Predicting faller status in persons with multiple sclerosis using the Multiple Sclerosis Walking Scale-12

Background: Persons with multiple sclerosis (PwMS) are at an increased risk for falling, making it necessary to identify useful screening tools. The aims of this study were to 1) determine a cut-off score for the 12-item Multiple Sclerosis Walking Scale (MSWS-12) for identifying PwMS as fallers and 2) evaluate its predictive ability of faller status after controlling for other potential contributing factors. Methods: Participant characteristics, MSWS-12, and falls in the last six months were collected on PwMS (n =171) during a single session. Fallers (53.8 %; n = 92) were individuals reporting >= 1 fall in the past six months. A receiver-operating-characteristic (ROC) curve was performed to estimate the classification accuracy (area under the curve; AUC) of the MSWS-12 at detecting fallers. Optimal cut-off scores were calculated using the Youden Index and Index of Union methods. The dichotomized MSWS-12 cut-off score was then entered into a logistic regression, with faller status as the outcome, and age, gender, body mass index, disease duration, and fatigue as covariates. Results: The MSWS-12 had a fair classification accuracy for identifying fallers (AUC = 0.74), with the cut-off score of >= 46 % having 76.1 % sensitivity and 64.6 % specificity. The MSWS-12 cut-off score remained a significant predictor of faller status in the adjusted model (adjusted odds ratio [aOR]: 3.77, 95 % CI: 1.75, 8.15, P = .001), along with higher fatigue (aOR: 1.11, 95 % CI: 1.02, 1.20, P = .015). Conclusion: PwMS with MSWS-12 scores >= 46 % were more likely to be fallers than those with lower scores. When used in conjunction with a clinician's judgement and other assessments, the MSWS-12 may be a useful screening tool for identifying PwMS who are fallers.