. Journal of Vocational Behavior 38, 65–77. Lee, R. T. & Ashforth, B. E. (1993) A longitudinal study of burnout among supervisors and managers: comparisons between the Leiter and Maslach (1988) and Golembiewski et al. (1986) models. Organizational Behavior and Human Decision Processes 54, 369–398. Lee, R. T. & Ashforth, B. E. (1996) A meta-analytic examination of the correlates of the three dimensions of job burnout. Journal of Applied Psychology 81, 123–133. Lehto, A.-M. & Sutela, H. (2005) Th reats and opportunities: fi ndings of Finnish quality of worklife surveys 1977–2003. Helsinki: Statistics Finland. Lehtonen, R., Djerf, K., Härkänen, T. & Laiho, J. (2003) Modelling complex health survey data: a case study. In R. Höglund, M. Jäntti & G. Rosenqvist (eds.) Statistics, Econometrics and Society: Essays in Honour of Leif Norberg, pp. 91–114. Research Reports 238. Helsinki: Statistics Finland. Leino, P. & Hänninen, V. (1995) Psychosocial factors at work in relation to back and limb disorders. Scandinavian Journal of Work, Environment & Health 21, 134–142. Leiter, M. P. (1993) Burnout as a developmental process: consideration of models. In W. B. Schaufeli, C. Maslach & T. Marek (eds.) Professional Burnout: Recent Developments in Th eory and Research, pp. 237–250. London: Taylor & Francis. Leiter, M. P. & Durup, J. (1994) Th e discriminant validity of burnout and depression: a confi rmatory factor analytic study. Anxiety, Stress and Coping 7, 357–373. Leiter, M. P. & Maslach, C. (1988) Th e impact of interpersonal environment on burnout and organizational commitment. Journal of Organizational Behavior 9, 297–308. Leiter, M. P. & Schaufeli, W. B. (1996) Consistency of the burnout construct across occupations. Anxiety, Stress and Coping 9, 229–243.

107 REFERENCES

Lemkau, J. P., Purdy, R. R., Raﬀ erty, J. P. & Rudisill, J. R. (1988) Correlates of burnout among family practice residents. Journal of Medical Education 63, 682–691. Leskelä, U. S., Melartin, T. K., Lestelä-Mielonen, P. S., Rytsälä, H. J., Sokero, T. P., Heikkinen, M. E. & Isometsä, E. T. (2004) Life events, social support, and onset of major depressive episode in Finnish patients. Journal of Nervous and Mental Diseases 192, 373–381. Lindblom, K. M., Linton, S. J., Fedeli, C. & Bryngelsson, I.-L. (2006) Burnout in the working population: relations to psychosocial work factors. International Journal of Behavioral Medicine 13, 51–59. Lindell, M. K. & Whitney, D. J. (2001) Accounting for common method variance in cross- sectional research designs. Journal of Applied Psychology 86, 114–121. Lindström, K., Elo, A.-L., Hopsu, L., Kandolin, I., Ketola, R., Lehtelä, J., Leppänen, A., Mukala, K., Rasa, P.-L. & Sallinen, M. (2005) Työkuormituksen arviointimenetelmä TIKKA [Assess- ment tool TIKKA for work strain]. Helsinki: Finnish Institute of Occupational Health. Lund, T., Kivimäki, M., Labriola, M., Villadsen, E. & Bang Christensen, K. (2007) Using administrative sickness absence data as a marker of future disability pension: the prospective DREAM study of Danish private sector employees. Occupational and Environmental Medicine, Epub ahead of print July 11. DOI: 10.1136/oem.2006.031393. Lundberg, U., Forsman, M., Zachau, G., Eklöf, M., Palmerud, G., Melin, B. & Kadefors, R. (2002) Eﬀ ects of experimentally induces mental and physical stress on motor unit recruit- ment in the trapezius muscle. Work & Stress 16, 166–178. Maier, S. F. & Watkins, L. R. (1998) Cytokines for psychologists: implications of bidirectional immune-to-brain communication for understanding behavior, mood, and cognition. Psy- chological Review 83–107. Marmot, M. G., Feeney, A., Shipley, M., North, F. & Syme, S. L. (1995) Sickness absence as a measure of health status and functioning: from the UK Whitehall II study. Journal of Epidemiology and Community Health 49, 124–130. Martin, F., Poyen, D., Bouderlique, E., Gouvernet, J., Rivet, B., Disdier, P., Martinez, O. & Scotto, J.-C. (1997) Depression and burnout in hospital health care professionals. Interna- tional Journal of Occupational and Environmental Health 3, 204–209. Maslach, C. (1976) Burned-out. Human Behavior 5, 16–22. Maslach, C. & Jackson, S. E. (1981) Th e measurement of experienced burnout. Journal of Oc- cupational Behavior 2, 99–113. Maslach, C. & Jackson, S. E. (1996) Maslach Burnout Inventory – Human Services Survey (MBI-HSS). In C. Maslach, S. E. Jackson & M. P. Leiter (eds.) Maslach Burnout Inventory Manual, 3rd ed., pp. 3–17. Palo Alto (CA): Consulting Psychologists Press. Maslach, C. & Leiter, M. P. (1997) Th e Truth about Burnout. How Organizations Cause Personal Stress and What To Do about It. San Francisco (CA): Jossey-Bass Publishers. Maslach, C. (2001) What have we learned about burnout and health? Psychology and Health 16, 607–611.

