A study reported in PLOS ONE examined whether including social determinants of health improves accuracy in predicting who will develop cardiovascular disease (CVD). Matthew Morris, PhD, Hamidreza Moradi, PhD, and study partners analyzed data from 3,980 Black participants in the Jackson Heart Study with no CVD history at baseline. They modeled 10-year CVD risk using machine learning algorithms and evaluated the predictive value of biological, psychosocial, socioeconomic and environmental factors.
Contrary to expectations, adding social determinants of health did not improve overall predictive accuracy beyond traditional CVD risk factors such as gender, nutrition, blood pressure, smoking and cholesterol. However, analyses of relative predictive importance found that eight social determinants were among the top 15 markers of CVD risk, including insurance status, experiences with discrimination and environmental factors reflecting access to physical activity resources and healthy foods.
While traditional factors may better indicate which individuals currently face higher CVD risk, the authors note that social determinants could help guide community prevention efforts by pinpointing where societal barriers exist to individuals adopting healthy lifestyle behaviors. The findings suggest an approach focused on both medical risks and social inequities is needed to alleviate heart disease disparities.
Morris is associate professor of Anesthesiology at Vanderbilt University Medical Center. Moradi, formerly at the University of Mississippi Medical Center in Jackson, is at North Carolina A&T State University in Greensboro. Others from Vanderbilt on the study include David Schlundt, PhD, and Kerry Kinney, PhD. The Jackson Heart Study is supported by the National Institutes of Health (R01HL117285).