Lexis, Monique A. S. MSc; Jansen, Nicole W. H. PhD; van Amelsvoort, Ludovic G. P. M. PhD; van den Brandt, Piet A. PhD; Kant, IJmert PhD
* Outline previous research findings on depressive complaints in the workforce, including the relationship between depression and work absences because of illness.
* Summarize the new findings on how different levels of depressive complaints are related to the rate and characteristics of sickness absence (including time to onset and duration of absences).
* Discuss the implications of the new findings on depressive complaints for efforts to prevent future sickness absence.
Depression is a common mental disorder affecting about 121 million people worldwide. According to the World Health Organization, depression will be the second leading cause of morbidity by the year 2020.1 Annually, 5.8% of the Dutch population meet the Diagnostic and Statistical Manual, 4th revision (DSM-IV) criteria of depressive disorder.2 Depending on the definition and measurement of depression reported prevalences of depression and depressive complaints greatly vary in the literature. In the current study, the focus is on a relatively mild level of depressive complaints, which may be viewed as a possible precursor for depressive disorder.3 Depressive complaints are characterized by relatively mild to moderate symptoms of depression which have not been classified as a depression according to the commonly used definitive standard, the diagnostic interview based on the DSM-IV criteria. In this perspective depressive complaints may be viewed as a continuum of complaints, ranging from no depressive complaints at one end of the spectrum to severe complaints at the other end. The continuum encompasses the total range of complaints that precedes the diagnosis of clinical depression. An annual Dutch report on self-reported depressive complaints showed a prevalence of approximately 10% percent in the general Dutch population over the period 2002–2005. A prevalence of 8.4% was found for men and 13.2% for women.4 In the working population depressive complaints are also highly prevalent. Studying the continuum of depressive complaints is particularly relevant in the labor force. Andrea et al found that depressive complaints, measured by the screening questionnaire Hospital Anxiety and Depression Scale, were even more prevalent in the working population than the general Dutch population. In their study, based on the same cohort as our study, a classification was made based on different levels of depressive complaints. For men, 18.7% of employees were classified with mild depressive complaints and 7.1% with moderate depressive complaints. For women, the prevalences were 15.8% and 6.2%, respectively.5
Depressive complaints among employees are a very important topic in the labor force, because they contribute considerably to health care use, loss of productivity, and disability days.3,6 Since employees with a clinical depression are disabled from work in many cases, a strategy aimed at early identification and treatment of employees with especially mild depressive symptoms might reduce the risk of developing severe depressive complaints and future long-term sickness absence. Consequently, this strategy may also reduce the costs of mental health problems in the workforce.3,7–9 Nevertheless, the majority of studies on depression among the workforce have focused on major depression as a risk factor for sickness absence.5,10–13 Depression is associated with an increased risk of sickness absence for both men and women.8,14,15 Pronounced gender differences are also reported in the literature concerning sickness absence due to depression. In most studies, the risk of sickness absence because of depression appeared higher among women than among men,13,16–19 but not all studies are consistent on this. Discrepancies between men and women regarding the duration of sickness absence also exist. Most studies show longer sickness absence spells in women than in men.20–22 However, other studies show opposite results.14,23
Although many studies have been conducted on the relation between depression and sickness absence, less is known about the impact of different levels of depressive complaints on sickness absence. In a longitudinal study, Stansfeld et al13 showed that minor psychiatric disorders, including minor depression, were one of the most common reasons for sickness absence among civil servants. In a prospective study conducted in the Danish general working population, a more than two times higher risk of long-term sickness absence was found in both men and women with severe depressive symptoms, compared with subjects without depressive symptoms.10 Many studies showed that a higher level of depressive symptoms was also related to longer sickness absence.12,24–26 Workers with major depression were found to have between 1.5 and 3.2 times more short-term disability days over a 30-day period than other workers.11
Information is also scarce on duration and time to onset of sickness absence. This is of particular interest, because duration and time to onset of sickness absence encompass different information. Especially, short-term sickness absence is regarded as a coping strategy to reduce health complaints and thus to prevent more serious ill health.27 Low or mild levels of health complaints, such as a cold, generally correspond with more frequent but short-term sickness absence. There is a strong relationship between absence frequency and time to onset of a sick leave episode. Short-term sickness absence is predictive of a quick onset of the first sickness absence spell. More severe health complaints, however, correspond with less frequent but long-term sickness absence.28 Individuals with more severe depressive symptoms are more likely to report days of work loss due to depression.29–31
The aim of the present article is to study both cross sectionally and prospectively the relationships between depressive complaints and sickness absence from work. We hypothesize that relatively mild depressive complaints will result in a longer duration and a shorter time to onset of the first sickness absence spell and that there will be a dose-response relationship between depressive complaints and sickness absence.