108 REFERENCES

Maslach, C., Schaufeli, W. B. & Leiter, M. P. (2001) Job Burnout. Annual Review of Psycho- logy 52, 397–422. Mattila, A. K., Ahola, K., Honkonen, T., Salminen, J. K., Huhtala, H. & Joukamaa, M. (2007) Alexithymia and occupational burnout are strongly associated in working population. Journal of Psychosomatic Research 62, 657–665. McCranie, E. W. & Brandsma, J. M. (1988) Personality antecedents of burnout among middle- aged physicians. Behavioral Medicine 14, 30–36. McEwen, B. S. & Stellar, E. (1993) Stress and the individual. Archieves of Internal Medicine 153, 2093–2101. McEwen, B. S. (1998a) Protective and damaging eff ects of stress mediators. Th e New England Journal of Medicine 338, 171–179. McEwen, B. S. (1998b) Stress, adaption, and disease. Allostasis and allostatic load. In S. M. McCann, J. M. Lipton, Sternberg, E. M., G. P. Chrousos, P. W. Gold & C. C. Smith (eds.) Neuroimmunomodulation: Molecular Aspects, Integrative Systems, and Clinical Advances, pp. 33–44. Annals of the New York Academy of Sciences, Volume 840. New York: Th e New York Academy of Sciences. McKnight, J. D. & Glass, D. C. (1995) Perceptions of control, burnout, and depressive symptomalogy: a replication and extension. Journal of Clinical and Consulting Psychology 63, 490–494. Meier, S. T. (1984) Th e construct of validity of burnout. Journal of Occupational Psychology 57, 211–219. Melamed, S., Kushnir, T. & Meir, E. I. (1991) Attenuating the impact of job demands: additive and interactive eff ects of perceived control and social support. Journal of Vocational Behavior 39, 40–53. Melamed, S., Kushnir, T. & Shirom, A. (1992) Burnout and risk factors for cardiovascular diseases. Behavioral Medicine 18, 53–60. Melamed, S., Ugarten, U., Shirom, A., Kahana, L., Lerman, Y. & Froom, P. (1999) Chronic burnout, somatic arousal and elevated salivary cortisol levels. Journal of Psychosomatic Research 46, 591–598. Melamed, S., Shirom, A., Toker, S., Berliner, S. & Shapira, I. (2006) Burnout and risk of cardiovascular disease: evidence, possible causal paths, and promising research directions. Psychological Bulletin 132, 327–353. Merikangas, K. R. (2005) Anxiety disorders: epidemiology. In D. J. Sadock & C. A. Sadock (eds.) Kaplan & Sadock's Comprehensive Textbook of Psychiatry, Volume 1, Eighth Edition, pp. 1720–1728. Philadelphia: Lippincott Williams & Wilkins. Michie, S. & Williams, S. (2003) Reducing work related psychological ill health and sickness absence: a systematic literature review. Occupational and Environmental Medicine 60, 3–9. Middeldorp, C. M., Cath, D. C. & Boosma, D. I (2006) A twin-family study of the association between employment, burnout and anxious depression. Journal of Aff ective Disorders 90, 163–169.

