Abstract Body (Do not enter title and authors here): Background: Cardiovascular disease (CVD) remains the leading cause of death globally, disproportionately affecting populations experiencing systemic barriers to healthcare. Although traditional CVD risk factors are well-established, the critical role of social determinants of health (SDOH) remains underutilized clinically. Composite measures capturing individual-level social risk may enhance cardiovascular risk prediction and address persistent health disparities.
Hypothesis: We hypothesized that a composite SDOH score derived through principal component analysis (PCA) of self-reported social determinants data would significantly associate with cardiovascular outcomes, providing robust predictive value independent of traditional cardiovascular risk factors.
Methods: This study included 117,549 participants from the All of Us Research Program who completed the SDOH survey. Eleven SDOH features were analyzed via PCA, yielding two principal components explaining 46% of variance in the features. A weighted cumulative SDOH score was created. Logistic regression assessed associations between the score and cardiovascular outcomes, adjusting for age, sex, and traditional risk factors: hypertension, hyperlipidemia, type II diabetes, and smoking status.
Results: All individual SDOH features were significantly associated with coronary artery disease (CAD). The PCA-derived SDOH score demonstrated significant associations with CAD (OR: 1.30; 95% CI: 1.27–1.33), peripheral artery disease (OR: 1.50; 95% CI: 1.40–1.59), and stroke (OR: 1.25; 95% CI: 1.22–1.29). Adjusting for traditional risk factors, the SDOH score retained its significance with CAD (OR: 1.17; 95% CI: 1.14–1.20). Notably, incorporating the SDOH score into the model reduced racial disparities in CAD outcomes.
Conclusion: Our findings establish the value of a novel PCA-derived individual-level SDOH score for cardiovascular outcomes, independent of traditional risk factors. Integrating this social deprivation metric into clinical risk assessment may improve cardiovascular disease prediction and promote targeted interventions to reduce health disparities.