Examining the Relationship between Local Governmental Expenditures on the Social Determinants of Health and County-Level Overdose Deaths, 2017-2020

Context Prior studies have examined the social determinants of health (SDOH) characteristics of counties with elevated rates of overdose deaths. However, this research often ignores variation in upstream governmental investment in these SDOH categories. Objective To examine the association of local governmental investments in SDOH with drug overdose mortality at the county level, controlling for demographic and socioeconomic characteristics. Design Longitudinal analysis from 2017 to 2020. Setting United States counties. Participants 3126 counties. Main Outcome and Measures The primary study outcome was county overdose death rates per 100 000 population taken from the Centers for Disease Control and Prevention. The main predictor of interest was the per capita average of local governmental expenditures in public health, social service, and education drawn from the US Census of Governments, categorized into quartiles. Covariates included county-level socioeconomic and demographic characteristics as well as county-level SUD services from the AHRQ SDOH Database. Results In longitudinal multivariate linear regression models with state and year fixed effects, counties in the highest 3 quartiles of SDOH spending had significantly lower rates of overdose deaths compared with counties in the lowest quartile of SDOH spending; in particular, counties in quartile 2 had an average of 9.09 fewer overdose deaths per 100 000 population (95% CI: -12.58, -5.60), counties in quartile 3 had an average of 13.18 fewer deaths per 100 000 population (95% CI: -17.56; -8.80), and counties in quartile 4 had an average of 13.92 deaths per 100 000 population (95% CI: -19.16, -8.68). Conclusions Our findings demonstrate that higher levels of local governmental investments in SDOH were associated with significantly lower rates of overdose death in US counties after controlling for socioeconomic and demographic characteristics. Going forward, research should focus on determining causality in this relationship, the mechanisms through which this relationship occurs, as well as assess the impact of local public health expenditures allocated toward SUD prevention and treatment specifically on substance use outcomes at the local level.