Duncan, Meredith

Research Activities

We are assessing the role of social determinants of health in predicting cardiovascular disease risk. We are using structural equation modeling, but because our datasets are somewhat large (~50,000 participants) and the method is complex, we are not able to run the analyses on our local machines in a timely manner (a single model takes over 24 hours and renders the laptop inoperable during that time). Future analyses will seek to build risk prediction models for cardiovascular disease which incorporate social determinants of health.

 

List of Projects

  • Identification of Social Determinants of Health Associated with Cardiovascular Disease

  • Assessing the Added Value of Social Determinants of Health in Cardiovascular Disease Risk Prediction

 

Computational Methods

  • structural equation modeling (implemented with the lavaan package in R, should be available)

  • Cox proportional hazards regression (can be implemented in either R or SAS)

 

List of Software

R

 

Collaborators

All collaborators are affiliated with UK: myself (Meredith Duncan), Anthony (Tony) Mangino, Kory Heier, and William Burton Burrows

Center for Computational Sciences