General Departmental Seminar Series
Statistical Challenges in the Study of Alzheimer's Disease
Ron Brookmeyer, Ph.D., Department of Biostatistics
The Johns Hopkins University
Friday, May 5, 2000, 12:00-1:00 pm
6225 Medical Science Center
1300 University Avenue
This talk will describe two very different statistical problems in the study of Alzheimer's disease. The first problem concerns methods for projecting the future prevalence and incidence of Alzheimer's disease in the United States and the potential impact of interventions to delay disease onset. The approach uses age specific disease incidence rates together with assumptions about survival to reconstruct disease prevalence. We found that the prevalence of Alzheimer's disease in the United States will nearly quadruple over the next fifty years.
The second problem concerns the analysis of multidimensional longitudinal data. A statistical challenge in the analysis of clinical trials for Alzheimer's disease is to summarize a treatment effect from data on multiple outcomes including cognitive, functional and emotional status. The question is, can we summarize the effect of the treatment on the multidimensional longitudinal data by one single parameter that has a simple interpretation? The basic idea of our proposed method is that disease progresses according to different internal clocks and the effect of treatment is to accelerate or decelerate the time scale of the internal clocks. This generates a rich family of parametric models to unify the analysis of continuous, binary and time to event data. The methods are applied to data from the Johns Hopkins Alzheimer's Disease Research Center.
Back to General Departmental Seminar Series