Burnout as a predictor of all-cause mortality among industrial employees: a 10-year prospective register-linkage study

Journal of Psychosomatic Research 69 (2010) 51 – 57 Burnout as a predictor of all-cause mortality among industrial employees: A 10-year prospective register-linkage study☆ Kirsi Aholaa,⁎, Ari Väänänena, Aki Koskinena, Anne Kouvonenb, Arie Shiromc aFinnish Institute of Occupational Health, Helsinki, Finland bInstitute of Work, Health, and Organizations, University of Nottingham, Nottingham, United Kingdom Received 3 August 2009; received in revised form 17 December 2009; accepted 5 January 2010 Objective: Burnout, a psychological consequence of pro- graphic factors and register-based health status according to longed work stress, has been shown to coexist with physical and entitled medical reimbursement and prescribed medication for mental disorders. The aim of this study was to investigate mental health problems, cardiac risk factors, and pain problems.
whether burnout is related to all-cause mortality among employ- Results: During the 10-year 10-month follow-up, a total of 199 ees. Methods: In 1996, of 15,466 Finnish forest industry employees had died. The risk of mortality per one-unit increase employees, 9705 participated in the ‘Still Working’ study and in burnout was 35% higher (95% CI 1.07–1.71) for total score 8371 were subsequently identified from the National Population and 26% higher (0.99–1.60) for exhaustion, 29% higher for Register. Those who had been treated in a hospital for the most cynicism (1.03–1.62), and 22% higher for diminished profes- common causes of death prior to the assessment of burnout were sional efficacy (0.96–1.55) in participants who had been under excluded on the basis of the Hospital Discharge Register, 45 at baseline. After adjustments, only the associations regarding resulting in a final study population of 7396 people. Burnout was burnout and exhaustion were statistically significant. Burnout measured using the Maslach Burnout Inventory–General Survey.
was not related to mortality among the older employees.
Dates of death from 1996 to 2006 were extracted from the Conclusion: Burnout, especially work-related exhaustion, may National Mortality Register. Mortality was predicted with Cox hazard regression models, controlling for baseline sociodemo- 2010 Elsevier Inc. All rights reserved.
Keywords: Burnout; Exhaustion; Forest industry; Mortality; Prospective; Work have been found to contribute to the risk of early death.
Furthermore, high experienced justice at work has been Extensive overwork is acknowledged as a serious health shown to buffer against premature death .
risk In Japan, the phenomenon of Karoshi—death from Burnout is a psychological reaction to chronic work stress overwork—has recently been recognized as a social concern with the core content of gradual depletion of employ- . Also, other psychosocial work conditions, such as low ees' intrinsic energetic resources over time The job control high job strain, and effort–reward most widely used conceptualization (e.g., Ref. ) views imbalance , and major downsizing in the organization burnout as consisting of three major components: exhaus-tion, cynicism, and diminished professional efficacy ☆ This work was supported by the Academy of Finland (grant number The estimated prevalence of severe burnout has ranged from 128089) and the Finnish Work Environment Fund (grant number 109395, 3% to 7% in representative working populations Several work characteristics, such as high workload, role ⁎ Corresponding author. Finnish Institute of Occupational Health, conflict and ambiguity, low predictability, lack of participa- Topeliuksenkatu 41 a A, FI-00250 Helsinki, Finland. Tel.: +358 30 474 tion and support, and experienced unfairness, have also been 0022-3999/10/$ – see front matter 2010 Elsevier Inc. All rights reserved.
doi: K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 In some European countries, burnout has entitled disease was emphasized in the group of workers less than 50 financial compensation and the use of rehabilitation services years of age Since burnout develops as a consequence for employees. In Sweden, for example, burnout has been a of prolonged work stress , it is possible also that the legitimate diagnosis for medical certificates , and in association between burnout and mortality might differ the Netherlands, burnout has been considered an occupa- according to the age of the workers.
