Online ISSN: 1099-176X Print
Copyright © 2001 ICMPE.
Income and Mental Health: Unraveling Community and Individual Level Relationships
|Carole Roan Gresenz,1* Roland Sturm2 and Lingqi Tang3|
D., Economist, RAND, Arlington, Virginia, USA
*Correspondence to: Carole Roan Gresenz, 1200 South Hayes Street,
Arlington, Virginia 22202-5050, USA
Tel.: +1 703-413-1100 x5419
Fax: +1 703-414-4726
Source of Funding: Financial support was provided from the Robert Wood
Johnson foundation, sponsor of Healthcare for Communities, and NIMH grant R01MH62124.
Background: The association between individual socioeconomic status (SES) and mental disorder is well-documented, but studies to date have provided limited and sometimes conflicting evidence on the relationship between aspects of socioeconomic environment, including the role of income inequality, and mental disorder.
Aims of the Study: This paper explores the relationships between mental disorder and individual SES and socioeconomic environment, with particular attention to both the level and dispersion of community income and to their interactions with individual income.
Methods: Cross sectional study using nationally representative, individual level data from the Healthcare for Communities survey merged with supplemental information. Dependent variable is individual mental health status, measured by the 5 item Mental Health Inventory (MHI-5; average 80.6) and an indicator of probable anxiety or mood disorder based on clinical screening instruments (positive for 14.3 percent of respondents in the sample).
Results: MHI-5 decreases (indicating worse mental health), and the probability of an anxiety or depressive disorder increases continuously from the highest to the lowest quintiles of family income. Compared to those in the highest income quintile, MHI-5 is more than 10 points lower and the probability of disorder is much greater among individuals in the lowest income quintile. Within-quintile own income level is also strongly associated with mental health among lower income individuals. We find no evidence that higher levels of income inequality are associated with poor mental health outcomes, measured either by the probability of disorder or MHI-5. Regarding income level, MHI-5 is 3.4 to 3.5 points higher among low income individuals in medium or high income states compared to those in low income states.
Discussion: The qualitative conclusions are stable across various specifications reported (two different measures of mental health, two geographic levels, and among all individuals and low income individuals alone), and in specifications with alternative parameterizations of the community variables (continuously measured, included as quintiles instead of tertiles, and using other indicators of inequality). Individual income is highly correlated with mental health status; level of state income has some association; community or state income inequality has no detectable relationship with mental health. This analysis provides no support for the hypothesis that income inequality is a stronger determinant of health than individual or family income, a hypothesis that in recent years has received much attention in the popular press and policy debates. Limitations of our analysis include the cross-sectional nature of the analysis, the sample sizes used in derivation of the site variables (though sensitivity analyses showed robust results) and the age of the state data.
Conclusions: The association between individual income and mental health is strong. No support for the income inequality hypothesis is found.
Implications for Health Policy Formulation: Our findings point to a need for better understanding of the relationship between individual income and mental health outcomes. Our research does not support the notion that policies aimed at diminishing income inequality are an important lever in improving mental health outcomes for individuals.
Implications for Further Research: This research does not address whether and how different sources of income-at the individual or community level-may affect mental health, and whether the associations observed cross-sectionally also bear out longitudinally. In addition, more research into the relationship between other community characteristics, such as service availability, and mental health outcomes is needed.
Received 26 November 2001; accepted 14 May 2002
Copyright © 2001 ICMPE