Although cross-sectional analyses using self-reported sickness absence data give insight into the effects that are already present at baseline measurement, the present study will additionally focus on the prospective relationship between depressive complaints and objective sickness absence by using data through company record linkage. We are particularly interested whether different levels of depressive complaints are predictive of increased sickness absence, in terms of time to onset of sickness absence and sickness absence duration.
When conducting a study on levels of depressive complaints in relation to sickness absence, one should carefully consider the multifactorial etiology of sickness absence.27 Hence, one should take into account differences in demographic characteristics, work-related factors and health status.
Maastricht Cohort Study
Data were used from an ongoing large-scale prospective epidemiological cohort study in The Netherlands, addressing a broad range of work-related and non–work-related factors, individual factors, and (mental) health status. At baseline in May 1998, this Maastricht Cohort Study on Fatigue at Work surveyed a population of 12,140 employees from 45 different companies and organizations, representing a baseline response rate of 45%. Employees were followed up by means of 10 consecutive self-administered questionnaires. Baseline characteristics of the study population and a non-response analysis have been described elsewhere.32
The scale used to measure depressive complaints was first included in the ninth questionnaire (T8) of the study, in January 2001. Therefore, T8 will constitute the baseline measurement for the current study. At T8, 8033 employees received the questionnaire and, of these, 7482 employees completed and returned the questionnaire (response 93.1%). First, 77 records with incomplete or incorrect data on self-reported sickness absence variables were removed from the database. Second, female employees who were pregnant at the time of completing the questionnaire (n = 28) were excluded to avoid counting spells specifically related to pregnancy leave, resulting in a study population of 7377 employees, 1992 women and 5385 men for the cross-sectional analyses.
Longitudinal data derived from register-based information on sickness absence from the participating companies and organizations in the Maastricht Cohort Study were collected between 1998 and 2001. For the original Maastricht Cohort Study, 45 companies agreed to provide us with sickness absence data until December 2000. Thirteen out of the 45 companies were able to provide us with sickness absence data for one additional year, the year 2001, which were needed for the current study. Objective sick leave data were available for 3580 participants at T8. In this case, selective dropout of participants is rather unlikely to occur since objective sickness absence data concern data on company level instead of an individual level. For the prospective analyses, the same exclusion criteria were applied as for the cross-sectional analyses, and we additionally excluded those employees who reported themselves fully or partially absent from work at T8 to study incident sickness absence exclusively. Women were also excluded if they were on sick leave because of pregnancy or maternity leave during follow-up. After application of the exclusion criteria, the study population for the prospective analyses included 3339 employees, of which 2666 were men and 673 were women.
The Hospital Anxiety and Depression (HAD) scale was used to measure the presence and severity of depressive complaints. The HAD is a 14-item self-report questionnaire that was originally developed to indicate the presence and severity of both anxiety (HAD-A) and depression (HAD-D) separately.33 Both the HAD-A and the HAD-D consist of 7 items which are scored on a four-point Likert scale (0 to 3), resulting in a range of 0 to 21. In this study, only the HAD-D was used.
