A Multi-state Stochastic Model for the Analysis of Longitudinal Dementia Data

• Activity Number: 21
• Type: Contributed
• Date/Time: Sunday, August 11, 2002 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section*
• Author(s): Sujuan Gao*+ and Jaroslaw Harezlak and Siu Hui
• Affiliation(s): Indiana University School of Medicine and Harvard School of Public Health and Indiana University School of Medicine
• Keywords: muti-state stochastic model ; longitudinal study ; nonignorable missing data

Abstract

Longitudinal dementia studies face many challenging issues, including multiple outcomes, panel data, nonignorable missing information due to death of elderly participants, and complex sampling design for ascertaining clinical outcomes. We propose a multi-state transitional stochastic model for the analysis of dementia data. The model includes three states for clinical outcomes: normal, cognitively impaired but not demented and demented state, and an absorbing state for death. Maximum likelihood approach will be used for parameter estimation from this model framework. Incidence of cognitive impairment and dementia will be estimated as well as mortality rates. Estimates of risk factor effects can be achieved by assuming proportional hazard model for the transitional hazards among the four states. The methodology can be extended to deal with data collected using complex sampling plans using the EM algorithm. We will illustrate our approach using data from the Indianapolis Dementia Study.