# IMPROVING THE PREDICTION OF FUNCTIONAL RECOVERY IN OLDER ADULTS WITH STROKE IN GERIATRIC REHABILITATION USING AN INERTIAL MEASUREMENT UNIT COMBINED WITH THE UTRECHT SCALE FOR EVALUATION OF REHABILITATION

## TL;DR

Using wearable devices to measure balance improves predictions of recovery in older stroke patients compared to traditional scales.

## Contribution

Combining inertial measurement units with the USER scale enhances functional recovery prediction accuracy in geriatric stroke rehabilitation.

## Key findings

- Combining IMU data with the USER scale increased explained variance in recovery prediction (R² = 0.61 vs 0.30).
- 12 postural sway features were selected for PCA from 35 reliable features across two balance conditions.
- 16.9% of patients could not perform the second balance condition due to insufficient balance.

## Abstract

Prediction of functional recovery in older adults recovering from stroke is typically based on observational scales, such as the Utrecht Scale for Evaluation of Rehabilitation (USER). Objectively measuring postural sway using inertial measurement devices (IMU) may complement or improve conventional approaches. The aim of this study was to evaluate whether integrating an IMU with USER data enhances the accuracy of predicting functional recovery at discharge.

This prospective cohort study included older adults (≥ 65 years) recovering from stroke. Postural sway was assessed using an IMU during 2 different balance conditions and analysed using principal component analysis (PCA). Using 3 different regression models, percentage explained variance was compared to assess predictive performance on functional recovery of USER vs an IMU.

The 71 patients included had a mean age of 78 (SD 7.6) and a median time since stroke of 16 days (IQR 19–60). Of the 71 patients, 12 (16.9%) were unable to perform balance condition 2 due to insufficient balance. Of 35 postural sway features displaying reliability for both balance conditions, 12 were selected for PCA. Incorporation of principal components for both balance conditions in the final model increased the explained variance compared to a model in which only USER-mobility at admission was used to predict delta-USER at discharge (R2 = 0.61 vs 0.30).

Sitting and standing balance as measured by an IMU improves the prediction of functional recovery at discharge compared to USER alone.

After a stroke, older adults often go through rehabilitation to regain mobility and perform daily tasks. To predict how well someone might recover, healthcare professionals typically use observational scales like the Utrecht Scale for Evaluation of Rehabilitation (USER). This study explores whether using small, wearable devices called inertial measurement units (IMUs) to measure balance while sitting and standing can improve predictions. The study included 71 patients with an average age of 78 years. By combining the USER scale with IMU measurements of balance, we found a improvement in predicting patients’ recovery by the time they were discharged. This approach explained more of the variance in recovery outcomes compared to using the USER scale alone.