As before, we ran one multivariable model without and one with state fixed effects. We used the same approach to examine the association between the absolute change in life expectancy from 1980 to 2014 and the change in the proportion of votes cast for the Republican party from 2016 to 2020. We used multivariable linear regression models, one without and one with state fixed effects, weighted by the number of two-party votes to assess the association between the absolute change in life expectancy from 1980 to 2014 and the proportion of votes for the Republican candidate in the 2020 Presidential election. We also excluded one county due to missing data and counties in Alaska and Kalawao County, Hawaii due to a discrepancy in county classifications between the data sources, so 3110 counties and Washington, D.C.

Across 3109 counties with any votes not attributed to the Democratic or Republican parties, the median percentage of votes for the third-party candidate was 2%, so we excluded votes for third-party and other candidates from the denominator. We obtained county-level voting results from the 2020 Presidential election from GitHub ( McGovern, 2020), which were scraped from results published by Fox News, Politico, and the New York Times, and calculated the Republican candidate's share of votes as the proportion of total votes cast for the Democratic and Republican candidates in each county and Washington, D.C. We obtained county-level data on life expectancy from the Institute for Health Metrics and Evaluation (Institute for Health Metrics and Evaluation, 2017) and calculated the absolute difference in life expectancy from 1980 to 2014. We also assess the association between the change in county-level life expectancy and the change in the proportion of votes cast for the Republican candidate from 2016 to 2020. In this brief report, we examine the association between change in county-level life expectancy from 1980 to 2014 and the proportion of votes cast for the Republican party in the 2020 U.S. Voters in those counties were more likely to vote for Republican candidates in 2016, irrespective of prior voting patterns.įollowing the 2020 Presidential election, we examined the election results to determine whether the pattern observed in 2016 persisted. One analysis found a 9.1 percentage point increase in the Republican vote share from 2008 to 2016 in counties in which life expectancy stagnated or declined from 1980 to 2014 ( Bor, 2017). Recent reports have described the association between changes in life expectancy and voting behavior, noting a shift toward the Republican presidential candidate in counties where life expectancy has declined or increased the least ( Guo, 2016 Bor, 2017 Bilal et al., 2018). A gap of more than 10 years in life expectancy existed between counties in the 1st and 99th percentile, an increase of 2.4 years from 1980 ( Dwyer-Lindgren et el, 2107).

In 2014, a 20-year gap existed between counties with the lowest and highest life expectancy. Large disparities in life expectancy among counties in the U.S. increased by almost 10 years over a nearly 60-year period from 1959 to 2016, the rate of increase slowed over time and life expectancy has declined after 2014 ( Woolf & Schoomaker, 2019).

Nytimes election map 2020