Effects of Economic Cycles and Job Insecurity on Depression in the U.S.: 1998-2007

The relationship between macroeconomic growth and personal health is moderated by a person's experience of the economy at a micro level (Catalano & Bellow, 2005, Dooley & Catalano, 2003; Jensen & Slack 2003). The experience or threat of job loss, underemployment, job insecurity or reduced income would provoke stress, increasing the risk for clinical depression or antidepressant treatment, with implications for physical comorbidities such as hypertension, hearth disease, or asthma. One might expect that personal employment interacts with the immediate local economic environment, such that job loss in a recessionary environment could be more or less stress-provoking than job loss in a growing economy (hypothesis 1). Unemployment insurance may buffer the effect in the very short term (12-16 weeks), although many “at will” workers do not qualify for benefits. Other economic risk factors for depression could be maintaining personal employment in a recessionary economy, potentially increasing job insecurity; reduced income from reduced hours; or movement to a lesser job (underemployment) (hypothesis 2). Finally, there is macroeconomic evidence that recessionary downturns, especially the most recent starting in 2007, disproportionately affect unskilled or low-wage workers and those in low income households. Therefore, we expect that the effects of job loss on workers in low status industries or in low income households have a moderating effect, increasing the risk for depression (hypothesis 3). We test these hypotheses using the Medical Expenditure Panel Survey (MEPS), a nationally representative survey conducted by the Agency for Healthcare Research and Quality since the mid-1990s. Data span 1998-2007, the latest available data with individual health ratings, health care services and expenditures, and detailed personal economic information including three annual reports of job status, industry of work, household income, and personal demographics unevenly affected by recessions (gender, age, race/ethnicity, region). We use multivariate seemingly unrelated regression analysis to jointly analyze all years of data. Data include expansionary periods (1999, 2004) and downturns (2001-02, 2007, the beginning of the recent Great Recession); 2008 data is released in December 2010. We test these hypotheses and examine age, gender, household income, job industry prestige, job type, marital status, and region as moderators of the relationship between job loss/underemployment and depression. In 2007 10% of U.S. adults ages 18-85 had been diagnosed with depression by a doctor. The association between job insecurity, underemployment, depression, and macroeconomic growth is important for prevention and targeted interventions in populations most affected by the Great Recession.