Predicting Incident Cardiovascular Disease among African-American Adults: A Deep Learning Approach To Evaluate Social Determinants of Health in the Jackson Heart Study
The Deep Neural Network model provided more accurate predictions regarding incident CVD than the random survival forest or cox proportional hazards models. Contrary to hypothesis, overall predictive accuracy did not improve when adding social determinants of health. However, social determinants of health comprised eight of the top 15 predictors of first CVD events. Findings suggest that understanding upstream social determinants of CVD risk could guide prevention efforts by identifying where and how policy and community-level interventions could be targeted to facilitate changes in individual health behaviors.