A score on the Likert scale represents a combination of the number and severity of depressive complaints. A higher score on the HAD-D is indicative for having more severe depressive complaints. Three categories of depressive complaints were originally defined by Zigmond and Snaith in 1983. A score of less than 8 points was defined as noncase of depression; a score of 8 points and more was defined as a possible case of depression; and 11 points or more as a subclinical case of depression.33,34 Although the HAD scale was originally developed to identify (possible) caseness of anxiety disorders and depression among patients in nonpsychiatric hospital clinics, the questionnaire was recently found to perform well in assessing the symptom severity and caseness of anxiety disorders and depression in somatic, psychiatric and primary care patients, and in the general population. Cronbach alphas for HAD-D were 0.86 and 0.85 for women and men, respectively.5 In subsequent research on the HAD scale, Snaith did not present the categories as a probability of the presence of depression but presented a different definition of the categories. These categories, which are mutually exclusive, were presented as a division of depressive complaints into three ranges: 1) normal (less than 8 points), further used as reference range in this study, 2) mild (8 to 10 points), and 3) moderate and severe (11 points or more).35 The latter description is in line with our approach of depressive complaints, as defined earlier, as a continuum.
Data about sickness absence were gathered from objective organizational absence records and also from the self-report questionnaire. In the present study, data on self-reported sickness absence were only used to perform the cross-sectional analyses and for exclusion of prevalent cases for the prospective analyses. As outcome measure for the longitudinal analyses, we used objective company-registered sickness absence data, which were obtained by record linkages with the company sickness absence registry systems. For the present study, sickness absence data from the year 2001 were used. After sending out the questionnaires in January 2001, a follow-up period of 10 months (March to December 2001) was maintained. A time lag of 2 months was used between the time of employees receiving the questionnaire and the measurement of sickness absence, as the time of the returning of the questionnaires and the processing of the data at most could take 2 months. All information regarding time to onset of first sickness absence spell and total number of days absent from work over the 10 months of follow-up was measured through record linkage on an individual level with the company registers on sickness absence.
Potential Confounding Factors.
Several potential confounding factors should be controlled for when studying the relationship between depressive complaints and sickness absence. Because of earlier reported gender differences both with regard to depressive complaints and sickness absence, all analyses will be conducted for men and women separately.
In this study, several domains of potential confounding factors were taken into account, that is demographic factors, health, work-related factors and lifestyle factors. Educational level (low, medium, high), smoking (yes/no), living alone (yes/no), the presence of a long-term illness (psychological illnesses excluded) (yes/no), and working in shifts (yes/no) were taken into account as potential confounding factors.6,27,36,37 Information on these factors was gathered through the questionnaires. Age and educational level were assessed at baseline T0 (May 1998), whereas the presence of a long-term illness and living situation were assessed in the T6 cohort questionnaire of May 2000. Information on smoking and working in shifts was present in the questionnaire T8.
Statistical analyses were performed with SPSS 13.0 and SAS. All analyses were stratified for gender. Logistic regression analyses were performed to study the cross-sectional relation between depressive complaints and sickness absence at the time of completing the questionnaire T8. Because the distribution of total number of days absent from work more than 10 months was skewed like a Poisson distribution, Poisson regression analyses were used to test differences in number of sickness absence days between employees with different levels of depressive complaints. Multivariate survival analyses using Cox regression were conducted to examine prospective effects of depressive complaints, in which we modeled the time to first sickness absence spell from work over the 10-month period following the questionnaire in January 2001. In all analyses, adjustments were applied in three steps. In the first step, odds ratios or hazard ratios (ORs and HRs, respectively) and 95% confidence intervals (95% CI) were calculated for depressive complaints adjusted for age. In the second step, additional adjustments were made for the presence of a long-term illness and for smoking. In the third step, additional adjustments were made for educational level, living alone, and working in shifts. Other statistical procedures included χ2 tests and analysis of variance.
We first studied the relationship between the level of depressive complaints (HAD-D continuous) and sickness absence (yes/no). The average score on the HAD-D for men who were on sick leave at T8 was 6.73 (SD = 4.53) and 5.84 (SD = 4.45) for women. The average HAD-D score for employees not on sick leave at T8 was 3.84 (SD = 3.65) for men and 3.37 (SD = 3.49) for women.
