Related Experiment Video
Updated: Jul 20, 2025
Quantitative Static and Dynamic Assessment of Balance Control in Stroke Patients
Published on: May 17, 2020
Clinical prediction rules for multi surfaces walking independence using the Berg Balance Scale or Mini-Balance
Shuntaro Tamura1, Sota Kobayashi2,3, Ren Takeda4
1Department of Rehabilitation, Fujioka General Hospital, Fujioka, Japan.
Background:
A Clinical prediction rule (CPR) for determining multi surfaces walking independence in persons with stroke has not been established.
Objectives:
To develop a CPR for determining multi surfaces walking independence in persons with stroke.
Methods:
This was a multicenter retrospective analysis of 419 persons with stroke. We developed a Berg Balance Scale (BBS)-model CPR combining the BBS, comfortable walking speed (CWS) and cognitive impairment, and a Mini-Balance Evaluation Systems Test (Mini-BESTest)-model CPR combining the Mini-BESTest, CWS, and cognitive impairment. A logistic regression analysis was conducted with multi surfaces walking independence as the dependent variable and each factor as an independent variable. The identified factors were scored (0, 1) based on reported cutoff values. The CPR's accuracy was verified by the area under the curve (AUC). We used a bootstrap method internal validation and calculated the CPR's posttest probability.
Results:
The logistic regression analysis showed that the BBS, CWS, and cognitive impairment were factors in the BBS model, and the Mini-BESTest was a factor in the Mini-BESTest model. The CPRs were 0-3 points for the BBS model and 0-1 points for the Mini-BESTest model. The AUCs (bootstrap mean AUC) of the CPR score were 0.89 (0.90) for the BBS model and 0.72 (0.72) for the Mini-BESTest model. The negative predictive value (negative likelihood ratio) was 97% (0.054) for CPR scores < 2 for the BBS model and 94% (0.060) for CPR scores < 1 for the Mini-BESTest model.
Conclusions:
The CPR developed herein is useful for determining multi surfaces walking independence.
