Abstract
Objectives. We determined the proportion of workers meeting criteria for major depressive episodes in the past year and examined the association between psychosocial work-stress variables and these episodes.
Methods. Data were derived from the Canadian Community Health Survey 1.2, a population-based survey of 24324 employed, community-dwelling individuals conducted in 2002. We assessed depressive episodes using the Composite International Diagnostic Interview.
Results. Of the original sample, 4.6% (weighted n=745948) met criteria for major depressive episodes. High job strain was significantly associated with depression among men (odds ratio [OR]=2.38; 95% confidence interval [CI]=1.29, 4.37), and lack of social support at work was significantly associated with depression in both genders (men, OR=2.70; 95% CI=1.55, 4.71; women, OR=2.37; 95% CI=1.71, 3.29). Women with low levels of decision authority were more likely to have depression (OR=1.59; 95% CI=1.06, 2.39) than were women with high levels of authority.
Conclusions. A significant proportion of the workforce experienced major depressive episodes in the year preceding our study. Gender differences appear to affect work-stress factors that increase risk for depression. Prevention strategies need to be developed with employers and employee organizations to address work organization and to increase social support.
Depression is the leading cause of disability and is projected to become the second leading cause of the global burden of disease by 2020.1 Lifetime and 12-month (having had an episode in the past year) prevalence rates of major depressive disorder are estimated at 12.8%2 to 16.6%3 (16.5% of women, 8.9% of men2) and 3.9%2 to 6.7%4 (5.0% of women, 2.6% of men2), respectively. A substantial proportion of those affected by this disorder are of working age, but little is known about its prevalence or risk factors within the general working population. One of the few epidemiological studies conducted on the subject reported that 15.7% of employed individuals (19.5% of women, 11.4% of men) met lifetime criteria for major depressive disorder and that 8.6% (10.2% of women, 5.9% of men) met 12-month criteria.5
Changes in workplace structure over the past 30 years may have contributed to increased stress and psychiatric morbidity.6 Aging of the population, coupled with a shortage of younger workers in some sectors, has also resulted in many people remaining in the workforce beyond usual retirement age. Depression at work reduces employees’ productivity,7 increases disability and depression-related absence, and may lead to premature early retirement.8 The prevalence and associated costs of depression necessitate an increased understanding of work-related factors that may contribute to this condition.
Previous studies had 2 common methodological limitations that our study addressed. First, most studies have focused on distinct occupational groups9,10 rather than on the general population.5,11 Second, most studies have relied on self-report questionnaires to measure depressive symptoms rather than on standardized diagnostic interviews. We determined the proportion of employed individuals who met the 12-month criteria for a major depressive episode as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV )12 and examined the relationship between psychosocial work characteristics and 12-month major depressive episodes. We hypothesized that individuals experiencing increased job strain (higher job demands and lower decision latitude) would be at higher risk for 12-month major depressive episodes.
METHODS
Data and Sample
The data for this study were obtained from the Canadian Community Health Survey 1.2: Mental Health and Well Being, a large epidemiological national survey of community-dwelling individuals aged 15 years and older conducted by Statistics Canada in 2002. Residents of the territories, Indian reserves, institutions, and certain remote areas and full-time members of the armed forces were not included. A multistage stratified cluster design was used to sample dwellings, and 1 person aged 15 years or older was randomly selected from each sampled household. Details of the design of the survey and the questionnaires are published elsewhere.13,14
The responding sample totaled 36984, with a response rate of 77%. Our study included individuals aged 15 to 75 years who were employed, which was defined as having worked at a job or business in the preceding 12 months, yielding an unweighted sample of 24324. To make estimates produced from the survey data representative of the covered population, and not just the sample itself, the data were weighted, and these were used in the calculations. Each respondent in the sample was given a sampling weight proportional to the portion of the Canadian population he or she represented. Therefore, our sample of 24 324 represents a weighted population of 16353 858 Canadians.
