FEBRUARY 7, 2025
Haiyi Xie, PhD
Professor of Biomedical Data Science and Community and Family Medicine
Statistician, Center for Technology and Behavioral Health
Geisel School of Medicine at Dartmouth
About the Presentation: Longitudinal follow-up studies are a common approach in health services research. However, time trends for longitudinal outcomes are seldom linear. When statistical models incorporate curvilinear (polynomial) terms and interaction terms, interpreting the results becomes complex. This challenge is further amplified when the outcome variable is categorical. The purpose of this talk is to highlight the challenges practitioners often encounter in interpreting nonlinear longitudinal modeling results and to provide recommendations for enhancing interpretability.
About the Presenter: Haiyi Xie, PhD is a Professor of Biomedical Data Science and Community and Family Medicine, Geisel School of Medicine at Dartmouth, and a statistician at the Center for Technology and Behavioral Health. He collaborates with the Centers’ Principal Investigators, and provides statistical consultation, analysis and other support for a variety of research projects. He has extensive experience in longitudinal and multilevel data analyses with generalized linear-mixed models, generalized estimating equations (GEE), and latent growth curve and trajectory class modeling.