Abstract 13270: Social Deprivation Index and Cardiovascular Disease Risk Among 2.5 Million Patients in the US

Abstract

Background: Neighborhood-level social determinants of health (SDOH) impact cardiovascular disease (CVD) risk. However, limited data are available from large, multi-ethnic, national samples with CVD risk factors on how place-based SDOH independently relates to CVD risk, and on whether accounting for place-based SDOH attenuates race/ethnicity differences in CVD risk. Methods: We studied 2,538,811 patients free of prior CVD in the Optum Labs Data Warehouse (OLDW). Billing addresses were geocoded to derive social deprivation index (SDI, deciles 0 to 10 as a composite measure of poverty, education, employment, home rental & crowding, single-parent households and no car ownership) at the zip code level. SDI was compared by race/ethnicity, as coded in the electronic health record. Multivariable Cox models quantified the adjusted relative hazard of incident CVD (using hospitalization discharge codes) associated with linearly modeled SDI in the overall population and stratified by race/ethnicity. We also assessed race/ethnicity differences in incident CVD risk before and after adjustment for SDI and CVD risk factors. Results: Mean age was 52 years, 57% were female, with a range of health insurance (75% commercial, 15% Medicare, 2.5% Medicaid). The highest SDI was seen among Black and Hispanic patients. SDI was associated with increased risk of CVD in all race/ethnicity groups (Table). Race/ethnicity differences in CVD risk were attenuated when adjusted for SDI and CVD risk factors (e.g., HR for HF in Black vs White patients decreased from 1.59 [95% CI: 1.47-1.72] adjusted for age and sex, to 1.29 [95% CI: 1.18-1.40] further adjusted for SDI, to 1.09 [95% CI: 1.00-1.19] adjusted for SDI and risk factors). Conclusions: Social deprivation levels are associated with CVD risk independent of traditional risk factors. Accounting for disparities in SDI, which reflect both present and past influences, attenuates differences in CVD risk associated with race and ethnicity.