In a study published in the American Journal of Health Behavior, researchers found an nuanced connection between obesity and voting behavior in the 2016 presidential election. Counties with increasing obesity rates initially showed rising support for Donald Trump until a threshold, after which support began to decline. This suggests a complex, non-linear relationship between local obesity prevalence and political voting patterns.
Obesity has become a major public health issue in the United States. The prevalence of obesity has been increasing significantly, and it is associated with various chronic diseases and substantial healthcare costs.
The condition has become a political issue with two predominant standpoints: the “personal-responsibility” perspective and the “environmental” perspective. These perspectives have different policy implications and are associated with different political parties.
While health disparities have been studied as a political outcome, the reverse relationship, i.e., how health behaviors or conditions influence political behavior, has not been extensively explored. Researchers Ruopeng An and Mengmeng Ji aimed to fill this gap in the literature.
To investigate the link between voting behavior during the 2016 presidential election and obesity rates, the researchers retrieved county-level 2016 presidential election data from the New York Times database. This data included the percentage of votes for the Republican Party presidential candidate (Donald Trump), the percentage of votes for the Democratic Party presidential candidate (Hillary Clinton), and the total number of votes in each county.
To ensure the accuracy of their election data, the researchers compared it with data from two other major election datasets maintained by The Guardian and Ballotpedia.org. They found that the state-specific percentage of votes for the Republican and Democratic candidates agreed across all three databases.
Data on county-level prevalence of adult obesity in 2013 were obtained from the U.S. Centers for Disease Control and Prevention (CDC)’s County Data Indicators (CDIs). This data was based on the Behavioral Risk Factor Surveillance System (BRFSS), an annual survey that collects information on health-related behaviors, chronic conditions, and preventive services.
To account for potential influences, they controlled a wide range of county-level sociodemographic factors, such as age groups, racial and ethnic composition, education, income, poverty rates, unemployment, metropolitan status, and total population.
The researchers used a spatial modeling approach called a spatial autoregressive regression (SAR) to examine the relationship between obesity rates and voting patterns. This approach allowed them to account for the spatial clustering of data, which is often seen in political behaviors. They analyzed information from 3,111 U.S. counties, excluding Alaska due to data limitations.
Contrary to previous studies that had suggested a linear relationship between obesity rates and votes for the Republican Party candidate, this research uncovered a non-linear, or more complex, connection.
The study found that as the obesity rate in a county increased from around 12% to approximately 34%, there was a steady rise in votes for the Republican Party presidential candidate. However, this trend plateaued when obesity rates reached about 36.1%.
Beyond this point, as obesity rates continued to rise, the vote margin for the Republican Party candidate started to decline. In simpler terms, there was a turning point – an obesity rate threshold – where further increases in obesity did not lead to more support for the Republican Party.
While this study provides valuable insights, it’s essential to recognize its limitations. The use of county-level data means that the findings don’t offer a direct link between individual obesity and voting behavior. Additionally, the data used for obesity rates was from 2013, which might not perfectly align with the 2016 election year.
The researchers cautioned against drawing causal interpretations from their findings, emphasizing the need for future research to explore the mechanisms underlying the relationship between obesity rates and voting behaviors.
“This study assessed the impact of county-level obesity prevalence on the 2016 presidential election,” the researchers concluded. “A quadratic association between county obesity rate and the vote margin for the Republican Party presidential candidate was identified. The vote margin initially increased with county obesity rate, but after reaching its peak, it started to decline as obesity rate further increased. Findings of this study indicate that disparities in obesity not only may serve as a political outcome, but also as an influence on political behavior. This study is observational and subject to measurement error, confounding, and ecological fallacy.”
“Future research is warranted to replicate the nonlinear association between county obesity rate and the vote margin for the Republican Party presidential candidate identified in this study, elucidate the mechanisms through which the obesity epidemic links to voting behaviors, and track the long-term trend for the relationship between obesity rate and the presidential election.”
The study, “Obesity Prevalence and Voting Behaviors in the 2016 US Presidential Election“, was published September 2018.