Measures
Psychiatric symptomatology was assessed with the World Mental Health 2000 version of the Composite International Diagnostic Interview,15,16 a fully structured interview designed for use by trained lay interviewers. Both lifetime (recurrent) and 12-month major depressive episodes were assessed by DSM-IV diagnosis. All interviews were conducted by trained interviewers who used a computer-assisted application; 86% were conducted in person, the remainder by telephone.
Karasek and Theorell’s job-strain model,17 the most widely tested and validated model of work-related stress and health, was used to assess psychosocial work stress. Six subscales were included: decision authority (3 items), skill discretion (3 items), psychological demands (3 items), work social support (4 items), job insecurity (2 items), and physical exertion (2 items).
The model posits 2 key dimensions: psychological job demands and decision latitude. Decision latitude comprises decision authority (control over work) and skill discretion (variety of work and the opportunity to use skills). According to the model, the worst combination for health is to have high demands and low decision latitude. There is consistent evidence that working in a high-strain job (one high in work demands but low in control and decision latitude) is associated with an increased risk of depressive symptoms.11,18,19 High levels of decision latitude and social support appear protective in both cross-sectional11 and longitudinal studies,18,19 and lack of decision authority is associated with depression.11
Income adequacy is divided into 5 groups, ranging from low to high, and is based on the number of people in the household and the total household income from all sources.
Respondents were asked if they experienced any chronic health conditions that had been present for a minimum of 6 months and had been diagnosed by a health professional. Responses were categorized by the presence or absence of a chronic condition.
Statistical Analysis
Weighted cross-tabulations were used to estimate the number and proportion of people with depression. Work-stress variables were categorized a priori into 5 equidistant categories of the range; however, because sample sizes in the 2 highest categories were small, these categories were collapsed for job strain, social support, job security, skill discretion, and decision authority. We used logistic regression to investigate associations between each work-stress variable and depression, adjusting for age, marital status, White race, income adequacy, presence of a chronic health condition, education, antidepressant medication use in the past 12 months, and other work-stress variables.
The significance of each work-stress variable was determined using the Wald χ2 test for the overall effect (i.e., linear trend). We used tolerances and variance inflation factors to investigate multicollinearity and the Hosmer–Lemeshow goodness-of-fit statistic to examine model fit.20 To account for survey-design effects, we estimated the variance used in the calculation of prevalence estimates, coefficients of variation (standard error of the estimate), and confidence limits using the bootstrap technique. All analyses were performed with SAS version 9.1 (SAS Institute Inc, Cary, NC).
RESULTS
The majority of the respondents were White (84%), married or cohabiting (65%), in the upper-middle or highest levels of income adequacy (77%), and had postsecondary education (55%). There were slightly more men than women in the sample (54% vs 46%), but there were no significant differences in demographic variables when analyzed by gender. Demographic characteristics of the sample, divided by major depressive episode status, are shown in Table 1 ▶.
