Change Point Models in Longitudinal Aging:
Methods, Results, and Challenges
Charles B. Hall
Department of Epidemiology and Population Health
Albert Einstein College of Medicine
Friday, October 23, 2009, 12:30–1:30 pm
GEMS classroom, 3rd Floor in
Shriner's Building
Coffee, tea, and cookies will be provided
Abstract
In longitudinal studies of aging, repeated observations over time are collected on such outcomes as memory and other cognitive domains, mobility, ability to perform activities of daily living, bone density, and mood. In some individuals, the natural history of the aging process, as manifested by such measures, undergoes a change, which can signal incipient disease, disability, or mortality. Change point models may be used to model the time at which such a change occurs along with the natural history before and after such a change. Furthermore, the change point itself may depend on risk factors that could potentially be targets of intervention studies to prolong healthy aging. In this talk we will survey much of the history of the use of change point models in aging and dementia research, highlighting the use of different methodologies to answer different substantive questions, and will describe challenges for future research.