A Functional Decline Model for Prevalent Cohort Data

Longitudinal designs are often used for studying the natural history of diseases. Data sets typically consist of short series of repeated measures on prevalent cases. We propose a growth model approach to the analysis of follow-up data to describe functional decline and associated risk factors in disease progression. We illustrate the model with an application to longitudinal data that describe the time-evolution of cognitive decline in a cohort of patients with Alzheimer's disease.