# Prognostic Equations and Accuracy of a Total Score of Functional Independence Measure at Discharge for Different Diseases in a Convalescent Rehabilitation Ward

## TL;DR

This study develops equations to predict patient recovery outcomes in rehabilitation, focusing on different diseases and their recovery accuracy.

## Contribution

The study introduces disease-specific prognostic equations for predicting functional recovery in convalescent rehabilitation.

## Key findings

- The study identified admission motor and cognitive FIM as key predictors for all four diseases.
- Prediction accuracy varied by disease, with TKA showing the highest accuracy followed by HF, stroke, and VCF.

## Abstract

Objectives: Prognosis and goal setting from admission in the convalescent rehabilitation ward, supported by a multidisciplinary team, enhance rehabilitation and discharge support. Predicting functional independence measure (FIM) outcomes can further optimize these processes. This study aimed to develop prognostic equations for the motor FIM at discharge for stroke, hip fracture (HF), vertebral compression fractures (VCFs), and total knee arthroplasty (TKA), which are common diseases in patients admitted to convalescent rehabilitation wards, using multiple regression analysis, and to clarify the difference in the accuracy of the predicted motor FIM according to the disease.

Methods: This study included 965 patients admitted to our hospital. The objective variable consists of the motor FIM at discharge, and the explanatory variables were age, sex, days from onset to admission, total admission motor FIM, and total admission cognitive FIM. A stepwise multiple regression analysis was performed. The analysis of the difference in the accuracy of predicted motor FIM by disease used the absolute value of the residuals.

Results: The total motor FIM and cognitive FIM at admission were extracted for all four diseases included in this study. The absolute value of the residuals appeared to be more accurate for TKA, HF, stroke, and VCF in that order.

Conclusions: Although differences in the accuracy of the prediction equation were observed by disease, this prediction equation can be used as an approach to review the details of rehabilitation and discharge and can be tailored to each case.