TABLE 1—
Sample Demographic Characteristics and Major Depressive Episodes: Canadian Community Health Survey 1.2, 2002
| Respondents With Depression, a No. (%) | OR (95% CI) | |
| Overall | 745 948 (4.6) | |
| Gender | ||
| Women | 450 061 (6.0) | 1.83 (1.53, 2.20) |
| Men | 295 887 (3.4) | 1.00 |
| Age, y | ||
| < 18 | 30 225 (5.0) | 1.08 (0.68, 1.72) |
| 18–24 | 139 271 (6.6) | 1.46 (1.13, 1.89) |
| 25–34 | 172 273 (5.1) | 1.1 (0.88, 1.38) |
| 35–44 | 213 479 (4.6) | 1.00 |
| 45–54 | 138 654 (3.8) | 0.81 (0.61, 1.08) |
| ≥55 | 52 046 (2.6) | 0.54 (0.38, 0.77) |
| Education | ||
| Less than high school | 114 620 (4.3) | 0.05 (0.04, 0.06) |
| High school graduate | 152 133 (4.7) | 1.00 |
| Some college | 86 352 (5.8) | 0.9 (0.69, 1.17) |
| College graduate | 392 842 (4.4) | 1.25 (0.92, 1.71) |
| Income quintileb | ||
| Lowest | 28 627 (9.2) | 1.71 (1.16, 2.52) |
| Lower middle | 54 320 (7.7) | 1.42 (1.00, 2.02) |
| Middle | 150 942 (5.6) | 1.00 |
| Upper middle | 280 762 (4.6) | 0.82 (0.65, 1.03) |
| Highest | 231 297 (3.5) | 0.62 (0.48, 0.79) |
| Marital Status | ||
| Married or common law | 348 332 (3.3) | 1.00 |
| Separated/divorced/widowed | 123 002 (8.9) | 2.9 (2.33, 3.62) |
| Single/never married | 274 613 (6.3) | 2 (1.65, 2.43) |
| Race/ethnicity, White | ||
| Yes | 652 675 (4.8) | 1.4 (1.01, 1.96) |
| No | 92 817 (3.4) | 1.00 |
| Had a chronic diseasec | ||
| Yes | 369 435 (6.3) | 1.82 (1.54, 2.14) |
| No | 376 512 (3.5) | 1.00 |
| Antidepressant use in previous 12 months | ||
| Yes | 287 469 (32.8) | 16.07 (13.33, 19.37) |
| No | 457 458 (2.9) | 1.00 |
Note. OR = odds ratio; CI = confidence interval.
aDefined as respondents who had a major depressive episode in the previous 12 months, as assessed by interviewers using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV).12
bLowest quintile = 1–4 people earning < $10 000 or ≥5 people earning < $15 000. Lower middle quintile = 1–2 people earning $ 10 000–$ 14 999, 3–4 people earning $10 000–$19 999, or ≥5 people earning $15 000–$29 999. Middle quintile = 1–2 people earning $ 15 000–$29 999, 3–4 people earning $20 000–$39 999, or ≥5 people earning $30 000–$59 999. Upper middle quintile = 1–2 people earning $30 000–$59 999, 3–4 people earning $40 000–$79 999, or ≥5 people earning $ 60 000–$79 999. High quintile = 1–2 people earning > $60 000 or ≥3 people earning > $80 000. All dollars are Canadian dollars.
c The chronic conditions included were arthritis or rheumatism, a back problem (excluding fibromyalgia or arthritis), chronic bronchitis, emphysema or chronic pulmonary disease, diabetes, epilepsy, heart disease, cancer, stomach or intestinal ulcers, continuing effects from a stroke, bowel disorder or Crohn’s disease, and thyroid condition.
Analysis showed that 4.6% (n = 745 948) of the sample met 12-month criteria for major depressive episodes, of whom 60.3% were women; 3.4% of all men and 6.0% of all women met 12-month major depressive episode criteria. The associations between categories of work-stress variables and the risk of 12-month major depressive episodes expressed as adjusted odds ratios (ORs) are shown in Table 2 ▶.
TABLE 2—
Results of Logistic Regression Analyses Relating Work Stress to Major Depressive Episodes During the Past 12 Months Among Men and Women: Canadian Community Health Survey 1.2, 2002
| Men With Depressiona | Women With Depressiona | |||