tional disease justifying work disability benefits . In The aim of the present study was to investigate whether Finland, burnout does not justify compensation, but it has burnout predicts all-cause mortality among forest industry been shown to independently associate with physician- employees. We took the baseline health status of the diagnosed sickness absences and to predict disability participants into account by excluding those with major pensions . Burnout-related absences tend to last for health problems and by adjusting for minor health relatively long periods , the excess risk being approx- problems. Following the above rationale, we examined imately 50–65 working days However, to our age group as a potential moderator of the relationship knowledge, there are no prospective studies linking burnout Burnout has been shown to often coexist with depressive, anxiety, and alcohol use disorders and musculoskel- etal disorders and cardiovascular diseases . In addition,in a 4-year prospective study of initially healthy men, burnout was associated with a twofold risk of developing amyocardial infarction Burnout has also been found to This study is a part of the ongoing ‘Still Working’ cohort predict diabetes , the common cold and sick leaves study examining work-related antecedents of health, mor- due to musculoskeletal disorders and diseases of the bidity, and mortality in a multinational private sector forest respiratory system . Furthermore, burnout has been industry corporation . The researchers gave each found to predict depressive symptoms , new cases of employee in the corporation an identification code, which insomnia and sickness absences due to mental and was marked in the questionnaire. The link between this identification code and the national personal identification The process between stress and health problems can be number given to all Finns at birth was known only to the direct (i.e., stress predisposing to illness or accelerating the researchers and used later to merge the questionnaire data disease process in its subclinical phase), indirect (i.e., stress collected in 1996 to data from several national Finnish increasing unfavorable health behavior), or even reversed registers till the end of 2006. During spring 1996, the (i.e., coping with illnesses at work further increasing stress) questionnaires which at that point did not contain any . As burnout reflects past accumulated exposure to a personal information were sent to the work units, distributed variety of stressful conditions at work , it may be a phase to employees by their supervisors, and, once completed, in the process between adverse psychosocial factors at work mailed directly to the Finnish Institute of Occupational and ill health. However, it is equally possible that the early Health. Participation was voluntary, and confidentiality was phases of illnesses reduce the possibilities to reach one's assured to all employees. Approval of the Ethics Committee goals at work, eventually leading to burnout .
of the Finnish Institute of Occupational Health was obtained For example, in a prospective study among Finnish dentists, a reciprocal relationship between burnout and depressionwas established. However, the relationship between job strain and depression was totally mediated by burnout, whilethe relationship between job strain and burnout was only Of the eligible Finnish employees aged 16–65 years partially mediated by depression The mechanisms (n=15,466) who returned the questionnaire (n=9705, re- linking chronic stress and strain with mortality are sponse rate 63%), 8371 could later be identified from the relevant in explaining the biological plausibility of the work database of the National Population Register Centre contain- ing the personal information. Of this base population, the In a long-term cohort study, the association between 542 people who before baseline, according to the Finnish burnout and diseases may differ between workers of Hospital Discharge Register, had already been treated as an separate age groups. Older workers are more likely to in-patient for the most common causes of death in Finland, leave the work force or retire during the follow-up period.
i.e., alcohol problems, heart disease, cancer, and suicide Therefore, those who continue working until the official attempt were excluded. Furthermore, 433 persons were retirement age may be characterized by especially high excluded due to missing values in the relevant items in the resiliency and effective health-enhancing coping strategies.
questionnaire leaving 7396 employees in the final study Supporting this line of reasoning, recent longitudinal population. Women (24% vs. 21%, Pb.001), nonmanual evidence showed that the association between chronic workers (38% vs. 19%, Pb.001), and married participants work-related stress and an increased risk of cardiovascular (65% vs. 60%, Pb.001) were overrepresented in the final K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 study sample compared to the original population, whereas age, gender, and marital status were obtained from the no difference emerged regarding age group, formed by National Population Register Centre. Marital status was dichotomizing the age distribution at its medium, i.e., 44 dichotomized as married vs. unmarried based on the register years (P=.146). In addition, mortality was higher among information. Occupational status was collected from those excluded compared to those in the final study employer's records. It was dichotomized as manual (i.e., production and maintenance) vs. nonmanual worker (i.e.,supervisory, research, developmental, and office work).
Burnout was measured using the Maslach Burnout Inventory–General Survey (MBI-GS) The MBI- The associations between the burnout and the subscales GS consists of the following three subscales: exhaustion and mortality were analyzed using Cox proportional hazard (five items, Cronbach's α=0.87), cynicism (five items, regression models. For each participant, person-days of α=0.76), and (diminished) professional efficacy (six items, follow-up were calculated from 1 March 1996 to either the α=0.83). Satisfactory reliability and validity of the MBI- event of death or 31 December 2006, whichever came first.
