High-Dimensional Fixed Effects Profiling Models: New Developments and Applications

• Activity Number: 54 - Recent Advances in Categorical Data Analytics
• Type: Topic Contributed
• Date/Time: Sunday, August 7, 2022 : 4:00 PM to 5:50 PM
• Sponsor: Biometrics Section
• Author(s): Danh Nguyen* and Jason Estes and Damla Senturk and Esra Kurum
• Companies: University of California, Irvine and Moutain View and University of California, Los Angeles and University of California, Riverside
• Keywords: profiling; fixed effects model; dialysis; time-dynamic; sparse data; hospital readmission

Abstract

Profiling analysis aims to evaluate health care providers, such as hospitals, nursing homes, or dialysis facilities etc., with respect to a patient outcome. fixed effects (FE) profiling methods have considered binary outcomes, such as 30-day hospital readmission or mortality. For the unique population of dialysis patients, (1) regular blood tests are required to evaluate the effectiveness of treatment and avoid adverse events, including dialysis inadequacy, imbalance mineral levels, and anemia among others, as well as (2) the need for continuous monitoring/care after transitioning to dialysis. We illustrate the versatility of FE profiling models through several applications in profiling dialysis facilities in the U.S. and recent FE model developments, including (a) time-varying/time-dynamic standardized readmission ratio, (b) profiling for adverse recurrent events, and (c) new insights on operating characteristics such performance of FE model under the low information context/sparse outcome data setting.