Psychiatric disorders and the labor market: an analysis by disorder profiles

A key societal cost of mental illness is its impact on the labor market. In examining the relationship between psychiatric disorders and the labor market, the literature to date either examines psychiatric disorders in broad classes or focuses on the impact of specific conditions.

The aim is to examine the relationships among meaningful profiles of concurrent past year disorders and labor market outcomes by gender.

Data are from the National Epidemiologic Survey on Alcohol and Related Conditions for 2001/2002 (NESARC), a representative sample of the noninstitutionalized population aged 18 or older residing in the United States. The analysis sample contains 18,429 women and 16,426 men (unweighted). We examined the relationship between profiles of psychiatric disorders and three labor market outcomes: labor force participation; employment, conditional on labor force participation; and working full-time conditional on being employed. Because no attempt was made to control for potential endogeneity between the labor market outcomes and the psychiatric profiles, we are unable to establish the causal direction of the associations estimated.

First, anxiety disorders among women appear to be associated with labor market outcomes (e.g., anxiety profile in employment outcome: OR=0.76, p<.05). Second, for employment among women large effects were seen for mood disorder and mood and anxiety; in contrast for men, these disorder profiles had significant associations with working full-time rather than employment. Third, for women, of the three labor market outcomes, employment status is particularly sensitive to the profiles of disorders. For men, no such pattern was found for any single labor market outcome.

Concurrent psychiatric disorder profiles affect men and women differently in the labor market. The greatest differences are in (i) the relationship between labor market outcomes and profiles exhibiting anxiety disorders, and (ii) which labor market outcomes are influenced. The main methodological limitation is that the approach does not attempt to assert a direction of causation between mental health conditions and the labor market outcomes. Unobserved heterogeneity and endogeneity are both possible and likely to some degree. Other limitations pertain to the data, which are cross-sectional and exclude some relatively rare disorders (e.g., schizophrenia).

The results may have significant implications for access to appropriate treatment, which may vary greatly by psychiatric disorder profile and by gender.

In the United States, private third-party coverage for treatment is obtained either directly through working full-time at a medium to large firm or through a spouse being employed. The findings thus suggest that mental health policy may need to account for labor market policy.

Additional research is needed to disentangle the findings presented. For example, it is important to separate the influence of substance use disorders from mood disorder and uncover the pathways by which the different conditions in substance use disorders may influence the job market.