Examining the Connection Between Health Outcomes, State Political Ideology, and Food Access in the United States

To estimate the impact of county-level income, access to food, availability of health resources, socioeconomic factors, and state political ideology on population obesity and mental health in US counties.

We compiled a county-level data set from the US Census, County Health Rankings, USDA Food Environment Atlas, the American Community Survey, and the State Ideology Database. We specify 2 multivariable regression models for county-level obesity rate and per capita poor mental health days and control for rurality, food access, income, availability of health care resources, state political ideology, and socioeconomic characteristics.

We find that higher food access reduces obesity in counties; an increase in per capita full-service restaurants by 1 unit is associated with reduction in obesity rate by 1.24 points and an increase in per capita grocery stores reduces poor mental health days by 0.14. We also find that counties in liberal-leaning states tend to have lower obesity rates. Access to primary care providers (increase in primary care physicians by 1 is associated with decline in obesity rate by 1.18 points and poor mental health days by 0.11 days), and recreational facilities (increase in recreational facilities per 1000 by 1 is associated with reduction in obesity rate by 3.16 points and poor mental health days by 0.47 days) reduces obesity rates and poor mental health days. Median income is associated with decrease in obesity rate and poor mental health days. Increase in median household income by 1% is associated with reduction in obesity rate 4.75% and reduction in poor mental health days by 1.39 days.

We find that access to food and health care at county level and state ideology through policy making affects health outcomes. Our analysis indicates that counties can improve access to food and health care by investing in these services thereby improving county-level health outcomes and save dollars in the process.