Counties where Trump outperformed Romney have more public health problems, say researchers

What do counties where the Republican Party did better in 2016 than in 2012 have in common? Poor health, according to a recent study.

Counties where the Republican presidential candidate did better in 2016 than in 2012 have substantially poorer public health than those where Democrats gained support, a new study finds. Conducted by researchers at Massachusetts General Hospital and MIT, the study compares election results and public health factors like food security, healthcare costs, and rates of diabetes and obesity. After controlling for other factors, they noticed a pattern: less healthy counties saw an increase in support for the Republican party, suggesting health may have played a role in the 2016 election results. We spoke with Dr. Jason Wasfy about the study.

ResearchGate: Why did you conduct this study, and what did you examine?

Jason Wasfy: We are broadly interested in the relationship of health to people’s lives and people’s decisions. In this study, we examined correlations between public health in counties and the shifting of votes in presidential elections between 2012 in 2016. Since this voting shift is known to be associated with race, education, and income, we adjusted for these factors to estimate an independent association of public health.

RG: What did you find?

Wasfy: We found that most counties shifted toward more Republican in the 2016 election, relative to 2012.  But some counties shifted in the other direction. The counties that shifted toward more Republican generally had substantially poorer public health. This election was a bit of a realignment. For example, Texas voted relatively more Democratic than any year in 20 years. But Wisconsin voted Republican for the first time in 32 years. Across the country, at a county level, even after accounting for many demographic factors, the counties moving toward Republican in 2016 were worse-off in terms of public health. This effect was even stronger in states for which Electoral College votes changed, suggesting at least a possible causal role of public health in the election.

 

“Even accounting for demographic factors, the counties moving toward Republican in 2016 were worse-off in terms of public health.”


 

RG: Do you have any ideas as to why voters in counties with poor public health were more likely to vote Republican in 2016 compared to 2012?

Wasfy: From this analysis alone, we cannot know why this shift occurred—we only know that it occurred. That being said, distressed communities with poor public health moving strongly towards the Republican in 2016 is a possibility.  Also possible is that sicker individual voters within counties staying home from elections may have led to a greater number of healthy voters voting for the Republican.

RG: What constitutes good or poor public health?

Wasfy: We considered a broad variety of factors to try to represent different aspects of public health.  Counties that shifted away from Republican voting were worse off with respect to only rates of violent crime. Every other measure was worse in counties that shifted toward Republican voting in 2016 – physically unhealthy days and mentally unhealthy days as captured by survey data, food insecurity, teen birth rate, rate of primary care doctors, age-adjusted mortality, violent crime rate, health care costs, and rates of diabetes and obesity.

 

“Although our results cannot prove causality, that is very striking, and suggests a possible role for poor health in explaining a surprising election result.”


 

RG: What were some of the factors you controlled for?

Wasfy: We adjusted for a broad variety of demographic factors that are also related to voting shifts.  For example, we adjusted for gender, age, education level, rural versus urban, English proficiency, race, income, and county population.

RG: How great was the disparity in public health you found?  

Wasfy: In counties that shifted relatively more Republican in 2016, the age-adjusted mortality is nearly a third-higher than in counties that shifted less Republican in 2016. Although our results cannot prove causality, that is very striking, and suggests a possible role for poor health in explaining a surprising election result.

 



Featured image courtesy of the IIP Photo Archive.