109 REFERENCES

Miranda, H., Viikari-Juntura, E., Heistaro, S., Heliövaara, M. & Riihimäki, H. (2005) A population study on diff erences in the determinants of a specifi c shoulder disorder versus nonspecifi c shoulder pain without clinical fi ndings. American Journal of Epidemiology 161, 847–855. Moch, S. L., Panz, V. R., Joff e, B. I., Havlik, I. & Moch, J. D. (2003) Longitudinal changes in pituitary-adrenal hormones in South African women with burnout. Endocrine 21, 267–272. Mommersteeg, P. M. C., Heijnen, C. J., Verbraak, M. J. P. M. & van Doornen, L. J. P. (2006a) Clinical burnout is not refl ected in the cortisol awakening response, the day-curve or the response to a low-dose dexamethasone suppression test. Psychoneuroendocrinology 31, 216–225. Mommersteeg, P. M. C., Heijnen, C. J., Verbraak, M. J. P. M. & van Doornen, L. J. P. (2006b) A longitudinal study on cortisol and complaint reduction in burnout. Psychoneuroendo- crinology 31, 793–804. Mommersteeg, P. M. C., Keijsers, G. P. J., Heijnen, C. J., Verbraak, M. J. P. M. & van Doornen, L. J. P. (2006c) Cortisol deviation in people with burnout before and after psychotherapy: A pilot study. Health Psychology 25, 243–248. Mor, V. & Laliberte, L. (1984) Burnout among hospice staff . Health and Social Work 9, 274–283. Moyle, P. (1995) Th e role of negative aff ectivity in the stess process: tests of alternative models. Journal of Organizational Behavior 16, 647–668. Musselman, D. L., Evans, D. L. & Nemeroff , C. B. (1998) Th e relationship of depression to cardiovascular disease: epidemiology, biology, and treatment. Archieves of General Psychiatry 55, 580–592. National Research Council and the Institute of Medicine (2001) Musculoskeletal Disorders and the Work Place: Low Back and Upper-extremities. Panel on Musculoskeletal Disorders and the Work Place. Comission on Behavioral and Social Sciences and Education. Washington: National Academy Press. Nowack, K. M. (1986) Type A, hardiness, and psychological distress. Journal of Behavioral Medicine 9, 537–548. Oliver, M. I., Pearson, N., Coe, N. & Gunnell, D. (2005) Help-seeking behaviour in men and women with common mental health problems: cross-sectional study. British Journal of Psychiatry 186, 297–301. Orth-Gomér, K. & Leineweber, C. (2005) Multiple stressors and coronary disease in women. Th e Stockholm Female Coronary Risk Study. Biological Psychology 69, 57–66. Parkes, K. R. (1990) Coping, negative aff ectivity and the work environment: additive and interactive predictors of mental health. Journal of Applied Psychology 75, 399–409. Pines, A. (1993) Burnout: an existential perspective. In W. B. Schaufeli, C. Maslach & T. Marek (eds.) Professional Burnout: Recent Developments in Th eory and Research, pp. 33–51. London: Taylor & Francis.

