THE STATISTICAL-ANALYSIS OF EVENT HISTORIES IN LONGITUDINAL-STUDIES OF AGING

Data from longitudinal studies have a number of advantages for geronto logical research, but the effects of attrition and the increased compl exity of data structures with multiple observations may pose some prob lems for statistical analysis. Proportional hazards models allow for t he examination of event histories using all observations of a dependen t variable, and these models can incorporate time-dependent covariates to increase their explanatory power. The assumptions and applications of event history analysis using proportional hazards models are descr ibed, and the analysis of mortality data from the Ontario Longitudinal Study of Aging provides a relevant example. Extensions of proportiona l hazards models and commercially available software are also discusse d.