Evaluating the contribution of reactive balance to prediction of fall rates cross-sectionally and longitudinally in persons with multiple sclerosis

Background: Falls are common in persons with multiple sclerosis (PwMS). Reactive postural control—one's response to a balance perturbation—is likely an aspect of fall risk; however, the relationship between reactive posture and falls is poorly understood in PwMS. Objective: We evaluated tibialis anterior muscle onset latency (TA latency) after balance perturbations as a predictor of fall rates in PwMS, controlling for clinical, functional, sensory, psychological, and cognitive factors. Method: At baseline of the 18-month cohort study, 122 participants with MS (EDSS = 2.23) were included. Assessments were conducted every 6 months. Results: Of the original 122 participants at the baseline collection, data were available from 41, 39, and 34 people at the 6, 12, and 18 month follow-ups, respectively. Percent fallers at the four time points were 35.3%, 12.2%, 15.4%, and 20.5%. Cross-sectionally (i.e., at baseline), the Timed Up-and-Go, Falls Efficacy Scale – International (FES-I), and TA latency after perturbations were significant predictors of retrospective falls rates using negative binomial regression. Longitudinally, random-effects negative binomial regression found that trait-level FES-I, Stroop Color-Word, and TA latency were significant predictors for falls rates. Conclusion: Delays in automatic postural responses seem to account uniquely for fall rates in PwMS—beyond clinical, balance, or mobility measures. These delays may contribute to the increased fall rate in PwMS. In addition to brief self-report instruments (FES-I) and cognitive assessments, muscle onset after balance perturbations may be a valuable tool for predicting falls in those with MS.