GS have been confirmed In the present study, one of Hazard ratios (HR) and their 95% confidence intervals (95% the cynicism items was left out due to its low correlation with CI) provided risk estimates associated with the 1-S.D.
the other items intended to measure cynicism (α=0.83 increase in the standardized burnout and subscale sum scores afterwards). The items were scored on a seven-point for overall mortality. The time-dependent interaction terms frequency rating scale ranging from 0 (never) to six (daily).
between the burnout and the subscales and logarithm of the High scores for exhaustion and cynicism and low scores for follow-up period were examined to confirm that professional efficacy are indicative of burnout. The items of the proportional hazards assumptions were justified. The professional efficacy were reversed (diminished professional analyses were adjusted stepwise for sociodemographic efficacy). One missing value per burnout subscale was factors and health problems at baseline. The significance of allowed. A sum score, in which exhaustion, cynicism, and interaction effects on mortality between total burnout and its lack of professional efficacy have different weights (0.4×ex- three subscales, and age group or gender of the respondents haustion+0.3×cynicism+0.3×lack of professional efficacy), was systematically tested by including interaction terms in the models. In case of a significant interaction, the analyses The date of dying from the death certificate was extracted were stratified. Two-tailed P values below .05 were from the National Mortality Register maintained by Statistics considered to indicate statistical significance. The analyses Finland from 1 March 1996 to 31 December 2006.
were performed using the SAS statistical program package, Registered medication use was taken as an indicator of version 9.1 (SAS Institute, Cary, NC, USA).
minor health problems affecting the coping resources of theworkers. All Finns are entitled to drug reimbursement as partof the national public health insurance . Firstly, approved drugs qualify for partial reimbursement under the basicrefund category. All outpatient drug purchases are contained The majority of the study participants were men (76%), in the Prescription Register according to the Anatomical manual workers (62%), and married (65%) employees. The Therapeutic Chemical classification code . Secondly, mean age of the participants was 43 (S.D. 9 years, median 44 drugs used to treat certain chronic illnesses are fully years, range 16–65 years). The detailed characteristics of the reimbursed under the special refund category. These study population are presented in according to the entitlements are contained in the Special Refund Entitlement Register. The totally reimbursed medication (for example, All of the time-dependent interaction terms between diabetes mellitus, chronic hypertension, chronic coronary burnout and the subscales and the logarithm of the follow-up artery disease, and hyperlipidemia) until 1996 and the period were nonsignificant (PN.20), indicating that the partially reimbursed medication for mental health problems association between burnout and mortality was not moder- (psycholeptics and antidepressants), cardiac risk factors ated by the time elapsed between survey and death, therefore (cardiac therapy, antihypertensive, and lipid-modifying justifying the use of proportional hazard models. Significant agents), and pain problems (anti-inflammatory and antirheu- interaction effects between burnout and the exhaustion matic products) during 1994 and 1995, as the most common component and age group (P=.02 and .04, respectively) were causes for work disability in Finland, were adjusted for in found. Therefore, the analyses were stratified by age group.
A total of 199 (3%) of the employees died during the Sociodemographic factors, i.e., age, gender, marital follow-up. The results showed that a one-unit increase in the status, and occupational status, were used as covariates in standardized burnout sum score was related to a 35% the analyses because they have been shown to relate to the increase in the risk of all-cause mortality among workers level of burnout as well as to mortality Data on younger than 45. The association attenuated (HR=1.31) but K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 predicted mortality among the younger workers. Each one- Characteristics of the study population by age group unit increase in the sum score was related to a 29% increase in the risk of mortality. However, this association attenuated after adjustment for sociodemographic factors and was nolonger statistically significant. The younger workers who reported diminished professional efficacy were at a 22% excess risk of mortality, but this association did not reach statistical significance before or after adjustments. The burnout subscales did not predict mortality among the older workers, i.e., those who had been at least 45 years old In this prospective cohort study linking various indepen- dent health-related national registers, overall burnout and the exhaustion component were related to all-cause mortality during the 10-year 10-month follow-up period among those forest industry workers who at baseline were under 45 years of age, after the sociodemographic factors and health problems of the participants at baseline had been taken into account. The cynicism component of burnout predicted mortality before but not after adjustments, suggesting thatthis association was affected by sociodemographic factors.
remained significant after all adjustments ().