110 REFERENCES

Pirkola, S. P., Isometsä, E., Suvisaari, J., Aro, H., Joukamaa, M., Poikolainen, K., Koskinen, S., Aromaa, A. & Lönnqvist, J. (2005) Th e DSM-IV mood-, anxiety-, and alcohol use disorders and their comorbidity in the Finnish general population – Results from the Health 2000 Study. Social Psychiatry and Psychiatric Epidemiology 40, 1–10. Poulin, J. E. & Walter, C. A. (1993) Burnout in gerontological social work. Social Work 38, 305–310. Pruessner, J. C., Hellhammer, D. H. & Kirschbaum, C. (1999) Burnout, perceived stress, and cortisol responses to awakening. Psychosomatic Medicine 61, 197–204. Punnet, L. & Herbert, R. (2000) Work-related musculoskeletal disorders: is there a gender dif- ferential, and if so, what does it mean? In M. B. Goldman & M. C. Hatch (eds.) Women and Health, pp. 474–492. New York: Academic Press. Puttonen, S., Ravaja, N. & Keltikanas-Järvinen, L. (2005) Cloninger’s temperament dimensions and aff ective responses to diff erent challenges. Comprehensive Psychiatry 46, 128–134. Quick, J. C. (1999) Occupational health psychology: the convergence of health and clinical psychology with public health and preventive medicine in an organizational context. Profes- sional Psychology: Research and Practice 2, 123–128. Raff erty, Y. (1987) A test of the job demands-control and person-environmental fi t models of occupational stress as predictors of burnout in human service workers [Doctoral dissertation, State University of New York]. Ann Arbor (MI): University Microfi lms International. Raitasalo, R. & Maaniemi, K. (2006) Mielenterveyden häiriöiden aiheuttamat haitat kasvaneet [Th e negative eff ects of mental health problems have increased]. Sosiaalivakuutus 1/2006, 38–41. Raitasalo, R. & Maaniemi, K. (2007) Alkoholi, kansallinen ongelma [Alcohol, a national problem]. Sosiaalivakuutus 1/2007, 46–50. Ravaja, N., Keltikangas-Järvinen, L. & Kettunen, J. (2006) Cloninger’s temperament dimensions and threat, stress, and performance appraisals during diff erent challenges among young adults. Journal of Personality 74, 287–310. Richardsen, A. M., Burke, R. J. & Leiter, M. P. (1992) Occupational demands, psychological burnout and anxiety among hospital personnel in Norway. Anxiety, Stress and Coping 5, 55–68. Rihmer, Z. & Angst, J. (2005) Mood disorders: Epidemiology. In D. J. Sadock & C. A. Sadock (eds.) Kaplan & Sadock's Comprehensive Textbook of Psychiatry, Volume 1, Eighth Edition, pp. 1575–1582. Philadelphia (PA): Lippincott Williams & Wilkins. Roelofs, J., Verbraak, M., Keijsers, G. P. J., de Bruin, M. B. N. & Schmidt, A. J. M. (2005) Psychometric properties of a Dutch version of the Maslach Burnout Inventory-General Survey (MBI-DV) in individuals with and without clinical burnout. Stress and Health 21, 17–25. Rogers, J. C. & Dodson, S. C. (1988) Burnout in occupational therapists. American Journal of Occupational Th erapy 42, 787–792.

111 REFERENCES

Roxburgh, S. (1996) Gender diff erences in work and well-being: eff ects of exposure and vulnerability. Journal of Health and Social Behavior 37, 265–277. Rugulies, R., Bültman, U., Aust, B. & Burr, H. (2006) Psychosocial work environment and incidence of severe depressive symptoms: prospective fi ndings from a 5-year follow-up of the Danish Work Environment Cohort Study. American Journal of Epidemiology 163, 877–887. Salokangas, R. K. & Poutanen, O. (1998) Risk factors for depression in primary care. Findings of the TAPED project. Journal of Aff ective Disorders 48, 171–180. Salokangas, R. K. R., Mattila, V. & Joukamaa, M. (1988) Intimacy and mental disorder in late middle age. Acta Psychiatrica Scandinavica 78, 555–560. Sanderson, K. & Andrews, G. (2006) Common mental disorders in the workforce: recent fi ndings from descriptive and social epidemiology. Canadian Journal of Psychiatry 51, 63–75. Sauter, S. L., Hurrel Jr., J. J., Fox, H. R., Tetrick, L. E. & Barling, J. (1999) Occupational health psychology: an emerging discipline. Industrial Health 37, 199–211. Schaufeli, W. B., Leiter, M. P., Maslach, C. & Jackson, S. E. (1996) Maslach Burnout Inventory – General Survey (MBI-GS). In C. Maslach, S. E. Jackson & M. P. Leiter (eds.) Maslach Burnout Inventory Manual, Th ird Edition, pp. 19–32. Palo Alto (CA): Consulting Psy- chologists Press. Schaufeli, W. & Enzmann, D. (1998) Th e burnout companion to study and practice: a critical analysis. London: Taylor & Francis. Schaufeli, W. B., Bakker, A. B., Hoogduin, K., Schaap, C. & Kladler, A. (2001) On the clinical validity of the Maslach burnout inventory and the Burnout Measure. Psychology and Health 16, 565–582. Schaufeli, W. B. & Greenglass, E. R. (2001) Introduction to special issue on burnout and health. Psychology and Health 16, 501–510. Schaufeli, W. B. (2004) Th e future of occupational health psychology. Applied Psychology: An International Review 53, 502–517. Schaufeli, W. B. & Bakker, A. B. (2004) Job demands, job resources, and their relationship with burnout and engagement: a multi-sample study. Journal of Organizational Behavior 25, 293–315. Schaufeli, W. B. & Taris, T. W. (2005) Th e conception and measurement of burnout: common ground and worlds apart. Work & Stress 19, 256–262. Schaufeli, W. B. (2006) Th e balance of give and take: toward a social exchange model of burnout. Revue International de Psychologie Sociale 19, 87–131. Schreurs, P. J. G. & Taris, T. W. (1998) Construct validity and the demand control model: a double cross validation approach. Work & Stress 12, 66–84. Schuckit, M. A. (2005) Alcohol-related disorders. In D. J. Sadock & C. A. Sadock (eds.) Ka- plan & Sadock's Comprehensive Textbook of Psychiatry, Volume 1, Eighth Edition, pp. 1168–1188. Philadelphia (PA): Lippincott Williams & Wilkins.