Table 1 presents descriptive characteristics of the study population, categorized by the level of depressive complaints.
As shown in Table 1, in both men and women, statistically significant differences were observed between the three categories of depressive complaints with respect to demographic, health, and work-related factors. In men, all factors, except living alone were statistically significant with depressive complaints, with employees reporting mild or moderate-severe complaints having a higher mean age, a higher percentage of lower educational level, a higher proportion of long-term illness, smokers, and shift workers. In women, those reporting mild or moderate- severe complaints had a statistically significant higher mean age, a higher percentage of long-term illness and a higher proportion of employees living alone compared with employees scoring within the reference range. The proportion of employees absent from work at the time of completing the baseline was also statistically significant associated with depressive complaints in both men and women, again with those reporting mild or moderate-severe complaints having a significantly higher proportion of sickness absence at the time of completing the questionnaire T8 compared with employees scoring within the reference range.
To explore the cross-sectional associations between depressive complaints and sickness absence at the time of completing the questionnaire, we first studied the effects of one point increase on the HAD-D scale to use the full continuum of the scale. In both men and women statistically significant associations were found. In men, the OR was 1.15 (95% CI = 1.11 to 1.19) and in women 1.16 (95% CI = 1.11 to 1.21) after adjusting for age, long-term illness, smoking, educational level, living alone, and shift work.
Table 2 shows the cross-sectional association between the categories of depressive complaints and sickness absence at the time of completing the baseline questionnaire. In both men and women ORs were statistically significant and relevant for social-medical counseling. ORs increased remarkably as the level of depressive complaints increased. In women, the highest odds of sickness absence was observed among employees with moderate-severe complaints compared with those scoring within the reference range. In men, the highest odds of sickness absence was observed among employees with mild complaints compared with those scoring within the reference range.
For the prospective analyses, over 10 months of follow up, we examined the effect of depressive complaints on time to onset of first sickness absence spell, irrespective of sickness duration, based on company-registered sick leave data. Again, we first studied the effects of one point increase on the HAD-D scale to use the full continuum of the scale. In both men and women, a significant association between depressive complaints and the time to onset of the first sickness absence spell was found. In men, the HR was 1.03 (1.01 to 1.05) and 1.04 (1.01 to 1.07) in women after adjusting for age, long-term illness, smoking, educational level, living alone, and shift work. The mean time to onset of first sickness absence spell in women was 212 days for employees scoring within the reference range (SD = 117.61), 190 days (SD = 119.03) for those with mild complaints and 183 days (SD = 118.96) for those with moderate-severe complaints. For men, this was 241 days for employees scoring within the reference range (SD = 105.88), 226 days for those with mild complaints (SD = 114.84), and 220 days for those with moderate-severe complaints (SD = 114.77).
Table 3 presents prospective relations between the three categories of depressive complaints and time to onset of first sickness absence spell in the 10 months (March to December) following the questionnaire in January 2001.
For both men and women, the HRs for the relation between depressive complaints and time to onset of the first sickness absence spell were all in the expected direction, in which those with mild or moderate-severe complaints had a higher risk of going on sick leave earlier than employees scoring within the reference range. For women, only one significant and relevant association was found for those with moderate-severe complaints when adjusted for age. This is probably due to the small number of employees having mild or moderate-severe complaints among women. For men, the HRs for moderate-severe depressive complaints, adjusted for all the confounding factors, were relevant and reached statistical significance.
To analyze duration of sickness absence, Poisson regression analyses were used. Again first, the effects of one point increase on the HAD-D scale were studied. Table 4 shows statistically significant β's in both men and women after adjusting for age, long-term illness, smoking, educational level, living alone, and shift work. These results show a statistically significant association between depressive complaints and sickness absence duration. Similar results were found after performing the analyses with the classification into the three categories of depressive complaints (results not displayed). In both men and women, the risk of a longer duration of sickness absence increased as the level of depressive complaints increased.
