322 - Reducing Falls in Stroke Rehabilitation: A Quality Improvement Case-Control Study

Thursday, February 19, 2026

5:00 PM - 6:30 PM AST

Presenting Author(s)

DM

Dylan McCaleb, MD

Resident Physician
UVA Health System
Charlottesville, Virginia, United States

Co-Author(s)

CW

Carley Whitt, MD

Resident Physician
University of Virginia Medical Center
Charlottesville, Virginia, United States
AG

Aileen Giordano, MD

Assistant Professor of PM&R
University of Virginia School of Medicine
Charlottesville, Virginia, United States

Submitter(s)

DM

Dylan McCaleb, MD

UVA Health System
Charlottesville, Virginia, United States

Objectives: Falls are a significant cause of morbidity among stroke patients in Inpatient Rehabilitation Facilities (IRFs). With limited evidence regarding stroke-specific features that contribute to fall risk, we aim to identify stroke-related factors that are associated with falls among stroke patients in an IRF in order to guide development of targeted fall risk stratification tools.

Design: A matched retrospective case-control study was conducted at our IRF between July 2024 and June 2025. Stroke patients who experienced a fall during their rehabilitation stay (n = 40) were compared to matched controls who did not fall (n = 45). Data extracted from the electronic medical record included demographics, stroke location, stroke-related deficits, comorbidities, and functional level. A comparative analysis using ANOVA testing was employed to evaluate the significance of associations between individual risk factors.

Results: Patient demographics, comorbidities, medications, stroke location, and functional level were similar between groups. However, the study revealed that of the examined fifteen stroke-specific factors, nine were significantly more common in the fall group (impaired orientation, impaired attention, impaired short-term memory, agitation, impulsivity, apraxia, hemineglect, expressive aphasia, and hemorrhagic stroke). The fall group had a mean of 4.22 of these significant risk factors present compared to 1.31 in the control group.

Conclusions: Nine stroke-specific deficits were identified as statistically more common in the fall group, suggesting these risk factors could be used to identify stroke patients at higher fall risk. Based on these results, the rehabilitation team will flag patients with 3 or more of the 9 identified risk factors present on admission and implement proactive measures to reduce their fall risk. By analyzing fall rates over time, we aim to validate a scoring tool that predicts which stroke patients are at the highest risk of falls, and, ultimately, we hope to reduce the fall rate among stroke patients in IRF.