Analysis of Recurrent Event Data in the Presence of Terminal Events and Missing Covariates

• Activity Number: 251
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Shankar Viswanathan*+ and Jianwen Cai
• Companies: The University of North Carolina at Chapel Hill and The University of North Carolina at Chapel Hill
• Keywords: Recurrent events ; Terminal events ; Missing data ; EM algorithm

Abstract

In many epidemiological studies, often interests are to examine a relationship between covariates and recurrent events (eg. infections). Often, some covariates have missing values for various reasons and complete case analysis performed may be biased or inefficient. We present a method for estimating the parameters in the marginal rate model for analyzing recurrent event data in the presence of terminal events. We adopt a weighted estimating equation approach with missing data assumed to be MAR. The parameters are estimated via weighted EM algorithm. Simulation studies showed that the estimates are approximately unbiased and the performance is better with smaller terminal event rates with various missing percentages. Our method is useful in reducing bias and improving the efficiency of the estimates. We applied the proposed method to Indian renal transplant cohort data for illustration.