| No. (%)b | OR (95% CI)c | No. (%)b | OR (95% CI)c | |
| Overall | 295 887 (3.4) | 450 061 (6.0) | ||
| Job straind | ||||
| Low (Ref) | 8985 (1.3) | 1.00 | 15 264 (4.4) | 1.00 |
| Middle low | 68 156 (2.2) | 1.33 (1.09, 1.64) | 93 529 (4.6) | 1.1 (0.95, 1.28) |
| Middle | 125 548 (3.7) | 1.78 (1.19, 2.67) | 194 403 (6.1) | 1.21 (0.90, 1.63) |
| Highe | 91 447 (5.6) | 2.38 (1.29, 4.37) | 145 705 (7.7) | 1.33 (0.85, 2.09) |
| Social support | ||||
| Low | 79 500 (7.5) | 2.70 (1.55, 4.71) | 102 421 (10.1) | 2.37 (1.71, 3.29) |
| Middle low | 86 364 (3.8) | 1.94 (1.34, 2.81) | 135 657 (7.2) | 1.78 (1.43, 2.21) |
| Middle | 88 327 (2.4) | 1.39 (1.16, 1.68) | 152 271 (5.1) | 1.33 (1.2, 1.49) |
| Highe (Ref) | 39 482 (2.3) | 1.00 | 56 708 (3.7) | 1.00 |
| Job security | ||||
| Low | 95 641 (7.6) | 2.66 (1.56, 2.66) | 85 644 (7.6) | 1.02 (0.70, 1.49) |
| Middle low | 33 965 (4.0) | 1.92 (1.34, 1.92) | 57 285 (8.0) | 1.02 (0.79, 1.30) |
| Middle | 103684 (2.7) | 1.38 (1.16, 1.38) | 175 467 (5.4) | 1.01 (0.89, 1.14) |
| Highe (Ref) | 62 128 (2.2) | 1.00 | 129 252 (5.5) | 1.00 |
| Decision authorityf | ||||
| Low | 49 775 (7.3) | 1.72 (0.97, 1.72) | 75 179 (9.2) | 1.59 (1.06, 2.39) |
| Middle low | 66 542 (4.2) | 1.44 (0.98, 1.44) | 118 700 (7.0) | 1.36 (1.04, 1.79) |
| Middle | 108 240 (3.0) | 1.20 (0.99, 1.20) | 162 333 (5.2) | 1.17 (1.02, 1.34) |
| Highe (Ref) | 70 861 (2.4) | 1.00 | 93 454 (5.0) | 1.00 |
| Skill discretiong | ||||
| Low | 49 967 (5.4) | 1.31 (0.73, 1.31) | 49 000 (5.3) | 0.56 (0.37, 0.56) |
| Middle low | 48 157 (3.3) | 1.19 (0.81, 1.19) | 94 123 (6.3) | 0.68 (0.51, 0.68) |
| Middle | 159 678 (3.3) | 1.09 (0.90, 1.09) | 233 618 (5.8) | 0.82 (0.72, 0.82) |
| Highe (Ref) | 37 616 (2.4) | 1.00 | 72 925 (6.8) | 1.00 |
| Psychological demands | ||||
| Low (Ref) | 5868 (1.6) | 1.00 | 8486 (2.9) | 1.00 |
| Middle low | 47 582 (2.4) | 1.17 (1.02, 1.17) | 80 257 (5.2) | 1.12 (0.99, 1.26) |
| Middle | 125 024 (3.3) | 1.38 (1.03, 1.38) | 172 290 (5.7) | 1.25 (0.99, 1.59) |
| Middle high | 68 046 (4.8) | 1.62 (1.05, 1.62) | 84 705 (6.4) | 1.40 (0.98, 2.00) |
| High | 47 616 (4.1) | 1.90 (1.06, 1.90) | 103 163 (8.2) | 1.57 (0.98, 2.52) |
| Physical exertion | ||||
| Low (Ref) | 49809 (4.6) | 1.00 | 76 950 (7.1) | 1.00 |
| Middle low | 72411 (2.9) | 0.92 (0.82, 1.04) | 139 569 (5.2) | 0.97 (0.89, 1.07) |
| Middle | 25416 (2.5) | 0.85 (0.67, 1.08) | 57 549 (7.0) | 0.94 (0.78, 1.13) |
| Middle high | 92863 (3.7) | 0.78 (0.67, 1.08) | 99 395 (5.2) | 0.92 (0.69, 1.21) |
| High | 54919 (3.2) | 0.72 (0.45, 1.16) | 76 116 (7.7) | 0.89 (0.62, 1.29) |
Note. OR = odds ratio; CI = confidence interval.
aDefined as respondents who had a major depressive episode in the previous 12 months, as assessed by interviewers using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV).12
bWeighted estimates. (Each respondent in the sample was given a sampling weight proportional to the portion of the Canadian population he or she represented. Therefore, our sample of 24 324 represents a weighted population of 16 353 858 Canadians.)
cORs were adjusted for age, marital status, chronic condition (yes or no), antidepressant medication use in the past 12 months, and the other work-stress variables in the Karasek and Theorell model (see “Methods” section).
dDefined as a combination of high psychological demands and low decision latitude.
eBecause of small sample sizes, middle high and high were collapsed into 1 category.
fDecision latitude comprised decision authority (control over work).
gSkill discretion was defined as having a variety of work and the opportunity to use skills.
