Prevalence and patterns of major depressive disorder in the United States labor force

Background and Aims of the Study: In this paper, we identify the 12-month and lifetime prevalence of major depressive disorder in and out of the labor force, and among the employed and unemployed. We examine whether prevalence by labor force and employment status varies by gender and over the life cycle. Finally, we examine whether people can ‘recover’ from depression with time by identifying patterns of labor force participation and employment as time since most recent episode passes. Methods: We examine data collected as part of the National Comorbidity Survey, a survey representative of the population of the United States designed to identify the prevalence of major mental illnesses. The National Comorbidity Study identified cases of major depression via the Composite International Diagnostic Interview. Using these data, we estimate univariate and bivariate frequency distributions of major depressive disorder. We also estimate a set of multivariate models to identify the effect of a variety of dimensions of major depression on the propensity to participate in the labor force, and be employed if participating. Results: Lifetime and 12-month prevalence rates of depression are similar in and out of the labor force. Within the labor force, however, depression is strongly associated with unemployment. The negative relationship between depressive disorder and employment is particularly strong for middle age workers. Depression and the number of depressive episodes have a differing pattern of effects on labor market outcomes for men and women. We find evidence that labor force participation and employment rates for people with a history of depression increase significantly over time in the absence of additional depressive episodes. Discussion: Labor market status represents an important dimension along which prevalence of major depression varies. The relationship between depression and employment status is particularly strong for middle aged persons, but becomes weaker as time passes since the last depressive episode. Continued exploration of the association between work (or lack of work) and depression may ultimately help in the prediction, treatment and assessment of the illness. Implications for Practice and Policy: These results present a basic set of facts about the relationship between major depressive disorder and labor market outcomes. We have not, however, attempted to sort out the complexities of this relationship here. These complexities arise at almost every turn. For instance, the high level of prevalence of depression among the unemployed may be due to the possibility that the stresses associated with unemployment trigger depressive episodes or to the possibility that workers who are depressed are more likely to be fired or quit. Implications for Further Research: Our continuing research attempts to address these problems. Understanding when and how depression affects labor market outcomes and when and how labor market outcomes affect depression is an important endeavor for those interested in treating the disease and understanding its consequences. © 1999 John Wiley & Sons, Ltd.