Online Program Home
My Program

Abstract Details

Activity Number: 401 - Lead with Statistics: Case Studies and Methods for Learning and Improving Healthcare Through EHRs
Type: Topic Contributed
Date/Time: Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract #329586
Title: Recalibrating Prognostic Risk Score Adapted to EHR Data
Author(s): Dandan Liu* and Hui Nian and Qingxia Chen
Companies: Vanderbilt University Medical Center and Vanderbilt University Medical Center and Vanderbilt University
Keywords: EHR; risk score; recalibration; loss function; prognostic index

Prognostic index is valuable to clinicians advising patients, as well as policy makers and epidemiologists' interest in risk adjustment. However, for prognostic index developed for outpatients, a fraction of risk measurements often relied on subjective descriptions of health status along with administrative data with variable resources, which limits their usage for larger populations. The increasingly adopted electronic health record (EHR) systems provides a feasible and valuable approach for automated implementation of a mortality risk prediction tool. When researchers carried the existing risk indexes to EHR, however, problems often arose including coarse or partial risk factors are available in EHR. Integrating the original mortality index with the EHR data, we proposed to estimate the optimal equivalent value for the partially collected risk factors by minimizing various loss functions. Simulation studies showed that the proposed methods outperformed the other commonly used approaches in practice. The proposed approaches were applied to the EHR data to develop an electronically-adapted mortality risk score index.

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

Back to the full JSM 2018 program