Development and Validation of Predictive Models of Self-care and Motor Function at Discharge from Post-acute Rehabilitation for Adults 1072

Thursday, October 30, 2025

10:45 AM - 11:00 AM

Location: ROOM: Private Dining Room #1 >>> DIRECTIONS: Third Floor >>> From the Main Elevator Bank, proceed left past Williford and Astoria. PDR #1 is located to the left in the small corridor.

Presenting Author(s)

Alison M. Cogan, PhD, OTR/L (she/her/hers)

Assistant Professor
University of Southern Californian
Los Angeles, California, United States

Slides Slides

Non-presenting Author(s)

DN

Dingyi Nie, n/a

Graduate Student, Department of Computer Science, University of Southern California
University of Southern California
Los Angeles, California, United States
DY

Dongze Ye

PhD student
University of Southern California
Los Angeles, United States
NS

Nicolas Schweighofer, PhD

Professor
University of Southern California
Los Angeles, California, United States

We will report on the development and validation of predictive models of self-care and mobility function at discharge from inpatient rehabilitaion for adults with stroke. We applied a machine learning analysis to a large national sample, and used variables that are collected as part of standard of care in inpatient rehabiliation settings. Implications for clinical practice will be discussed.

Learning Objectives:

Identify key variables that predict self-care and mobility outcomes at discharge from inpatient rehabilitation facilities for adults with stroke.
Describe the general premise of random forest analysis.
Identify potential applications of prediction models for clinical practice.