Community Public Health Conditions and Political Participation

• In Event: Race, Health Policy, and Disparities in Political Voice

Abstract

Background
Research has demonstrated that individuals in poor health are less likely to participate politically through voting, signing petitions, and engaging with elected officials, compared to those in better health status (Brown et al., 2020; Burden et al., 2017). These studies have examined measures of physical and mental well-being such as self-rated health and insurance status (Brown et al., 2020). Inconsistent research findings indicate that the mechanisms through which health impacts voter turnout and other political behaviors are still unclear. Previous work also demonstrates that health may have different relationships with different political behaviors (Burden et al., 2017). The study of health and political behavior in the U.S. is increasingly important and complex given the major health inequities that persist across race and income (Baciu et al., 2017). Black Americans experience drastically lower wages, lower upward economic mobility, and experience much higher rates of chronic conditions and premature death compared to white Americans (Baciu et al., 2017; Chetty et al., 2020). These disparities in income and health disproportionately experienced among people of color have a profound impact on individuals’ everyday life and ability to engage in political activity; this holds critical implications for whose interests are represented in politics.

Research Motivation
Public health literature focuses on community conditions and structural determinants to explain these disparities in individual health by race and income, but few studies have examined associations between community-level health and political participation. Those few studies that currently exist tend to focus only on voting behavior. For instance, one study found that poor county-level community health was associated with county-level changes in voting patterns at the national-level from 2012 to 2016 and in a follow up study found that the relationship persisted, comparing the 2016 presidential election to the 2018 U.S. House of Representatives election (Wasfy et al., 2017, 2020). However, both of these studies examined voter turnout at the population-level, rather than estimating the effect of county-level conditions on individual-level political participation. Another study found that less reduction in the age-adjusted death rate at the county-level from 2008 to 2016 was associated with an increased percentage of votes for the Republican candidate for president (Goldman et al., 2019). These works suggest that health at the county-level may impact individual-level political participation and that these changes in participation impact election outcomes. However, further research is necessary that uses individual measures of political behavior to establish this relationship, rather than using participation data aggregated to the population-level.

Methods
I will use the 2018 Cooperative Congressional Election Survey (CCES) and the 2018 Robert Wood Johnson Foundation (RWJF) County Health Rankings (CHR), to examine relationships between county-level public health and the likelihood of individuals engaging in four political behaviors: (1) turnout in the 2016 U.S. national election, (2) contacting a public official, (3) working for a candidate or campaign, and (4) attending a political protest, march, or demonstration. I will perform exploratory factor analysis to summarize several items from the RWJF CHR to estimate the latent variable of county-level public health, measured across four dimensions – population health, access to healthcare, environmental factors, and social factors. I will merge this data with the CCES data and compute descriptive statistics. I will estimate associations between each outcome variable and county-level public health using bivariate logistic regression. I will employ a mixed-effects multivariate logistic regression model for each outcome variable with the county-level public health indicator as the key independent variable. I will estimate stratified logistic regression models to assess whether the county public health and political participation relationship varies within each income and racial group. I will use interaction terms to assess whether the relationship between county public health and participation varies between income and racial groups, respectively.

Implications
This study will address gaps in the literature on political participation and health by examining associations between four forms of political participation and a multidimensional measure of community public health. Understanding how and why citizens participate is increasingly important given the current polarized political climate in the U.S. and the global COVID-19 pandemic, which has exacerbated health inequities and imposed restrictions on resources, such as income and education, which are known determinants of political participation. This study will produce evidence which can inform community-level health interventions and efforts to increase political participation, including voter turnout.

Authors

• Cydney Marie McGuire, University of Minnesota School of Public Health

• Sarah E. Gollust, University of Minnesota