Time trends between vaccination coverage and voting patterns before and during the COVID-19 pandemic: analysis of COVID-19 and flu surveys in the United States

Background We assessed the relationship between vaccination coverage and voting patterns: how has the association between COVID-19 vaccination and voting patterns changed during the pandemic, how does it compare to the association between flu vaccination coverage and voting patterns, and what can the time trends between flu vaccination and voting patterns tell us about the broader relationship between vaccination coverage and voting patterns. Methods We analyzed survey data on flu and COVID-19 vaccination coverage utilizing National Immunization Surveys for flu (NIS-FLU; years 2010-2021) and for COVID (NIS-ACM; 2021-2022), CDC surveillance of COVID-19 vaccination coverage (2021-2022) and US COVID-19 Trends and Impact Survey (CTIS; 2021-2022). We described the association between state-level COVID-19 and flu vaccination coverage and state-level voting patterns using Pearson correlation coefficient. We examined individual-level characteristics of people vaccinated for COVID-19 and for flu using logistic regression among responses in CTIS during April-June 2022. We analyzed flu vaccination coverage by age in NIS-FLU between 2010-2021, and its relationship with voting patterns to see whether there has been a departure from the secular pre-pandemic trend during the pandemic. Results Between May 2021 – June 2022 there was a strong and consistent correlation between state-level COVID-19 vaccination coverage and voting patterns for the Democratic party in the 2020 presidential elections. Pearson correlation coefficient was around 0.8 in NIS-ACM, CTIS and CDC surveillance with a range of 0.76–0.92. COVID-19 vaccination coverage in June 2022 was higher than flu vaccination coverage in all states and it had a stronger correlation with voting patterns (R=0.90 vs. R=0.60 in CTIS). There was a small reduction in the flu vaccination coverage between 2020-2021 and 2021-2022 flu seasons. In the individual-level logistic regression, vaccinated people were more likely to be living in a county where the majority voted for the Democratic candidate in 2020 elections both for COVID-19 (aOR .18, 95%CI 2.12-2.24) and for flu (aOR 1.38, 95%CI 1.36-1.41). We demonstrate a longstanding correlation between voting patterns and flu vaccination coverage. It varied by age with the strongest correlation in the youngest age groups. During the 2020-2021 flu season, all age groups, except for 5-12 years old, had a stronger correlation coefficient with voting patterns than in the previous years. However, the observed and predicted vaccination coverage show relatively modest differences in their correlation with vote share. Conclusions There are existing pre-pandemic patterns between vaccination coverage and voting patterns as demonstrated by the flu vaccination coverage for 2010-2021. During the pandemic COVID-19 vaccination has been more strongly correlated with vote share than the correlation observed for flu during and before the pandemic. The findings align with other research that has identified an association between adverse health outcomes and the political environment in the United States.