Department of Biostatistics Seminar/Workshop Series

Biostatistics Student Research Forum:

Modeling Progressive Disease Using Longitudinal Panel Data

Jacquelyn Neal, Ph.D. Candidate, Department of Biostatistics

Vanderbilt University, School of Medicine

Alzheimer's disease (AD) is a slow, progressive disorder, with no fixed events that define its onset. Identifying older adults at highest risk of AD progression could benefit these patients through early interventions to prevent or delay the onset of AD. Panel data, or repeated measurements at pre-scheduled times, can be used to investigate transitions between disease stages. Ultimately, we would like to predict the risk of disease progression in an individual based on their current clinical measurements and any previous data we have on the subject. We have applied existing methodology from the literature to data from the National Alzheimer's Coordinating Center (NACC). Logistic regressions on a cross-sectional subset of the data or transition models with baseline covariate values are the most common methods used when examining transitions between disease stages in AD. However, the assumptions of these models do not hold when considering the pathology of the disease and its progressive nature. We propose using partly conditional models to investigate transitions between disease stages, as these models are highly flexible. These models do not rely on the Markov assumption and time-dependent covariate information can be incorporated. We present work in progress in this area as well.