The association between the third subscale of burnout, i.e., Exhaustion sum score was related to a 26% increase in the diminished professional efficacy, and mortality was weak risk of mortality among younger workers, but the association and did not reach statistical significance. Among employees failed to reach significance. However, when adjusted for aged 45 or over, burnout was not associated with mortality.
sociodemographic factors, this association was statistically Burnout is a result of chronic work stress . The significant. After further adjustment for health problems at present results concerning the association between burnout baseline, the excess risk of mortality related to a 1-S.D.
and subsequent mortality are in accordance with Japanese findings that working too much can lead to death, i.e., the Table 2Hazard ratios of one-unit increase in the standardized burnout subscales for overall mortality among Finnish forest industry workers between 1996 and 2006 byage group HR, Hazard ratio; CI, confidence interval; PE, professional efficacy.
a Model 1 is unadjusted.
b Model 2 is adjusted for baseline gender, marital status, and socioeconomic status.
c Model 3 is adjusted for baseline gender, marital status, socioeconomic status, and common risk factors for health and work ability (by register-based medication, for example, for coronary artery disease, depression, diabetes, hyperlipidemia, hypertension, and pain).
K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 Karoshi phenomena—death by overwork Also, other , is the core of burnout and especially related to aspects of overload have in prospective studies shown lethal the process of deteriorating health .
consequences for workers, for example, high job strain and The association between burnout and exhaustion and effort–reward imbalance , low job control , mortality was observed only among younger workers, a experienced injustice and major downsizing in the group formed by dichotomizing the baseline age distribution at its medium, i.e., 44 years. In the WOLF Stockholm Study, The most common specific causes of death among it was suggested that the inclusion of older employees may working-age Finns are alcohol-related causes, coronary dilute the association between job strain and cardiovascular artery disease, accidents, suicide, and specific cancers (breast disease as a result of the healthy worker survivor bias cancer in women and lung cancer in men) . The possible Because work disability increases with age, it has been found pathways linking burnout to mortality should be further that men in manual jobs in particular are healthier than those explored in future research. There may be several plausible who drift outside the work force . Therefore, the level of mechanisms to account for the observed associations. For job strain may decrease in the group still working and its example, there is a body of evidence supporting the effects may be weaker among the remaining participants association between burnout and the risk factors for leading to a lower risk of burnout-related mortality in the cardiovascular disease: at least several components of the group of workers who were older at baseline. The association metabolic syndrome, i.e., change in stress hormone levels, between exhaustion and mortality was not statistically low-grade inflammation, impairment of the immune system, significant until adjusted for gender, marital status, and blood coagulation, and fibrinolysis, have been found to be occupational status. This indicates that there may be other associated with burnout . The evidence on burnout and factors, in addition to age, which affect the exhaustion- health-risk behaviors is scarce and inconsistent , but the results that have linked work stress to a higher likelihood There are several limitations to be taken into consider- of adverse health behaviors support the possibility of ation in this study. First, we used a nonrandom sample in an additional, indirect path between burnout and cardiovas- only one occupational branch, the forest industry. Although our target organization, a large private sector company, Furthermore, burnout is related to an increase in the level employs heterogeneous personnel ranging from production of depressive symptoms and it often coexists with to managerial work, the majority of the study sample depressive disorders . Depression may in turn predispose comprised manual workers. However, it has been shown that to suicide and hazardous behavior leading to accidents burnout can evolve in all types of work and that the as well as to cardiovascular diseases It also process of burning out is similar among blue- and white- has high comorbidity with alcohol problems These collar workers In the final study sample, women were observations support the mental health-related link between overrepresented compared to the original, male-dominated burnout and mortality as another possible mechanism.
personnel. However, the risk of burnout was shown not to To our knowledge, there is no specific evidence on the differ by gender in the Finnish working population mechanism linking burnout to cancer or other kinds of Furthermore, there was no statistically significant interaction neoplasm. It is possible that, in addition to the cause-specific between gender and burnout on mortality.