Increased levels of job strain were significantly associated with an increased risk of major depressive episodes among men (OR = 2.38; 95% confidence interval [CI] = 1.29, 4.37). A similar, although non-significant, trend was seen among women (OR = 1.33; 95% CI = 0.85, 2.09). For both genders, lack of social support at work was significantly related to depression (men, OR = 2.70; 95% CI = 1.55, 4.71; women, OR = 2.37; 95% CI = 1.71, 3.29). Low levels of job security were associated with increased risk of 12-month major depressive episodes among men (OR = 2.66; 95% CI = 1.56, 2.66) but not among women (OR = 1.02; 95% CI = 0.70, 1.49).
Results were contradictory for the 2 factors that compose decision latitude—decision authority and skill discretion. Neither factor was significantly associated with depression among men; however, women in the lowest decision-authority category were more likely to have major depressive episodes (OR=1.59; 95% CI = 1.06, 2.39) than were women with higher levels. Unexpectedly, low skill discretion was associated with decreased risk of 12-month major depressive episodes among women (OR = 0.56; 95% CI = 0.37, 0.56). Increased psychological demands were associated with 12-month major depressive episodes among men (OR = 1.90; 95% CI = 1.06, 1.90). A similar nonsignificant trend was shown among women (OR = 1.57; 95% CI = 0.98, 2.52).
DISCUSSION
We found a consistent association between work stress and major depressive episodes as assessed by DSM-IV criteria. The relations between specific components of work stress and depression differed between men and women. High job strain, low levels of social support in the workplace, low job security, and increased psychological demands were associated with major depressive episodes among men. Among women, lower levels of social support and lack of decision authority were associated with major depressive episodes. Interestingly, having lower skill discretion was inversely associated with major depressive episodes among women.
Comparisons With Other Studies
Our findings support the theoretical model put forward by Karasek and Theorell,17 which associates psychosocial aspects of the work environment with an increased risk of experiencing depression. Our results agree with those of previous longitudinal and cross-sectional studies that found high job strain, low levels of social support within the workplace, low job security, and increased psychological demands were associated with an increased risk of experiencing depression.11,18,19 However, we noted some gender differences that have not been reported in the literature. Job strain was not associated with major depressive episodes among the women in our study, as it has been in previous studies.5,9–11 The reason for this is unclear. It may be because the inverse association between skill discretion (a component of the job-strain score) and major depressive episodes diminished this relation, in spite of the direct association observed with the other component (psychological demands). These results diminished the effect of job strain on major depressive episodes among women, whereas the results for decision authority and demands showed the same association among women as among men.
The gender differences we report here may indicate that some issues are more pertinent for men than for women. Women are more likely to take part-time jobs to balance work and family commitments.21,22 It may be that the lower pay, routine tasks, and limited advancement opportunities that, in general, characterize many part-time jobs22 mean that job strain is not as significant in the relationship with depression for these women. Further work is needed to elucidate the nature of this relationship.
In our study, 4.6% of respondents met the 12-month criteria for major depressive episodes. This is lower than the 8.6% previously reported in a sample of US workers.5 As expected, a higher proportion of women experienced depression than of men (6.0% of women vs 3.4% of men). Although these rates were lower than those previously reported in workers (11.4% of women and 5.9% of men),5 the male-to-female ratio is similar. The lower rates of depression in our study compared with the study of Marcotte et al.5 may be attributable to the more-stringent DSM-IV criteria used to assess major depressive episodes. Many respondents with milder depression, who could have been identified in other studies, may have been excluded from the diagnosis in our data. In addition, unlike the US study, ours did not include respondents who were actively seeking work. It has been consistently shown that rates of depression are significantly higher among unemployed than among employed persons.23,24 Differences in sample sizes and other population characteristics could also contribute to the different rates of depression reported.
