Search results

The past quarter-century has seen research on the economic impacts of mental illness flourish. Innovations in measurement and the release of several community-based and often nationally representative data sets containing valid and reliable diagnostic information have enabled researchers to make substantial advances in understanding the myriad ways that mental illness impacts the economic lives of the ill and their families. Among the most interesting and persistent findings in this literature is that mental illness affects women and men differently. Not only do women and men have very different rates of prevalence for various diseases, but mental illness is also commonly found to have different effects in their economic lives.

Mental illness, in its various forms, is common in the United States. Tens of millions of Americans are afflicted by an episode of mental illness every year. Estimates of the 12-month prevalence of mental disorders in the U.S. (including alcohol and substance abuse or dependence) indicate that 22–30 persons per 100 in the adult population are afflicted each year.¹ An episode of a psychiatric disorder, like a physical disorder, is debilitating – often disrupting the ability of the afflicted to carry on normal personal, social, and work activities. Mental illness also commonly results in large medical expenses. In addition, a number of recent papers have found that mental illness imposes large labor market losses on the ill, decreasing the likelihood of employment and limiting earnings for the employed.² In particular, research by two of the authors indicates that depressive disorders cause significant reductions in the labor force participation of women and the earnings of both men and women.³

In a number of recent studies, it has been demonstrated that mental illness imposes real and large costs over and above the direct expenses of care and treatment. Each year in the U.S., 5–6 million workers between 16 and 54 years of age lose, fail to seek or cannot find employment as a consequence of mental illness. Among those who do work, it is estimated that mental illness decreases annual income by an amount between USD 3500 and USD 6000 (Marcotte & Wilcox-Gök, 2001). Similar results have been shown in a number of studies (Ettner et al., 1997).

Mental disorders collectively account for 4 of the 10 leading causes of disability and represent more than 15% of the overall burden of disease in the United States (SAMHSA, 1999). The first Surgeon General’s Report on Mental Health reported that in 1999 nearly 20 million American adults (9.5% of the population) were clinically depressed and that, at any one time, 1 in every 20 employees is experiencing depression (SAMHSA, 1999). The indirect costs of mental disorders to the American economy amounted to an estimated $79 billion in 1990, with loss of productivity because of illness accounting for about 80% of these costs ($63 billion) (Rice & Miller, 1996). Additionally, significant costs may accrue from decreased productivity due to symptoms that sap energy, affect work habits, and cause problems with concentration, memory, and decision-making (SAMHSA, 1999).

In this research we use data from the National Comorbidity Survey to examine the effects of affective mental disorders on the income and employment of women and men. We consider separately the effects of three major categories of affective illness: depression, bipolar illnesses, and dysthymia.We use a two-stage instrumental variables procedure to estimate the labor market effects of affective disorders in the presence of unmeasured heterogeneity between those with and without illness. In the first stage, we draw on respondents' family histories with illness to construct instrumental variables, which are theoretically and empirically relevant. These instruments are used in second stage regression and logistic analyses of annual income and employment.We find evidence that some affective disorders result in earnings losses for both women and men. For women, depression is associated with substantial earnings losses for working women, on the order of more than $6,000 per year. But, these losses are mitigated over time. Moreover, we find evidence that depression has substantial negative employment effects for women. For men, we find evidence of income losses associated with dysthymia, but we find no evidence of significant employment effects.

Women are nearly twice as likely as men to suffer a major depressive episode (Kessler et al., 1994). Risk of onset for single mothers is twice that of married mothers and financial hardship also doubles the risk of becoming depressed (Brown & Moran, 1997). If diagnosed, depression can be effectively treated, typically with pharmacotherapy or psychotherapy or some combination of the two (Goldman et al., 1999; Sirey et al., 1999). But a sizable majority of sufferers remain undiagnosed and untreated (Lennon et al., 2001). Such treatment can be prohibitively expensive to patients who lack health insurance, particularly those with few financial resources. Although most low-income women have a safety net in Medicaid, welfare reform’s delinking of Medicaid from welfare cash assistance has left uncovered many who are eligible for the benefits (Garrett & Holahan, 2000).

It is well known that mental health disorders cause substantial functional limitations and disability (Surgeon General, 1999). Less well known is the central role that mental health plays in economic disparities. The prevalence of depressive disorders is almost 3 times as high among individuals in the bottom 20% than among individuals in the top 20% of the income distribution, a much steeper gradient than for hypertension, heart disease, arthritis, chronic pain, or the number of medical problems (Sturm & Gresenz, 2002). In addition, individuals with mental disorders are less likely to have savings than individuals with physical health problems and the disparity widens with advancing age (Gresenz & Sturm, 2000).

During the past two decades, there has been a growing awareness of the impact of mental illness on the population (Regier et al., 1988; U.S. Department of Health and Human Services, 1999; World Health Organization, 2001). However, only recently have issues surrounding its effect on the labor force been raised (Berndt et al., 2000; Dewa & Lin, 2000; Kessler et al., 1999; Marcotte et al., 1999; Stewart et al., 2003).