112 REFERENCES

Schutte, N., Toppinen, S., Kalimo, R. & Schaufeli, W. (2000) Th e factorial validity of the Maslach Burnout Inventory-General Survey (MBI-GS) across occupational groups and nations. Journal of Occupational and Organizational Psychology 73, 53–66. Sears Jr., S. F., Urizar Jr., G. G. & Evans, G. D. (2000) Examing a stress-coping model of burnout and depression in extension agents. Journal of Occupational Health Psychology 1, 56–62. Seidman, S. A. & Zager, J. (1991) A study of coping behaviours and teacher burnout. Work & Stress 5, 205–216. Selye, H. (1956) Th e stress of life. New York: McGraw-Hill Book Company. Sennett, R. (2006) Th e culture of the new capitalism. New Haven (CT): Yale University Press. Shirom, A. (1989) Burnout in work organizations. In C. L. Cooper & I. T. Robertson (eds.) International Review on Industrial and Organizational Psychology, pp. 25–48. Chichester: John Wiley & Sons. Shirom, A., Westman, M. & Carel, R. S. (1997) Eff ects of work overload and burnout on cholesterol and triglyserides levels: the moderating eff ects of emotional reactivity among male and female employees. Journal of Occupational Health Psychology 4, 275–288. Shirom, A. (2003) Job-related burnout: a review. In J. C. Quick & L. E. Tetrick (eds.) Hand- book of Occupational Health Psychology. Washington (DC): American Psychological Association. Shirom, A., Melamed, S., Toker, S., Berliner, S. & Shapira, I. (2005) Burnout and health review: current knowledge and future research directions. In G. Hodgkinson & F. Ford (eds.) In- ternational Review of Industrial and Organizational Psychology, pp. 269–308. Chichester: John Wiley & Sons. Siegrist, J. (2000) Place, social exchange and health: proposed sociological framework. Social Science & Medicine 51, 1283–1293. Siegrist, J. (2002) Eff ort-reward imbalance at work and health. In D. Ganster & P. Perrewe (eds.) Historical and current perspectives on stress and health, Volume 2, pp. 261–291. Amsterdam: Elsevier Science Ltd. Siegrist, J. & Rödel, A. (2006) Work stress and health risk behavior. Scandinavian Journal of Work, Environment & Health 32, 473–481. Soares, J. J. F. & Jablonska, B. (2004) Psychosocial experiences among primary care patients with and without musculoskeletal pain. European Journal of Pain 8, 79–89. Sonnentag, S., Brodbeck, F. C., Heinbokel, T. & Stolte, W. (1994) Stressor-burnout relationship in software development teams. Journal of Occupational and Organizational Psychology 67, 327–341. Sparks, K., Faragher, B. & Cooper, C. L. (2001) Well-being and occupational health in the 21st century workplace. Journal of Occupational and Organizational Psychology 74, 489–509.