Translated to daily practice the higher risk for a longer duration of sickness absence revealed itself in the mean number of absent days in the three categories of depressive complaints, as is shown in Table 5. This Table shows substantial and relevant differences in the average number of absent days from work over 10 months of follow-up among employees with different levels of depressive complaints. The average number of absent days between the different levels of depressive complaints was statistically significant, except in women with moderate-severe complaints. This may be due to the low power in this subgroup. The results were most pronounced for women with mild depressive complaints, where the average number of sickness absence days over 10 months of follow-up was almost 2.5 times as high as the average number of sickness absence days for employees scoring within the reference range.
Although several studies investigated and found a relation between major depression and sickness absence, this study has shown clear cross-sectional and prospective relations between levels of depressive complaints and future sickness absence, even after adjusting for several important confounding factors. Hence, this study suggests that less severe levels of depressive complaints also constitute risk factors for sickness absence, which is an important finding for the development of preventive measures with respect to sickness absence.
Because cross-sectional associations do not allow assertions on the specific causality of associations between depressive complaints and sickness absence, the prospective analyses suggest that depressive complaints can be considered as a predictor of sickness absence. In particular, the cross-sectional associations were very pronounced. This may indicate that a large part of the effect of depressive complaints on sickness absence may already have been present at the baseline measurement (time of completing questionnaire T8). Because of these high cross-sectional associations, the results from the longitudinal analyses may have been underestimated.
In 2002, the annual total sickness absence rate in the general Dutch population was 4.8% in men and 6.0% in women.38 The percentages of employees scoring within the reference range of depressive complaints and who were absent from work at the time of completing T8 are fairly consistent with these data, 4.1% for men and 7.4% for women. In men, 17.2% of the employees with moderate-severe complaints and 9.7% of the employees with mild depressive complaints were absent from work at the time of completing the questionnaire T8. In women, this was 28.0% of the employees with moderate-severe complaints and 16.0% of the employees with mild depressive complaints. Firm conclusions about gender differences in the relationship between depressive complaints and sickness absence cannot be drawn yet, partly because of the small number of women who participated, especially in the mild and moderate-severe complaints groups.
Although a reasonable association was found with respect to level of depressive complaints and time to onset of first sickness absence spell, pronounced differences between employees with a higher level versus a lower level of depressive complaints were observed with respect to the average number of days absent over the 10-month follow-up period. It appears that only employees with moderate-severe depressive complaints report themselves ill at an earlier point in time. The duration of sickness absence, however, increases most when employees with mild complaints and employees scoring within the reference range were compared, with twice as many absent days in men and two and a halve times more in women. This provides an indication that sickness absence as a consequence of depressive complaints occurs earlier when a person already experiences more severe complaints. For instance, when a person with mild depressive complaints does decide to go on sick leave, the duration of sickness absence increases significantly compared with the sickness absence duration of employees scoring within the reference range. Employees with mild complaints may be able to continue their work much longer without experiencing considerable restraints or they might not be aware of their psychological complaints when they first appear compared with employees with moderate-severe complaints.
In this study, depressive complaints were measured with a questionnaire instead of using a diagnostic interview. The Hospital Anxiety and Depression (HAD-D) scale was used to measure depressive complaints in the working population and to identify employees with different severity levels of depressive complaints. Bjelland et al34 found that the HAD-D performed well in assessing the severity and caseness of anxiety disorders and depression in the general population. Less is known about application of the HAD-D in the working population. Although based on self-report, Andrea et al5 showed that the HAD-D and HAD-A can be used as separate constructs to measure depressive complaints and anxiety in the working population and it is therefore sufficient to use the HAD-D in this study only. The surplus value of using the HAD scale in our study is that it enabled us to investigate both the whole continuum and the distinguished categories of depressive complaints.
