An Estimating Function Approach to the Analysis of Recurrent and Terminal Events

• Activity Number: 59
• Type: Topic Contributed
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Douglas E Schaubel*+ and John Kalbfleisch and Yining Ye and Qi Gong
• Companies: University of Michigan and University of Michigan and Amgen, Inc. and University of Michigan
• Keywords: Cox model ; Estimating functions ; Frailty ; Recurrent event ; Terminating event

Abstract

In biomedical studies, the event of interest is often recurrent (e.g., hospitalization). It is often the case that a terminal event (e.g. death) may occur and hence stop the recurrent event process. We consider the recurrent/terminal event setting and model the dependence through a shared gamma frailty. Conditional on the frailty, a model is specified only for the marginal recurrent event process, hence avoiding the strong Poisson-type assumptions traditionally used. Analysis is based on estimating functions that allow for estimation of covariate effects on the conditional recurrent event rate (given survival) and the terminal event hazard. The method also permits estimation of the degree of association between the two processes. Asymptotic variance estimators are proposed. The proposed method is evaluated through simulations to assess the applicability of the asymptotic results in finite samples and the sensitivity of the method to its underlying assumptions. The methods can be extended in straightforward ways to accommodate multiple types of recurrent and terminal events. The methods are illustrated in an analysis of hospitalizations and death among dialysis patients.