Cardiovascular Health & Data Science: Apply Bayesian Tools to Real-World Public Health Data

We are offering a student research opportunity that applies advanced data science techniques to important public health questions. Specifically, we will use Bayesian Profile Regression (BPR)—a method I co-developed—to analyze complex exposure profiles in studies that previously used more traditional or ad hoc approaches. Our primary focus will be on the American Heart Association’s (AHA) seven cardiovascular health metrics and their association with outcomes in the Atherosclerosis Risk in Communities (ARIC) study. Subject to data availability, we will apply BPR to these multivariate health metrics to uncover latent exposure profiles and their relationship to cardiovascular outcomes. For background on the AHA metrics and ARIC data: https://pmc.ncbi.nlm.nih.gov/articles/PMC3685848/ For background on the BPR method: https://academic.oup.com/biostatistics/article-abstract/11/3/484/257407?login=false This is an excellent opportunity for students interested in the intersection of data science, epidemiology, and public health.