links between burnout and mortality, part of the burnout- Second, even though the original response rate (63%) was related mortality risk is general in nature, i.e., associated with of a satisfactory level in relation to observational studies in the depletion of total health resources. A previous study on general the final identified sample with no missing the current cohort showed that, in addition to associating information at baseline covered only 48% of all employees.
with disability pensions due to mental disorders, exhaustion- Fortunately, the mortality data were complete, i.e., covered related chronic work disability was related to the combined all employees, and were therefore not influenced by the group of miscellaneous disorders as the diagnosed cause of degree of burnout. The final study sample suffered from the healthy worker effect, i.e., that the participants were healthier In previous population-based studies, all three burnout according to mortality than those missing. This observation components have been shown to relate to health when suggests increased pathology among those excluded.
indicated by coexisting illnesses , while the exhaus- Therefore, the most likely consequence, if any, is a possible tion and cynicism subscales have mediated the association underestimation of the association between burnout and with work disability . However, in the present mortality. However, caution is warranted regarding gener- prospective study in an industrial sample, in addition to alization of these results to other, dissimilar populations.
burnout syndrome, only the exhaustion subscale predicted Thirdly, even though the sample was large, mortality was mortality after adjustments. The components of burnout have quite rare during the approximately 10-year follow-up period been shown to relate differently to various aspects of the which may have caused diminished statistical power in the work environment and individual well-being It has stratified analyses. All in all, these findings need to be previously been suggested that the exhaustion subscale, replicated in representative samples with a larger number closely resembling vital exhaustion and chronic fatigue K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 We took employee's health status at baseline into account [8] Elovainio M, Leino-Arjas P, Vahtera J, Kivimäki M. Justice at work in two ways. Firstly, we included only workers who have not and cardiovascular mortality: a prospective cohort study. J PsychosomRes 2006;61:271–4.
already been treated as in-patients in a hospital due to alcohol [9] Melamed S, Shirom A, Toker S, Berliner S, Shapira I. Burnout and risk problems, cancer, cardiovascular diseases, or suicide, which of cardiovascular disease: evidence, possible causal paths, and are the most common causes of death in Finland .
promising research directions. Psychol Bull 2006;132:327–53.
Secondly, we adjusted for other health problems at baseline [10] Shirom A. Job-related burnout: a review. In: Quick JC, Tetrick LE, with register-based medication use. This adjustment covered, editors. Handbook of occupational health psychology. Washington(DC): American Psychological Association, 2003. pp. 245–64.
for example, diabetes, hypertension, hyperlipidemia, and [11] Kristensen TS, Borritz M, Villadsen E, Christensen KB. The depression and cardiac and pain problems which most Copenhagen Burnout Inventory: a new tool for the assessment of probably affect the coping resources of the workers. It must burnout. Work Stress 2005;19:192–207.
be acknowledged that these factors did not cover all possible [12] Roelofs J, Verbraak M, Keijsers GPJ, de Bruin MBN, Schmidt AJM.
health problems and left out untreated illnesses in particular.
Psychometric properties of a Dutch version of the Maslach BurnoutInventory General Survey (MBI-DV) in individuals with and without In future studies, the baseline health assessment should cover clinical burnout. Stress Health 2005;21:17–25.
all participants regardless of their need for treatment. On the [13] Schaufeli WB, Enzmann D. The burnout companion to study and other hand, our study benefited from the use of independent practice: a critical analysis. London: Taylor & Francis, 1998.
national register data for exclusion, adjustment, and asses- [14] Maslach C, Jackson SE. Maslach Burnout Inventory–Human Services sing the outcome, which helped us to avoid common method Survey (MBI-HSS). In: Maslach C, Jackson SE, Leiter MP, editors.
Maslach burnout inventory manual. Palo Alto, CA: Consulting bias. Finally, we were not able to control for the health Psychologists Press, 1996. pp. 3–17.
behaviors (such as physical activity, smoking, or heavy [15] Schaufeli WB, Leiter MP, Maslach C, Jackson SE. Maslach Burnout drinking) of the employees because they were not Inventory–General Survey (MBI-GS). In: Maslach C, Jackson SE, covered in the data collection phase in 1996 in the ‘Still Leiter MP, editors. Maslach burnout inventory manual. Palo Alto, CA: Consulting Psychologists Press, 1996. pp. 19–32.