Limitations
Our study had several limitations. Because of the cross-sectional nature of the study, we cannot comment on the direction of the relationship between work factors and depression. Although it is plausible that work stressors are risk factors for depression, there is also evidence for pathways leading from psychological symptoms to work characteristics.25–27 Previous depressive illness or current depressed mood may lead to response bias because of negative perceptions of the workplace regarding demands, support, and control.27–29 This could lead to spurious associations between work stressors and depression. Earlier depressive illness could also tend to steer employees into less favorable work-places with high work stressors because of depression-related problems with work functioning and frequent absences that may affect chances of employment or promotion.30
Another issue is the influence of personality on the association between work stressors and depression. It has been hypothesized that personality traits such as negative affectivity—the tendency to report negatively on environmental factors and health—may steer people into less favorable occupations31 or, more likely, may influence the reporting of work characteristics and be an independent predictor of depressive illness.32–34 Because we had no measures of personality in this study, we were not able to test this possibility. However, concern about this source of bias may be overestimated. Although the association between work stressors and depression may be partly explained by negative affectivity, several longitudinal studies have adjusted for personality variables and still found substantial associations between work stressors and mental health19,35,36; this is less likely to be a source of error in instances when depressive illness is measured by standardized interview rather than by self-report questionnaire. Observer ratings of work, rather than reliance on self-reports of working conditions, also remove this bias, and these studies continue to find associations between work stressors and mental health.37
Our sample did not include those who were institutionalized, native populations, or members of the armed forces. Also, the respondents in our sample had more education and higher income and were more likely to be married and White compared with the general population. Although data on the types of occupations these individuals held are sparse, they suggest a predominantly white-collar population. The limitations these factors impose on our ability to generalize our results to other populations are those generally associated with the use of population surveys.
Notably, our study sample represents those individuals who remained in the workforce. It is well established that individuals with severe mental illness are less likely to sustain employment and that rates of depression are higher among the unemployed.38 Therefore, this sample may represent people with personal, social, and economic support systems that enabled continued employment.38 It may also indicate that these individuals had better coping mechanisms, access to social support or treatment, or to more effective treatment than did those who did not remain the workforce.38
Although the interviewers underwent extensive training, they were not clinicians. Therefore, some cases of major depressive episodes caused by a physical condition or drug use may have been included. Nevertheless, the rates we reported were lower than those reported in a previous population survey of employed workers, and it is therefore likely that only a small number of cases were improperly included and thus did not affect our rates significantly.
Conclusions
Depression in the workplace is a major public health problem that requires intervention yet remains underrecognized and under-treated. Effective treatment with medication may reduce the likelihood of experiencing major depressive episodes or enable those who are at risk to cope better with work stress. In our sample, two thirds of those who took antidepressants in the preceding 12 months did not experience a depressive episode. Primary prevention may be a good strategy to improve mental health in the workplace; workplace mental health promotion has been seen as one way of facilitating economic growth and ensuring the sustainability of overburdened social welfare systems. Many of the same factors that influence the risk of major depressive episodes also influence the risk of taking sickness absence, namely, high demands, low support, low skill discretion, and low decision authority. More research is needed to study the types and levels of work stress related to DSM-IV–defined depression in the workforce and their relation to absenteeism. Prevention and treatment strategies could then be tailored to specific factors for individual employees. Both primary preventive approaches and high-quality treatments by primary care, occupational health, and mental health professionals can be used to reduce the burden of depression in the workplace.
Our study is one of the few to show a relation between work stress and major depressive episodes assessed with DSM-IV criteria in a large survey representative of the general working population. The finding that men and women differ in the components of work stress related to major depressive episodes is notable and underlines the importance of considering gender differences in designing future research studies as well as prevention and treatment programs to address this major public health issue.
Acknowledgments
This study was funded in part by the Canadian Institutes of Health Research (grant 03–0415).
