Estimating the Effect of a Time-Dependent State Transition on the Recurrent Event Rate

• Activity Number: 672
• Type: Invited
• Date/Time: Thursday, August 13, 2015 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Douglas Earl Schaubel* and Abigail Smith and Nathan Goodrich
• Companies: University of Michigan and University of Michigan and Arbor Research Collaborative for Health
• Keywords: Organ failure ; Recurrent event ; State transition ; Time-dependent

Abstract

Recurrent events (e.g., hospitalizations) are often of interest in observational studies. In the setting we consider, all subjects begin follow-up in the same state (e.g., without renal failure), with some subjects subsequently experiencing a state transition (e.g., renal failure onset); in this sense, the transition is time-dependent. Primary interest lies in the impact of the state transition on the recurrent event rate. The essential contrast is within subjects who experience the state transition, and between the post-transition event rate and the counterfactual event rate that would have occurred in the absence of the transition. Asymptotic properties are developed and assessed in finite samples through simulation. The proposed methods are applied to estimate the effect on the hospitalization rate of end-stage renal disease onset among liver transplant recipients.