The authors of this study developed a single risk score derived from multiple social determinants of health that predicts county-level cardiovascular disease mortality.
Using data culled from government resources, they identified U.S. counties with a high risk of CVD based on a three-year average mortality rate.
They used a 50 percent random sample of 3,026 counties to develop a risk score based on seven social determinants of health factors: proportion of non-white population, poverty rate, proportion of population without high school diploma, grocery store ratio, fast-food restaurant ratio, after-tax soda price and primary care physician supply. The remaining 50 percent of the counties served to validate the measure.
The resulting index had better predictive performance for CVD burden than common single-measure area-level indexes (e.g., only measuring poverty). The authors conclude that their multivariable SDoH risk score can identify counties with high CVD risk and has the potential to improve CVD risk prediction and interventions for vulnerable populations at the county level.