Jason H Wasfy1, Emma W Healy2, Jinghan Cui2, Charles Stewart3. 1. Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. Electronic address: jwasfy@mgh.harvard.edu. 2. Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. 3. Massachusetts Institute of Technology, Cambridge, MA, USA.
Abstract
BACKGROUND: In the U.S. presidential election of 2016, communities with poorer public health shifted votes to the Republican party. Whether this trend has persisted beyond 2016 is unclear. METHODS: We created a county-level measure of public health (the "unhealthy" component) by performing principal component analysis on 9 health statistics. We then estimated shifting of votes by defining "net vote shift" as the percentage of Republican votes in the 2018 U.S. House of Representatives election minus the percentage of Republican votes in the same election in 2016. Finally, we performed linear regression to assess the independent, county-level association of the unhealthy component with net vote shift after adjusting for county-level demographic factors. RESULTS: The mean county-level net vote shift was -6.4 percentage points (SD 12.6 percentage points), consistent with a mean net vote shift toward the Democratic party. After adjustment for demographic covariates, the unhealthy score was associated with higher net vote shift (17.7 percentage points shift toward Republican per unit unhealthy, p = .0323). CONCLUSIONS: In the 2018 congressional elections, despite an overall shift toward the Democratic Party there is evidence of ongoing shifting of community voting in unhealthy communities toward the Republican party.
BACKGROUND: In the U.S. presidential election of 2016, communities with poorer public health shifted votes to the Republican party. Whether this trend has persisted beyond 2016 is unclear. METHODS: We created a county-level measure of public health (the "unhealthy" component) by performing principal component analysis on 9 health statistics. We then estimated shifting of votes by defining "net vote shift" as the percentage of Republican votes in the 2018 U.S. House of Representatives election minus the percentage of Republican votes in the same election in 2016. Finally, we performed linear regression to assess the independent, county-level association of the unhealthy component with net vote shift after adjusting for county-level demographic factors. RESULTS: The mean county-level net vote shift was -6.4 percentage points (SD 12.6 percentage points), consistent with a mean net vote shift toward the Democratic party. After adjustment for demographic covariates, the unhealthy score was associated with higher net vote shift (17.7 percentage points shift toward Republican per unit unhealthy, p = .0323). CONCLUSIONS: In the 2018 congressional elections, despite an overall shift toward the Democratic Party there is evidence of ongoing shifting of community voting in unhealthy communities toward the Republican party.