We thank the staff of Statistics Canada at the Toronto Research Data Centre, particularly Veronica Yei, Angela Prencipe, and Byron Lee for their assistance with this project.
Human Participant Protection This study was approved by the institutional review board of Statistics Canada.
Peer Reviewed
Contributors E. Robertson Blackmore, S. A. Stansfeld, and I. Weller originated the study and supervised all aspects of its implementation. D. E. Stewart was instrumental in obtaining funding. I. Weller and B. M. Zagorski completed the analyses. S. Munce assisted with the analyses.
References
- 1.World Health Organization. Information on mental health disorders management: depression. Available at: http://www.who.int/mental_health/management/depression/definition/en. Accessed October 9, 2006.
- 2.Alonso J, Angermeyer MC, Bernert S, et al. Prevalence of mental disorders in Europe: results from the European Study of the Epidemiology of Mental Disorders (ESEMeD) project. Acta Psychiatr Scand. 2004; 420:21–27. [DOI] [PubMed] [Google Scholar]
- 3.Kessler RC, Berglund P, Demler O, Jin R, Merikangas KR, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:593–602. [DOI] [PubMed] [Google Scholar]
- 4.Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:617–627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Marcotte DE, Wilcox-Gok V, Redmon DP. Prevalence and patterns of major depressive disorder in the United States labor force. J Ment Health Policy Econ. 1999;2:123–131. [DOI] [PubMed] [Google Scholar]
- 6.Stansfeld S. Work, personality and mental health. Br J Psychiatry. 2002;181:96–98. [DOI] [PubMed] [Google Scholar]
- 7.Kessler RC, Frank RG. The impact of psychiatric disorders on work loss days. Psychol Med. 1997;27:861–873. [DOI] [PubMed] [Google Scholar]
- 8.Wang J. Perceived work stress and major depressive episodes in a population of employed Canadians over 18 years old. J Nerv Ment Dis. 2004;192:160–3. [DOI] [PubMed] [Google Scholar]
- 9.Stansfeld SA, Fuhrer R, Head J, Ferrie J. Work and psychiatric disorder in the Whitehall II Study. J Psychosom Res. 1997;43:73–81. [DOI] [PubMed] [Google Scholar]
- 10.Weinberg A, Creed F. Stress and psychiatric disorder in healthcare professionals and hospital staff. Lancet. 2000;355:533–537. [DOI] [PubMed] [Google Scholar]
- 11.Mausner-Dorsch H, Eaton WW. Psychosocial work environment and depression: epidemiologic assessment of the demand–control model. Am J Public Health. 2000;90:1765–1770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Washington, DC: American Psychiatric Association; 1994.
- 13.Gravel R, Béland Y. The Canadian Community Health Survey: Mental Health and Well-Being. Can J Psychiatr. 2005;50:573–579. [DOI] [PubMed] [Google Scholar]
- 14.Statistics Canada. Canadian Community Health Survey (CCHS): Mental Health and Well-Being cycle 1.2. Available at: http://www.statcan.ca. Accessed August 21, 2006.
- 15.Kessler RC, Abelson J, Demler O, et al. Clinical calibration of DSM-IV diagnoses in the World Mental Health (WMH) version of the World Health Organization (WHO) Composite International Diagnostic Interview (WMHCIDI). Int J Methods Psychiatr Res. 2004; 13:122–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Kessler RC, Ustun TB. The World Mental Health (WMH) Survey Initiative Version of the World Health Organization (WHO) Composite International Diagnostic Interview (CIDI). Int J Methods Psychiatr Res. 2004; 13:93–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Karasek RA, Theorell T. Healthy Work: Stress, Productivity and the Reconstruction of Working Life. New York, NY: Basic Books; 1990.
- 18.Niedhammer I, Goldberg M, Leclerc A, Bugel I, David S. Psychosocial factors at work and subsequent depressive symptoms in the Gazel cohort. Scand J Work Environ Health. 1998;24:197–205. [DOI] [PubMed] [Google Scholar]
- 19.Stansfeld SA, Fuhrer R, Shipley MJ, Marmot MG. Work characteristics predict psychiatric disorder: prospective results from the Whitehall II Study. Occup Environ Med. 1999;56:302–307. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hosmer D, Lemeshow S. Applied Logistical Regression. New York, NY: John Wiley and Sons; 1989.
