Search

Dec 04 Public Health Grand Rounds: “Moving from Collaboration to Team Science: One Statistician’s Perspective”

December 4, 2018 by zrb8mf@virginia.edu

[UHW/MSB Room 3182 (inside Room 3181] Public Health Grand Rounds is a series of lunch-and-learn seminars about current research projects in the area of Public Health. The format will be a 30-40 minute presentation followed by discussion. Lunch will be served to the first 25-30 attendees.

GoToMeeting link for remote viewing: https://www.gotomeet.me/UVA_PHS_Conference

“Moving from Collaboration to Team Science: One Statistician’s Perspective”
Mary D. Sammel, ScD.
Professor of Biostatistics
Department of Biostatistics, Epidemiology and Informatics
Center for Clinical Epidemiology and Biostatistics
University of Pennsylvania Perelman School of Medicine

Abstract
My experience work on the PENN Ovarian Aging Study (POAS) has contributed to my career in many ways. In this talk I describe how my work on this project has shaped my views on team science and collaboration. I will illustrate some examples of how this research team has influenced my interests and methodological contributions.
POAS tracked a population-based cohort of pre-menopausal women age 35-47 years over 14 years to determine associations between reproductive hormones and menopausal symptoms. Well timed annual visits over the study period afforded a unique opportunity to examine not only longitudinal trends in hormones over the menopausal transition but allowed us to examine the question of whether fluctuations in hormones contribute to symptom onset. I will describe the development of a joint modeling approach that utilizes individual-level longitudinal measurements of follicle stimulating hormone (FSH) to predict the risk of severe hot flashes. In this work we developed two approaches, a shared random effects and a latent class framework for the primary-outcome model. Simulation studies are developed to compare and contrast model performance.
Key words: Joint model; Latent class; Long-term trend; Short-term variability