The prevalence of employees with moderate-severe depressive complaints in our study was 6.9% for men and 6.2% for women. The prevalence for men in our study population seems comparable with other studies based on DSM-III-R classifications and presenting annual depression prevalences for men and women separately.5 Nevertheless, the prevalence of moderate-severe depressive complaints for women in our study seems to be lower. This lower prevalence may be caused by the fact that more women than men work part-time and less women work in shifts. Thus, they may have already adapted their work to their health status or private situation. Another explanation may be that women with mild or moderate-severe depressive complaints have already left the labor force at the time the questionnaire was distributed.
In the analyses, several adjustments were made to investigate whether the effects of depressive complaints on sickness absence could be ascribed to depressive complaints or should be attributed to other confounding factors, known to be associated with sickness absence. Therefore, we adjusted the analyses in three steps for several factors. Although the ORs in the cross-sectional analyses were reduced after controlling for these factors, similar trends generally remained. In the prospective analyses, small changes in observed effects also occurred after controlling for these confounding factors. The observed effects on sickness absence after controlling for confounding factors could then more likely be ascribed to depressive complaints. Although, the strength of the associations was hardly reduced after controlling for the selected group of confounding factors, it could be argued that there may be other important risk factors which contribute to the effects found, given the multifactorial etiology of sickness absence.27 Due to practical limitations data on all potential confounding factors were not available in the questionnaire. For this reason, it is important to control for educational level, as it is often viewed as a proxy for other factors related to occupation or lifestyle.39
The results of this study are based on data from a large-scale prospective cohort study, enabling us to study the cross-sectional and prospective relationship between levels of depressive complaints and long-term sickness absence among employees over a 10-month follow-up period. Sickness absence was measured through linkage on an individual level with the company records on sickness absence, providing us with objective sickness absence data. A time lag of 2 months was used between the time employees received the questionnaire and the measurement of sickness absence, as the time of the returning of the questionnaires and the processing of the data would take a maximum of 2 months. For the prospective analyses, it may be possible, however, that respondents were already on sick leave because of depressive complaints before the measuring of sickness absence began. This means there might be a shorter time between depressive complaints and occurrence of sickness absence as a consequence than our predefined 2 months lag time. These persons were excluded from the analyses, because prevalent cases at the start of the measurement were excluded. Therefore, the effect of depressive complaints on sickness absence could be underestimated.
The baseline population of the present study consists of participants of the Maastricht Cohort Study after 2 years of follow-up. This might raise questions about selective dropout of participants during follow-up. Analyses on the effect of nonresponse during follow-up with respect to sickness absence indicated that employees reporting themselves sick four or more times showed a lower response rate on the questionnaires as compared with those reporting no sickness absence or calling in sick once or twice. This results in an underestimation of the incidence of sick leave.40 For our study, this might also indicate an underestimation of sickness absence.
Sickness absence data that were derived from the questionnaire were used only for the cross-sectional analyses. Although measured by self-report, we argue that no recall bias has occurred since workers were asked for sickness absence at the time of completing the questionnaire. From the present study, we conclude that a clear relationship exists between depressive complaints and sickness absence. The high cross-sectional association at baseline measurement and the prospective association between depressive complaints and sickness absence indicate that the effects of mild or moderate-severe depressive complaints on sickness absence may occur rather quickly over time, resulting in a shorter time to onset of sickness absence and much longer periods of sickness absence.
Because of this and the high prevalence of mild and moderate-severe depressive complaints in the labor force, prevention of depressive complaints might be beneficial in preventing future sickness absence.
The Maastricht Cohort Study is part of the Netherlands concerted research action on “Fatigue at Work” granted by the Netherlands Organization for Scientific Research.
This study was supported by the Health Research and Development Council (Zorg Onderzoek Nederland) grant no. 62200024, and by the School for Public Health and Primary Care (CAPHRI), Maastricht, The Netherlands.