[16] Ahola K, Honkonen T, Isometsä E, Kalimo R, Nykyri E, Aromaa A, In conclusion, our study showed that work-related et al. The relationship between job-related burnout and depressive exhaustion was a risk factor for all-cause mortality for disorders — results from the Finnish Health 2000 Study. J Affect over 10 years among young industrial employees, i.e., those less than 45 years of age. Regarding the other [17] Hallsten L, Josephson M, Torgén M. Performance-based self-esteem.
subscales of burnout, the relationship between cynicism A driving force in burnout processes and its assessment. Stockholm:National Institute for Working Life, 2005.
and mortality was affected by sociodemographic factors [18] Schaufeli WB, Bakker AB. Job demands, job resources, and their and between diminished efficacy and mortality it was weak.
relationship with burnout and engagement: a multi-sample study. J Burnout did not predict mortality among the older employees. These results point to the importance of [19] Borritz M, Bültmann U, Rugulies R, Christensen KB, Villadsen E, tackling exhausting work-related problems. The association Kristensen T. Psychosocial work characteristics as predictors forburnout: findings from 3-year follow-up of the PUMA study. J Occup between burnout and mortality needs to be studied further in a representative sample with possibilities to explore the [20] Maslach C, Leiter MP. Early predictors of job burnout and engagement. J Appl Psychol 2008;93:498–512.
[21] Andersson B. Depression and other mental disorders as causes of sickness absenteeism and work disability pensions in Sweden. In: Järvisalo J, Andersson B, Boedeker W, Houtman I, editors. Mentaldisorders as a major challenge in prevention of work disability.
[1] Sokejima S, Kagamimori S. Working hours as a risk factor for acute Experiences in Finland, Germany, the Netherlands and Sweden.
myocardial infarction in Japan: case-control study. BMJ 1998;317: Helsinki: The Social Insurance Institution of Finland, 2006.
[2] Iwasaki K, Takahashi M, Nakata A. Health problems due to long [22] Geurts S, Kompier M, Gründemann R. Curing the Dutch disease? working hours in Japan: working hours, workers' compensation Sickness absence and work disability in the Netherlands. Int Soc Secur (Karoshi), and preventive measures. Ind Health 2006;44:537–40.
[3] Kivimäki M, Leino-Arjas P, Luukkonen R, Riihimäki H, Vahtera J, [23] Houtman I, Desczca S, Brenninkmeijer V. Sickness absenteeism and Kirjonen J. Work stress and risk of cardiovascular mortality: disability due to mental health problems in the Netherlands. In: Järvisalo prospective cohort study of industrial employees. BMJ 2002;325:857.
J, Andersson B, Boedeker W, Houtman I, editors. Mental disorders as a [4] Johnson JV, Stewart W, Hall EM, Fredlund P, Theorell T. Long-term major challenge in prevention of work disability. Experiences in psychosocial work environment and cardiovascular mortality among Finland, Germany, the Netherlands and Sweden. Helsinki: The Social Swedish men. Am J Pub Health 1996;86:324–31.
Insurance Institution of Finland, 2006. pp. 80–105.
[5] Amick BC, McDonough P, Chang H, Rogers W, Pieper CF, Duncan [24] Ahola K, Honkonen T, Virtanen M, Koskinen S, Kivimäki M, G. Relationship between all-cause mortality and cumulative working Lönnqvist J. Occupational burnout and medically certified sickness life course psychosocial and physical exposures in the United States absence: a population-based study of Finnish employees. J Psychosom labor market from 1968 to 1992. Psychosom Med 2002;64:370–81.
[6] Vahtera J, Kivimäki M, Pentti J, Linna A, Virtanen M, Virtanen P, [25] Ahola K, Gould R, Virtanen M, Honkonen T, Aromaa A, Lönnqvist J.
et al. Organizational downsizing, sickness absence, and mortality: 10- Occupational burnout as a predictor of disability pension: a population- town prospective cohort study. BMJ 2004;328:555–60.
based cohort study. Occup Environ Med 2009;66:284–90.