113 REFERENCES

Stansfeld, S. A., Rael, E. G., Head, J., Shipley, M. & Marmot, M. (1997) Social support and psychiatric sickness absence: a prospective study of British civil servants. Psychological Medicine 27, 35–48. Stansfeld, S. (2002) Work, personality and mental health. British Journal of Psychiatry 181, 96–98. Stansfeld, S. & Candy, B. (2006) Psychosocial work environment and mental health – a meta- analytic review. Scandinavian Journal of Work, Environment & Health 32, 443–462. Statistics Finland (1999) Ammattiluokitus 1997 [Occupational classifi cation 1997]. Handbook 14. Helsinki: Statistics Finland. Statistics Finland (2000) Suomen tilastollinen vuosikirja 2000 [Statistical yearbook of Finland 2000]. Hämeenlinna: Statistics Finland. Suls, J. & Bunde, J. (2005) Anger, anxiety, and depression as risk factors for cardiovascular disease: the problems and implications of overlapping aff ective dispositions. Psychological Bulletin 131, 260–300. Taris, T. W., Schreurs, P. J. G. & Schaufeli, W. B. (1999) Construct validity of the Maslach Burnout Inventory – General Survey: two sample examination of its factor structure and correlates. Work & Stress 13, 223–237. Taris, T. W., Le Blanck, P. M., Schaufeli, W. B. & Schreuers, P. J. G. (2005) Are there causal relationships between the dimensions of the Maslach Burnout Inventory? A review and two longitudinal tests. Work & Stress 19, 238–255. Taylor, S. E., Cousino Klein, L., Lewis, B. P., Gruenewald, T. L., Gurung, R. A. R. & Updegraff , J. A. (2000) Biobehavioral responses to stress in females: tend-and-befriend, not fi ght-or- fl ight. Psychological Review 107, 411–429. Tennant, C. (2001) Work-related stress and depressive disorders. Journal of Psychosomatic Research 51, 697–704. Th e Social Insurance Institution of Finland (2000) Kansaneläkelaitoksen tilastollinen vuosikirja [Statistical yearbook of the Social Insurance Institution of Finland]. Helsinki: Th e Social Insurance Institution of Finland. Th e Social Insurance Institution of Finland (2005) Kansaneläkelaitoksen tilastollinen vuosikirja [Statistical yearbook of the Social Insurance Institution on Finland]. Helsinki: Th e Social Insurance Institution of Finland. Th eorell, T. & Karasek, R. A. (1996) Current issues relating to psychosocial job strain and cardiovascular disease research. Journal of Occupational Health Psychology 1, 9–26. Th orensen, C. J., Kaplan, S. A., Barsky, A. P., de Charmont, K. & Warren, C. R. (2003) Th e aff ective underpinnings of job perceptions and attitudes: a meta-analytic review and integra- tion. Psychological Bulletin 129, 914–945. Tinsley, H. E. A. (2000) Th e congruence myth: an analysis of the effi cacy of the person-environment fi t model. Journal of Vocational Behavior 56, 147–179.

114 REFERENCES

Toker, S., Shirom, A., Shapira, I., Berliner, S. & Melamed, S. (2005) Th e association between burnout, depression, anxiety, and infl ammation biomarkers: C-reactive protein and fi brinogen in men and women. Journal of Occupational Health Psychology 10, 344–362. Toppinen-Tanner, S., Kalimo, R. & Mutanen, P. (2002) Th e process of burnout in white-collar and blue-collar jobs: eight-year prospective study of exhaustion. Journal of Organizational Behavior 23, 555–570. Toppinen-Tanner, S., Ojajärvi, A., Väänänen, A., Kalimo, R. & Jäppinen, P. (2005) Burnout as a predictor of medically certifi ed sick-leave absences and their diagnosed causes. Behavioral Medicine 31, 18–27. Tselebis, A., Moulou, A. & Ilias, J. (2001) Burnout versus depression and sense of coherence: Study of Greek nursing staff . Nursing and Health Sciences 3, 69–71. Turnipseed, D. L. (1998) Anxiety and burnout in the health care work environment. Psycho- logical Reports 82, 627–642. Tyrka, A. R., Mello, A. F., Mello, M. F., Gagne, G. G., Grover, K. E., Anderson, G. M., Price, L. H. & Carpenter, L. L. (2006) Temperament and hypothalamic-pituitary-adrenal axis function in healthy adults. Psychoneuroendocrinology 31, 1036–1045. Töyry, S. (2005) Burnout and self-reported health among Finnish physicians [Doctoral dissertation, University of Kuopio]. Kuopio University Publications D. Medical Sciences 365. Kuopio: Kuopio University. van der Doef, M. & Maes, S. (1999) Th e job demand-control(-support) model and psychological well-being: a review of 20 years of empirical research. Work & Stress 13, 87–114. van Dierendonck, D., Schaufeli, W. B. & Buunk, B. P. (2001) Toward a process model of burnout: results from a secondary analysis. European Journal of Work and Organizational Psychology 10, 41–52. Virtanen, M., Honkonen, T., Kivimäki, M., Ahola, K., Vahtera, J., Aromaa, A. & Lönnqvist, J. (2007) Work stress, mental health and antidepressant medication: fi ndings from the Health 2000 Study. Journal of Aff ective Disorders 98, 189–197. Vredenburgh, L. D., Carlozzi, A. F. & Stein, L. B. (1999) Burnout in counselling psychologists: type of practice setting and pertinent demographics. Counselling Psychology Quarterly 12, 293–302. Väänänen, A., Toppinen-Tanner, S., Kalimo, R., Mutanen, P., Vahtera, J. & Peiró, J. M. (2003) Job characteristics, physical and psychological symptoms, and social support as antecedents of sickness absence among men and women in the private industrial sector. Social Science & Medicine 57, 807–824. Väänänen, A., Kevin, M. V., Ala-Mursula, L., Pentti, J., Kivimäki, M. & Vahtera, J. (2004) Th e double burden of and negative spill-over between paid and domestic work: associations with health among men and women. Women and Health 40, 1–18. Warr, P. (1987) Work, unemployment and mental health. Oxford: Clarendon Press.

