New Approach for Analyzing Longitudinal Data on Health, Physiological State, and Survival Collected Using Different Observational Plans

• Activity Number: 75
• Type: Contributed
• Date/Time: Sunday, July 31, 2011 : 4:00 PM to 5:50 PM
• Sponsor: Section on Health Policy Statistics
• Author(s): Anatoliy Yashin*+ and Igor Akushevich and Konstantin Arbeev and Alexandr Kulminski and Svetlana Ukraintseva
• Companies: Duke University and Duke University and Duke University and Duke University and Duke University
• Keywords: combining data ; health histories ; physiological age trajectories ; gaussian approximation ; stochastic differential equations ; stopping times

Abstract

In a number of longitudinal studies individual health status and physiological/biological variables are repeatedly measured during the life course for a large number of study subjects. The data collected in such studies could be used for studying dynamic connection between these two components of aging related changes in humans. In this paper, we develop new method of statistical modeling appropriate for joint analyses of longitudinal data on aging, health, and longevity collected using different observational plans and applied it to analyses of Framingham data. The method is based on a mathematical model describing evolution of a stochastic process with two mutually dependent components. Available datasets are characterized by different observational plans which play a crucial role in specifying the parameter estimation procedure. The results of our analyses indicate that parameters of