Abstract W P134: Inpatient Rehabilitation Facility Admission Outcomes Specific for Patients with Stroke Predict Community Discharge

Abstract

Objective: To use IRF admission outcomes specific for patients with stroke to predict community discharge. Design: Retrospective study Setting: Inpatient Rehabilitation Hospital Participants: Patients hospitalized for a stroke Interventions: N/A Main Outcomes Measures: Admission data from the Functional Independence Measure (FIM) and Stroke Rehabilitation Assessment of Movement (STREAM) were used to determine predictive factors for a community discharge and FIM walk / wheelchair score of 5 or above. A discriminant analysis, logistic regression, and chi-square analysis were used to identify admission variables. Results: 296 subjects were included in this retrospective analysis; 24% of subjects did not have a community discharge and 54% did not discharge with a FIM walk / wheelchair score of 5 or above. Five admission factors were identified as predictive for a community discharge or a walk / wheelchair FIM score of 5 or above; the Stream, Motor FIM, Total FIM, FIM Bladder control, and FIM social interaction. The sensitivity for each variable ranged from 61% - 94% while the specificity was between 40% and 65%. When all five variables were combined 83% of non-community discharges were identified (77% of patients with a walk wheelchair score of below 5). Conclusions: By using admission outcomes, community discharge can be predicted with significant sensitivity. Patients identified as at risk for a non-community discharge may be targeted for specialized interventions. Prospective evaluation of the predictive ability of the above-identified factors is warranted.