115 Watson, D. & Clark, L. A. (1984) Negative aff ectivity: Th e disposition to experience aversive emotional states. Psychological Bulletin 96, 465–490. Weber, A. & Jaekel-Reinhard, A. (2000) Burnout syndrome: a disease of modern societies. Oc- cupational Medicine 50, 512–517. Wilhelm, K., Kovess, V., Rios-Seidel, C. & Finch, A. (2004) Work and mental health. Social Psychiatry and Psychiatric Epidemiology 39, 866–873. Winwood, P. C., Winefi eld, A. H. & Lushington, K. (2003) Th e role of occupational stress in the maladaptive use of alcohol by dentists: a study of South Australian general dental practitioners. Australian Dental Journal 48, 102–109. Wittchen, H.-U., Lachner, G., Wunderlich, U. & Pfi ster, H. (1998) Test-retest reliability of the computerized DSM-IV version of the Munich-Composite International Diagnostic Interview (M-CIDI). Social Psychiatry and Psychiatric Epidemiology 33, 568–578. World Health Organization (1946) Defi nition of Health. Preamble to the Constitution of the World Health Organization as Adopted by the International Health Conference, New York 19–22 June, 1946; entered into force on 7 April, 1948. (http://www.who.int/entity/hac/ techguidance/training/induction/defi nitions%20list.pdf) Downloaded on August 16, 2007. World Health Organization (1992) Th e ICD-10 Classifi cation of Mental and Behavioural Disorders: Clinical Descriptions and Diagnostic Guidelines. Geneva: World Health Or- ganization. World Health Organization (2002) Global Burden of Disease 2002: DALYs by Age, Sex, and Cause for the Year 2002. WHO region: EURO. (http://www.who.int.healthinfo/ bodgbd2002revised /en/index.html) Downloaded on March 8, 2007. Zapf, D., Dormann, C. & Frese, M. (1996) Longitudinal studies in organizational stress research: A review of the literature with reference to methodological issues. Journal of Occupational Health Psychology 2, 145–169. Zapf, D. (2002) Emotion work and psychological well-being. A review of the literature and some conceptual considerations. Human Resource Management Review 12, 237–268. Zellars, K. L., Perrewé, P. L. & Hochwarter, W. A. (2000) Burnout in health care: the role of the fi ve factors of personality. Journal of Applied Social Psychology 30, 1570–1598. Zimmermann, U., Spring, K., Kunz-Ebrecht, S. R., Uhr, M., Wittchen, H. U. & Holsboer, F. (2004) Eff ects of ethanol on hypothalamic-pituitary-adrenal system response to psychosocial stress in sons of alcohol-dependent fathers. Neuropsychopharmacology 29, 1156–65.

116