Unhealthy Behaviors, Prevention Measures, and Neighborhood Cardiovascular Health: A Machine Learning Approach

摘要

This study identifies and ranks predictors of cardiovascular health at the neighborhood level in the United States. We merged the 500 Cities Data and the 2011-2015 American Community Survey to create a new data set that includes sociodemographic characteristics, health behaviors, prevention measures, and cardiovascular health outcomes for more than 28 000 census tracts in the United States. We used random forest to rank predictors of coronary heart disease and stroke. For coronary heart disease, the top 5 ordered predictors were the prevalence of taking medicine for high blood pressure control, binge drinking, being aged 65 years or older, lack of leisure-time physical activity, and obesity. For stroke, the top 5 ordered predictors were the prevalence of obesity, lack of leisure-time physical activity, taking medicine for high blood pressure, being black, and binge drinking. Machine learning approaches have the potential to inform policy makers on important resource allocation decisions at the neighborhood level.