Improved Health Policy Decisionmaking via Efficiency Gains with Recurrent Events

• Activity Number: 257
• Type: Topic Contributed
• Date/Time: Tuesday, August 5, 2008 : 10:30 AM to 12:20 PM
• Sponsor: Section on Health Policy Statistics
• Author(s): Edsel A. Pena*+ and Akim M. Adekpedjou and Jonathan Quiton
• Companies: University of South Carolina and University of Missouri-Rolla and Western Kentucky University
• Keywords: recurrent monitoring ; Koziol-Green model ; sum-quota accrual

Abstract

Health policy decisions are typically based on the outcomes of statistical analysis of observed health and biomedical data. With reliable inference from such health data, then reliable policy decisions will be made. Oftentimes, in the context of failure time monitoring, statistical inference methods are usually based on the time-to-first event, and the usual methods of analysis are based on the Kaplan-Meier estimator and Cox regression. However, in many settings, the event of interest is recurrent. In this situation, we illustrate the potential gain in efficiency when doing inference by considering all the observed recurrences, and how this could translate to improved health policy decisions. We do this efficiency analysis by extending the Koziol-Green model, which was used in efficiency analysis for single-event models.