Social determinants of health and atrial fibrillation hospitalizations: a comparative analysis of rural and urban counties in north carolina

Background Atrial fibrillation (AF) is a major contributor to hospitalizations and healthcare costs. The relationship between the social determinants of health (SDOH) and AF hospitalizations remains relatively unexplored, particularly between urban and rural populations. Hospitalization rates may reflect differences in healthcare access rather than true disease burden. Purpose This study examines the association between AF hospitalizations and 8 SDOH across all 100 counties in North Carolina. Methods AF hospitalization data for Medicare beneficiaries (65+) were obtained from the CDC Atlas of Heart Disease and Stroke (2019-2021) and the US Census Bureau. Eight county-level SDOH variables (uninsured rates, homeownership, median household income, primary care physician (PCP) density, low-income and low-access (LILA) areas (food deserts), food insecurity, limited access to healthy foods, and the food environment index were sourced from the USDA Food Access Research Atlas and the North Carolina Rankings and Roadmaps. Based on the US Office of Management and Budget, counties were classified as urban (n=46) or rural (n=54). Pearson correlation coefficients were calculated overall and separately for rural and urban counties to determine associations between the 8 SDOH factors and AF hospitalizations. Results In NC counties, we found that AF hospitalizations were negatively correlated with homeownership (r=-0.36, p<0.001) and food insecurity (r=-0.41, p<0.001) and positively correlated with median household income (r=0.50, p<0.001) and PCP density (r=0.86, p<0.001). In the urban NC counties, AF hospitalizations were negatively correlated with uninsured rates (r=-0.34, p<0.001), homeownership (r=-0.47, p<0.001) and food insecurity (r=-0.42, p=0.004) and positively correlated with median household incomes (r=0.45, p<0.001) and PCP density (r=0.86, p<0.001). In contrast, AF hospitalization was positively correlated only with PCP density (r=0.69, p<0.001) in rural NC counties. No significant correlations were found overall or after stratification between AF hospitalizations and LILA areas, percentage of population with limited access to healthy foods, or the food environment index. Conclusion These findings suggest that economic and social factors drive AF hospitalizations in urban NC counties, while primary care availability is the stronger determinant in rural NC counties. The lower AF lower hospitalization rates in older adults living in rural NC counties may indicate underdiagnosis due to a lack of access to primary care rather than a lower disease burden. These findings suggest that reducing economic barriers in urban NC counties and expanding primary care in rural NC counties could improve access to AF treatment and influence AF hospitalizations.