- 21.Lennon MC. Women, work and wellbeing: the importance of work conditions. J Health Soc Behav. 994; 35:235–247. [PubMed] [Google Scholar]
- 22.Mishel L, Bernstein J, Allegretto S. The state of working America 2006/2007. Washington, DC: Economic Policy Institute; 2007. Available at: http://www.stateofworkingamerica.org. Accessed January 15, 2007.
- 23.Dooley D, Catalano R, Wilson G. Depression and unemployment: panel findings from the Epidemiologic Catchment Area study. Am J Community Psychol. 1986; 22:745–765. [DOI] [PubMed] [Google Scholar]
- 24.Linn MW, Sandifer R, Stein S. Effects of unemployment on mental and physical health. Am J Public Health. 1985;75:502–506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tennant C. Work-related stress and depressive disorders. J Psychosomatic Res. 2001;51:697–704. [DOI] [PubMed] [Google Scholar]
- 26.Taris T, Kompier M. Challenges of longitudinal designs in occupational health psychology. Scand J Work Environ Health. 2003;29:1–4. [DOI] [PubMed] [Google Scholar]
- 27.de Lange AH, Taris TW, Kompier MAJ, Houtman ILD, Bongers PM. The relationships between work characteristics and mental health: examining normal, reversed, and reciprocal relationships in a 4-wave study. Work Stress. 2004;18:149–166. [Google Scholar]
- 28.de Lange AH, Taris TW, Kompier MAJ, Houtman ILD, Bongers PM. Different mechanisms to explain the reversed effects of mental health on work characteristics. Scand J Work Environ Health. 2005;31:3–14. [DOI] [PubMed] [Google Scholar]
- 29.Garst H, Frese M, Molenaar PCM. The temporal factor of change in stressor-strain relationships. J Appl Psychol. 2000;8:417–438. [DOI] [PubMed] [Google Scholar]
- 30.Stansfeld SA, Head J, Marmot MG. Explaining social class differences in depression and well-being. Soc Psychiatry Psychiatr Epidemiol. 1998;33:1–9. [DOI] [PubMed] [Google Scholar]
- 31.Kohn ML, Schooler C. Job conditions and personality: a longitudinal assessment of their reciprocal effects. Am J Sociol. 1982;87:1257–1286. [Google Scholar]
- 32.Parkes KR. Coping, negative affectivity, and the work environment: additive and interactive predictors of mental health. J Appl Psychol. 1990;75:399–409. [DOI] [PubMed] [Google Scholar]
- 33.Noor NM. Work and family roles in relation to women’s well-being: the role of negative affectivity. Pers Individ Dif. 1997;23:487–499. [Google Scholar]
- 34.Naswall K, Sverke M, Hellgren J. The moderating role of personality characteristics on the relationship between job insecurity and strain. Work Stress. 2005; 19:37–49. [Google Scholar]
- 35.Spector PE, Chen PY, O’Connell BJ. A longitudinal study of relations between job stressors and job strains while controlling for prior negative affectivity and strains. J Appl Psychol. 2000;85:211–218. [DOI] [PubMed] [Google Scholar]
- 36.Paterniti S, Niedhammer I, Lang T, Consoli SM. Psychosocial factors at work, personality traits and depressive symptoms. Longitudinal results from the GAZEL Study. Br J Psychiatry. 2002;181:111–117. [PubMed] [Google Scholar]
- 37.Leitner K, Resch MG. Do the effects of job stressors on health persist over time? A longitudinal study with observational stressor measures. J Occup Health Psychol. 2005;10:18–30. [DOI] [PubMed] [Google Scholar]
- 38.Mintz J, Mintz LI, Arruda MJ, Hwang SS. Treatments of depression and the functional capacity to work. Arch Gen Psychiatry. 1992;49:761–768. [DOI] [PubMed] [Google Scholar]