State level determinants of Poor Health in the United States: Findings from the Behavioral Risk Factor Surveillance System Study

Background: Although there have been studies on predictors of poor health, few analyses have investigated factors associated with poor health at the state level in the United States (US) within a single cohort large enough to accommodate this analysis.

Methods: Self-reported data from respondents aged 18-years and older from the 2010 wave of the Behavioral Risk Factor Surveillance System (BRFSS) study was used for this analysis. Data for 51 US states including the District of Columbia were summarized by computing rates at the state level. Linear regression analysis was used for modeling, with the dependent variable represented by the rate of self-rated poor health, and the following rates as independent variables: self-rated life satisfaction, education-level, income-level, employment status, physical activity in the past 30 days, black race, obesity status, number of children in household, type of health-care coverage, home ownership status, and history of smoking, heavy drinking, anxiety diagnosis, and depression diagnosis. Backward elimination procedure was applied to select the best model that explains poor health.

Results: Predictors of self-rated poor health included low income level (<$25000 per annum), unemployment, and no physical activity engagement in the past 30 days (p<0.0001 for all). These predictors explained 92% of the variation in the rate of self-rated poor health among US states.

Conclusion: These results may reflect the need for greater social support by states and their department of health for its citizenry and possibly explains the need for job creation, increase in wages and provision of recreational centers that encourage physical activities.