1. World Health Organization. World Health Report 2000. Geneva, Switzerland: World Health Organization; 2000.
2. Bijl RV, Ravelli A, van Zessen G. Prevalence of psychiatric disorder in the general population: results of The Netherlands Mental Health Survey and Incidence Study (NEMESIS). Soc Psychiatry Psychiatr Epidemiol. 1998;33:587–595.
3. Cuijpers P, de Graaf R, van Dorsselaer S. Minor depression: risk profiles, functional disability, health care use and risk of developing major depression. J Affect Disord. 2004;79:71–79.
4. Statistics Netherlands. Zelfgerapporteerde medische consumptie, gezondheid en leefstijl. [Self-Reported Medical Consumption, Health and Lifestyle]. Heerlen/Voorburg, The Netherlands: Statistics Netherlands. Available at: http://www.statline.cbs.nl
. Accessed 20 June 2008.
5. Andrea H, Bultmann U, Beurskens AJ, Swaen GM, van Schayck CP, Kant IJ. Anxiety and depression in the working population using the HAD Scale—psychometrics, prevalence and relationships with psychosocial work characteristics. Soc Psychiatry Psychiatr Epidemiol. 2004;39:637–646.
6. Lerner D, Henke RM. What does research tell us about depression, job performance, and work productivity? J Occup Environ Med. 2008;50:401–410.
7. Jenkins R. Mental illness and work. In: Floyd M, Povall M, Watson G, eds. Mental Health at Work. London: Jessica Kingsley; 1994:77–79.
8. Simon GE, Chisholm D, Treglia M, Bushnell D. Course of depression, health services costs, and work productivity in an international primary care study. Gen Hosp Psychiatry. 2002;24:328–335.
9. Smit F, Beekman A, Cuijpers P, de Graaf R, Vollebergh W. Selecting key variables for depression prevention: results from a population-based prospective epidemiological study. J Affect Disord. 2004;81:241–249.
10. Bultmann U, Rugulies R, Lund T, Christensen KB, Labriola M, Burr H. Depressive symptoms and the risk of long-term sickness absence: a prospective study among 4747 employees in Denmark. Soc Psychiatry Psychiatr Epidemiol. 2006;41:875–880.
11. Kessler RC, Barber C, Birnbaum HG, et al. Depression in the workplace: effects on short-term disability. Health Aff (Millwood). 1999;18:163–171.
12. Nieuwenhuijsen K, Verbeek JH, de Boer AG, Blonk RW, van Dijk FJ. Predicting the duration of sickness absence for patients with common mental disorders in occupational health care. Scand J Work Environ Health. 2006;32:67–74.
13. Stansfeld S, Feeney A, Head J, Canner R, North F, Marmot M. Sickness absence for psychiatric illness: the Whitehall II Study. Soc Sci Med. 1995;40:189–197.
14. Laitinen-Krispijn S, Bijl RV. Mental disorders and employee sickness absence: the NEMESIS study. Netherlands Mental Health Survey and Incidence Study. Soc Psychiatry Psychiatr Epidemiol. 2000;35:71–77.
15. Lerner D, Adler DA, Chang H, et al. Unemployment, job retention, and productivity loss among employees with depression. Psychiatr Serv. 2004;55:1371–1378.
16. Hensing G, Alexanderson K, Akerlind I, Bjurulf P. Sick-leave due to minor psychiatric morbidity: role of sex integration. Soc Psychiatry Psychiatr Epidemiol. 1995;30:39–43.
17. Kivimaki M, Vahtera J, Thomson L, Griffiths A, Cox T, Pentti J. Psychosocial factors predicting employee sickness absence during economic decline. J Appl Psychol. 1997;82:858–872.
18. Laaksonen M, Martikainen P, Rahkonen O, Lahelma E. Explanations for gender differences in sickness absence: evidence from middle-aged municipal employees from Finland. Occup Environ Med. 2008;65:325–330.
19. Kopp MS, Skrabski A, Szedmak S. Socioeconomic factors, severity of depressive symptomatology, and sickness absence rate in the Hungarian population. J Psychosom Res. 1995;39:1019–1029.