[7] Maslach C, Schaufeli WB, Leiter MP. Job burnout. Ann Rev Psychol [26] Ahola K, Toppinen-Tanner S, Huuhtanen P, Koskinen A, Väänänen A.
Occupational burnout and chronic work disability: an eight-year cohort K. Ahola et al. / Journal of Psychosomatic Research 69 (2010) 51–57 study on pensioning among Finnish forest industry workers. J Affect [47] Martikainen J, Rajaniemi S. Drug imbursement systems in EU Member States, Iceland and Norway. Helsinki: The Social Insurance Institution [27] Ahola K. Occupational burnout and health [dissertation]. People and Work Research Reports 81. Helsinki: Finnish Institute of Occupational [48] World Health Organization Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC Classification and DDD [28] Honkonen T, Ahola K, Pertovaara M, Isometsä E, Kalimo R, Nykyri E, et al. The association between burnout and physical illness in the [49] Ahola K, Honkonen T, Kivimäki M, Virtanen M, Isometsä E, Aromaa general population — results from the Finnish Health 2000 Study. J A, et al. Contribution of burnout to the association between job strain and depression: the Health 2000 Study. J Occup Environ Med 2006;48: [29] Appels A, Schouten E. Burnout as a risk factor for coronary heart [50] Ahola K, Honkonen T, Virtanen M, Aromaa A, Lönnqvist J. Burnout [30] Melamed S, Shirom A, Toker S, Shapira I. Burnout and risk of type 2 in relation to age in adult working population. J Occup Health 2008;50: diabetes: a prospective study of apparently healthy employed persons.
[51] Martelin T, Koskinen S, Valkonen T. Mortality. In: Koskinen S, [31] Mohren DC, Swaen GM, Kant IJ, van Amelsvoort LG, Born PJ, Aromaa A, Huttunen J, et al, editors. Health in Finland. Helsinki: Galama JM. Common infections and the role of burnout in a Dutch National Public Health Institute, 2006. pp. 48–51.
working population. J Psychosom Res 2003;55:201–8.
[52] Siegrist J, Rödel A. Work stress and health risk behavior. Scand J [32] Toppinen-Tanner S, Ojajärvi A, Väänänen A, Kalimo R, Jäppinen P.
Work Environ Health 2006;32:473–81.
Burnout as a predictor of medically certified sick-leave absences and [53] Kouvonen A, Kivimäki M, Väänänen A, Heponiemi T, Elovainio M, their diagnosed causes. Behav Med 2005;31:18–27.
Ala-Mursula L, et al. Job strain and adverse health behaviors: the [33] Ahola K, Hakanen J. Job strain, burnout and depressive symptoms: a Finnish public sector study. J Occup Environ Med 2007;49:68–74.
prospective study among dentists. J Affect Disord 2007;104:103–10.
[54] Wulsin LR, Vaillant GE, Wells VE. A systematic review of the [34] Armon G, Shirom A, Shapira I, Melamed S. On the nature of burnout– mortality of depression. Psychosom Med 1999;61:6–17.
insomnia relationships: a prospective study of employed adults. J [55] Cuijpers P, Smit F. Excess mortality in depression: a meta-analysis of community studies. J Affect Disord 2002;72:227–36.
[35] Kivimäki M, Virtanen M, Elovainio M, Kouvonen A, Väänänen A, [56] Palmer KT, Harris EC, Coggon D. Chronic health problems and risk of Vahtera J. Work stress in the etiology of coronary heart disease — a injury in the workplace: a systematic literature review. Occup Environ meta-analysis. Scand J Work Environ Health 2006;32:431–42.
[36] Kilpatrick B, Ross DE. The trap of protopathic bias in neuropsychiatric [57] Musselman DL, Evans DL, Nemeroff CB. The relationship of research. Biol Psychiatry 1997;41:257–8.
depression to cardiovascular disease: epidemiology, biology, and [37] de Lange AH, Taris TW, Kompier MA, Houtman IL, Bongrs PM.
treatment. Arch Gen Psychiatr 1998;55:580–92.
