Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 349 - Lifetime Data Science Student Awards
Type: Topic-Contributed
Date/Time: Thursday, August 12, 2021 : 10:00 AM to 11:50 AM
Sponsor: Lifetime Data Science Section
Abstract #317159
Title: Analysis of Readmissions Data Taking Account of Competing Risks
Author(s): Wenbo Wu* and Kevin He and Xu Shi and Douglas E. Schaubel and John D. Kalbfleisch
Companies: University of Michigan and University of Michigan and Department of Biostatistics, University of Michigan and University of Pennsylvania and University of Michigan
Keywords: Discrete survival model; Cause-specific hazard; Provider profiling; Standardization; Robust inference

The 30-day unplanned hospital readmission rate has been used in provider profiling for evaluating hospital-facility care coordination, medical cost-effectiveness, and patient quality of life. Current profiling analyses use logistic regression to model readmission as a binary outcome, and the presence of competing risks (e.g., death) is not explicitly considered. Overlooking competing risks leads to an underestimation of the true readmission rate and invalid profiling analysis. To address these drawbacks, we propose a discrete competing risk model of readmission within a cause-specific hazards framework. Hazards of the event processes are sequentially formulated to exempt the nuisance competing risk hazard from a full specification. To foster model fitting with high-dimensional parameters and facility-effect inference with patient-level clustering, we also develop a Blockwise Inversion Newton algorithm with scalability and memory efficiency, and a stabilized robust score test suitable even for facilities with extreme outcomes. Evidence from simulations and application demonstrates the superior performance of our proposed methods over existing analyses.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program