One-Leg Standing Time of the Affected Side Moderately Predicts for Postdischarge Falls in Community Stroke Patients.

Background: The purpose of the present study was to investigate the predictive accuracy of one-leg standing time at hospital discharge on falls in stroke patients.

Methods: This was a retrospective cohort study. Participants included stroke patients (n = 65) who could walk when discharged from inpatient rehabilitation ward. To investigate the relationship between one-leg standing time and falls, logistic analysis was utilized with a criterion variable including the presence or absence of falls after 1-year hospital discharge as well as explanatory variables including Brunnstrom stage, knee extension strength on the affected side, Barthel Index, 10-m walking speed, and one-leg standing time on both sides. The accuracy of prediction by one-leg standing time was measured by the area under the curve of the receiver operating characteristic curve.

Results: Among the 65 patients, 38 (58.5%) experienced a fall 1 year after discharge. One-leg standing time of the affected side was not significantly associated with the falls (odds ratio: .89; 95% confidence interval: .79-1.01). When the fall incidents were assessed by area under the curve of the receiver operating characteristic curve, one-leg standing time of the affected side was observed to have increased marginally to .93 (95% confidence interval: .87-.99) as compared to the traditional prediction mode area under the curve (area under the curve .88; 95% confidence interval: .81-.97).

Conclusions: One-leg standing time of the affected side may be considered as a moderately effective and simple assessment method for predicting postdischarge falls in a clinical setting.