Different mechanisms to explain the reversed effects of mental health [58] Everson-Rose SA, Lewis TT. Psychosocial factors and cardiovascular on work characteristics. Scand J Work Environ Health 2005;31:3–14.
diseases. Annu Rev Public Health 2005;26:469–500.
[38] Chandola T, Britton A, Brunner E, Hemingway H, Malik M, Kumari [59] Kitzlerová E, Anders M. The role of some new factors in the M, et al. Work stress and coronary heart disease: what are the pathophysiology of depression and cardiovascular disease: overview of mechanisms? Eur Heart J 2008;29:640–8.
recent research. Neuro Endocrinol Lett 2007;28:832–40.
[39] Kiecolt-Glaser JK, McGuire L, Robles TF, Glaser R. Emotions, [60] Wilk J, West JC, Rae DS, Regier DA. Relationship of comorbid morbidity, and mortality: New perspectives from psychoneuroimmu- substance and alcohol use disorders to disability among patients in nology. Annu Rev Psychol 2002;53:83–107.
routine psychiatric practice. Am J Addict 2006;15:180–5.
[40] Kivimäki M, Theorell T, Westerlund H, Vahtera J, Alfredsson L. Job [61] Lee RT, Ashforth BE. A meta-analytic examination of the correlates of strain and ischaemic disease: does the inclusion of older employees in the three dimensions of job burnout. J Appl Psychol 1996;81:123–33.
the cohort dilute the association? The WOLF Stockholm Study. J [62] Appels A, Falger PR, Schouten EG. Vital exhaustion as a risk indicator Epidemiol Community Health 2008;62:372–4.
for myocardial infarction in women. J Psychosom Res 1993;37: [41] Kalimo R, Toppinen S. Organizational well-being: ten years of research and development in a forest industry corporation. In: Kompier [63] Lyall M, Peakman M, Wessly SA. A systematic review and critical M, Cooper C, editors. Preventing stress, improving productivity: evaluation of the immunology of chronic fatigue syndrome. J European case studies in the work place. London: Routledge, 1999.
[64] Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job [42] Väänänen A, Koskinen A, Joensuu M, Kivimäki M, Vahtera J, demands–resources model of burnout. J Appl Psychol 2001;86: Kouvonen A, et al. Lack of predictability at work and risk of acute myocardial infarction: an 18-year prospective study of industrial [65] Bakker AB, Demerouti E, de Boer E, Schaufeli WB. Job demands and employees. Am J Public Health 2008;98:2264–71.
job resources as predictors of absence duration and frequency. J Vocat [43] Statistics Finland. PX-Web-data bases. Helsinki: Statistics Finland, [66] Bartley M, Owen C. Relation between socioeconomic status, employment, and health during economic change, 1973–93. BMJ [44] Kalimo R, Hakanen J, Toppinen-Tanner S. Maslachin yleinen työuupumuksen arviointimenetelmä MBI-GS. (The Finnish version [67] Leiter MP, Schaufeli WB. Consistency of the burnout construct across of Maslach's Burnout Inventory–General Survey). Helsinki: Finnish occupations. Anxiety Stress Coping 1996;9:229–43.
Institute of Occupational Health, 2006.
[68] Toppinen-Tanner S, Kalimo R, Mutanen P. The process of burnout in [45] Taris TW, Schreurs PJG, Schaufeli WB. Construct validity of the white-collar and blue-collar jobs: eight-year prospective study of Maslach Burnout Inventory–General Survey: two sample examination exhaustion. J Organ Behav 2002;23:555–70.
of its factor structure and correlates. Work Stress 1999;13:223–37.
[69] Baruch Y, Holtom BC. Survey response rate levels and trends in [46] Schutte N, Toppinen S, Kalimo R, Schaufeli WB. The factorial validity organizational research. Human Relations 2008;61:1139–60.
of the Maslach Burnout Inventory–General Survey (MBI-GS) across [70] Kujala UM, Kaprio J, Koskenvuo M. Modifiable risk factors as occupational groups and nations. J Occup Organ Psychol 2000;73: predictors of all-cause mortality: the roles of genetics and childhood environment. Am J Epidemiol 2002;156:985–93.

Source: http://www.management.tau.ac.il/Eng/_Uploads/dbsAttachedFiles/RP_157_Shirom.pdf