Quantifying directional variability during sit-to-stand for post-stroke fall risk stratification: a pilot study

BACKGROUND: The Berg Balance Scale (BBS) has limited measurement granularity due to floor/ceiling effects and ordinal scoring, compromising fall prediction precision. Sensor-based sit-to-stand (STS) assessments yield continuous quantitative data, thereby overcoming the granularity limitations of ordinal scales. STS directional variability may reflect impaired feedforward motor control, serving as a sensitive indicator of fall risk in stroke survivors. OBJECTIVE: To determine whether quantifying STS directional variability using a pressure sensor provides added predictive value for post-stroke falls beyond the BBS. METHODS: This prospective observational cohort study enrolled healthy controls and stroke patients able to sit independently and follow two-step verbal instructions, excluding those with balance-affecting comorbidities. Clinical evaluations, including the BBS, were performed upon transfer to rehabilitation. STS directional variability was quantified using a high-density pressure sensor mat to calculate the STS Direction Range, reflecting trial-to-trial center of pressure trajectory angle variability. Falls were assessed 3 months post-discharge. RESULTS: Twenty-four stroke patients (median age 74.0 years; 41.7% female) and 16 healthy controls (median age 62.5 years) were enrolled. Baseline BBS scores were higher in non-fallers, but the difference was not statistically significant. Fallers exhibited higher STS directional variability (median 150.0° vs. 115.0°; Cohen's d = 0.91). Adding STS Direction Range to the baseline model (Age + BBS) improved predictive performance (AUC: 0.574 to 0.768). CONCLUSION: Quantifying STS directional variability via high-density pressure sensors enhances post-stroke fall risk stratification beyond the BBS alone, offering valuable methodological insights for stroke rehabilitation.