20. Koopmans PC, Roelen CA, Groothoff JW. Sickness absence due to depressive symptoms. Int Arch Occup Environ Health. 2008;81:711–719.
21. North F, Syme SL, Feeney A, Head J, Shipley MJ, Marmot MG. Explaining socioeconomic differences in sickness absence: the Whitehall II Study. BMJ. 1993;306:361–366.
22. Feeney A, North F, Head J, Canner R, Marmot M. Socioeconomic and sex differentials in reason for sickness absence from the Whitehall II Study. Occup Environ Med. 1998;55:91–98.
23. Hensing G, Alexanderson K, Allebeck P, Bjurulf P. Sick-leave due to psychiatric disorder: higher incidence among women and longer duration for men. Br J Psychiatry. 1996;169:740–746.
24. Hardy GE, Woods D, Wall TD. The impact of psychological distress on absence from work. J Appl Psychol. 2003;88:306–314.
25. Spijker J, de Graaf R, Bijl RV, Beekman AT, Ormel J, Nolen WA. Duration of major depressive episodes in the general population: results from The Netherlands Mental Health Survey and Incidence Study (NEMESIS). Br J Psychiatry. 2002;181:208–213.
26. Vaananen A, Toppinen-Tanner S, Kalimo R, Mutanen P, Vahtera J, Peiro JM. Job characteristics, physical and psychological symptoms, and social support as antecedents of sickness absence among men and women in the private industrial sector. Soc Sci Med. 2003;57:807–824.
27. Alexanderson K. Sickness absence: a review of performed studies with focused on levels of exposures and theories utilized. Scand J Soc Med. 1998;26:241–249.
28. Janssen N, Kant IJ, Swaen GM, Janssen PP, Schroer CA. Fatigue as a predictor of sickness absence: results from the Maastricht cohort study on fatigue at work. Occup Environ Med. 2003;60:71–76.
29. Kessler RC, Akiskal HS, Ames M, et al. Prevalence and effects of mood disorders on work performance in a nationally representative sample of U.S. workers. Am J Psychiatry. 2006;163:1561–1568.
30. Kessler RC, Frank RG. The impact of psychiatric disorders on work loss days. Psychol Med. 1997;27:861–873.
31. Lerner D, Adler DA, Chang H, et al. The clinical and occupational correlates of work productivity loss among employed patients with depression. J Occup Environ Med. 2004;46:46–55.
32. Kant IJ, Bultmann U, Schroer KA, Beurskens AJ, Van Amelsvoort LG, Swaen GM. An epidemiological approach to study fatigue in the working population: the Maastricht Cohort Study. Occup Environ Med. 2003;60:32–39.
33. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67:361–370.
34. Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res. 2002;52:69–77.
35. Snaith RP. The Hospital Anxiety and Depression Scale. Health Qual Life Outcomes. 2003;1:29.
36. Allebeck P, Mastekaasa A. Swedish Council on Technology Assessment in Health Care (SBU). Chapter 3. Causes of sickness absence: research approaches and explanatory models. Scand J Public Health Suppl. 2004;63:36–43.
37. Allebeck P, Mastekaasa A. Swedish Council on Technology Assessment in Health Care (SBU). Chapter 5. Risk factors for sick leave—general studies. Scand J Public Health Suppl. 2004;63:49–108.
38. Statistics Netherlands. National Statistics on Sick Leave, Frequency, Period of Absence
. Heerlen/Voorburg, The Netherlands: Statistics Netherlands. Available at: http://www.statline.cbs.nl
. Accessed at 11 June 2008.
39. Statistics Netherlands. Standaard Beroepenclassificatie 1992 [The Netherlands Standard Classification of Occupations 1992]. 's-Gravenhage: SDU/uitgeverij; 1993.
40. Mohren D, Jansen N, Amelsvoort L, Kant IJ. An Epidemiological Approach of Fatigue and Work: Experiences from the Maastricht Cohort Study. Maastricht: Programma Epidemiologie van Arbeid en Gezondheid; 2007:78–79.