Skip to content

Abstract Search

Health Services/Policy

The cross-sectional association between state-level public health funding per capita and physical health among adults in the United States. Stephen Hunter* Stephen Hunter Sze Y Liu Daniel M Cook Kia Davis Brendan T Smith Roman Pabayo

Background: Funding is necessary for public health agencies to achieve their goals of providing essential public health services. To date, there is limited research on whether public health funding affects individuals’ overall physical health.

Methods: Data from the 2018 Behavioral Risk Factor Surveillance System (BRFSS) was used. Participants reported the number of days they felt were physically unhealthy in the last 30 days. This was categorized into <14, ≥ 14 days of poor physical health. Individual-level confounders were self-reported in the BRFSS, while area-level confounders were from the 2018 American Community Survey. Since respondents were clustered within states, multilevel logistic regression was used to investigate the association between public health funding and poor physical health while adjusting for individual and area-level confounders. We also tested whether household income moderated any observed associations. Sampling weights were also applied to obtain estimates generalizable to the US population.

Results: State-level public health funding per capita did not have a main association with physical health (OR = 0.95, 95% CI: 0.89, 1.00). However, heterogeneity across household income was observed, where greater public health funding per capita was associated with lower odds of poor physical health (OR = 0.92, 95% CI: 0.86, 0.98) among respondents with low household incomes ($75,000 USD) household incomes.

Conclusion: State-level public health funding per capita was not associated with physical health at a population level. However, it may be protective against poor physical health in individuals with lower household incomes. Therefore, public health funding might help reduce physical health inequities. Future research is needed to determine whether this association is causal.