OBJECTIVES: The rising prevalence of diabetes remains a concern in the US, and the Mid-Atlantic region. Although the relationship between social determinants of health (SDoH) and disease outcomes has been studied more recently, more specific actionable steps are needed to improve the precision of diabetes management interventions. We aim to quantify the association between SDoH and the prevalence of diabetes in the Mid-Atlantic area.

METHODS: County-wise data were collected from US Census and clinical sources via PolicyMap. Data from all counties of the seven states in the Mid-Atlantic area were included: Maryland, Delaware, Virginia, West Virginia, New York, New Jersey, Pennsylvania, and Washington, DC. Multivariate analyses and t-tests were used to explore the impact of SDoH (0.05 level of significance) on diabetes prevalence. The SDoH examined were: age, race, gender, access to healthcare, nativity (foreign vs US born), social vulnerability index (SVI), and median household income.

RESULTS: Results from multiple linear regression model with backward elimination suggest that county-level rate of white residents, access to health care, and gender ratio (men: women) had a significantly negative impact on the prevalence of diabetes, while the median age of the population had a significantly positive impact. Diabetes prevalence rates were significantly lower in counties where the SVI was very low and low when compared to high. Two sample t-tests showed that there is significant difference of prevalence in diabetes between low (12.53%) and very low (11.45%) counties [95% CI (0.52, 1.62)]; low and moderate (13.12%) counties [95% CI (0.96,2.38)].

CONCLUSIONS: SDoH have significant impact on the prevalence of diabetes and addressing them can implement proper policy-based interventions. Race, access to health care and nativity have significant impact on diabetes prevalence. Future studies should expand to patient level data to better understand the drivers of disparities in diabetes.