809 Views
Health Services Research
Quality Improvement and Implementation Science
Cross-Cutting
Oral Presentation
Matt Malcolm, PhD, OTR/L
Associate Professor
Colorado State University
Fort Collins, Colorado
James Graham, PhD
Director, Center for Community Partnerships; Professor, Department of Occupational Therapy
Colorado State University
Fort Collins, Colorado
Julia Sharp, PhD
Director, Graybill Statistical Laboratory; Associate Professor
Colorado State University
Fort Collins, Colorado
This study identified patient characteristics which predict acute care rehabilitation utilization. Through a cross-sectional study of secondary, de-identified electronic health records (EHR) data we found that patients who were racial/ethnic minorities, female, or who lacked a significant other appeared to experience disparity in rehabilitation utilization.
Objective:
Identify patient characteristics which predict acute care rehabilitation utilization.
Design:
This was a cross-sectional study of secondary, de-identified electronic health records (EHR) data. Statistics: linear mixed and generalized linear mixed models.
Setting :
Five large acute care hospitals.
Participants (or Animals, Specimens, Cadavers) : EHR data were obtained from all patients admitted to the five hospitals during June 2014-June 2018.
Interventions : Interventions: Not applicable.
Main Outcome Measure(s):
Dependent variables: receipt of occupational therapy (OT), receipt of physical therapy (PT), onset of OT and PT, number of OT and PT encounters. Predictor variables: race/ethnicity, gender, age, presence of significant other (SO), length of stay (LOS), fall risk score, initial self-care and mobility scores.
Results:
Regression results revealed significant predictors of receiving OT were male gender (odds ratio (OR) 1.2), older age (OR 1.1), no SO (OR 1.2), longer LOS (OR 1.1), and higher fall risk scores (OR 1.9). While these same variables significantly predicted receipt of PT, being of racial/ethnic minority status also predicted PT receipt (OR 1.1). Significant predictors (all p<.001 except gender) of a later OT onset (relative to hospital admission) were racial/ethnic minority, female gender (p=.04), younger age, presence of SO, lower fall risk score, longer LOS, and higher initial self-care score. Similar results occurred when predicting PT onset. Receiving fewer PT encounters was predicted by being a minority, older age, no SO, lower fall risk, shorter LOS, and lower mobility score (p<.001). Similar findings were obtained for OT encounters, however minority status, age, and SO were not significant predictors.
Conclusions : Patients who were racial/ethnic minorities, female, or who lacked a significant other appeared to experience disparity in rehabilitation utilization. Further research is needed to identify why these disparities exist.