Status: Active

Each year, more than 1 million U.S. women experience adverse pregnancy outcomes (APOs), including preeclampsia, gestational hypertension, gestational diabetes and preterm delivery. These APOs are associated with a two-fold higher risk of future maternal cardiovascular disease. However, clinicians do not yet know how to identify which women with a history of an APO are most at risk for cardiovascular disease and no tools exist to help direct care for women after they experience an APO.

 In collaboration with researchers from Brigham and Women’s Hospital and the University of North Carolina-Chapel Hill, Duke AI Health faculty are applying machine learning methods to better identify which women who experience APOs are at risk of developing cardiovascular disease risk factors, such as chronic hypertension, hypercholesterolemia, and Type 2 diabetes in the short term, as we all which women are at risk of longer term cardiovascular events like heart attack and stroke. To build risk prediction models for the study, the team is utilizing data from the Nurses’ Health Study on risk factors for major chronic diseases in women, as well as electronic health record data from women who have given birth within the Duke and UNC health systems. Once the risk prediction modeling is complete, the team will develop and pilot test a clinical decision support tool designed to help inform clinicians and patients of the risks of cardiovascular disease after an APO and recommend tailored, best practice prevention strategies.

This research is supported by NIH/NHLBI 1R01HL167960-01A1.