Background: Seventy-five percent (75%) of full-time wheelchair or scooter users with multiple sclerosis (MS) experience at least one fall in a period of 6 months. Falls are detrimental for the independence, quality of life, and community participation. No previous prospective study has evaluated fall risk factors in this segment of MS community.

Objective: To develop a multivariable falls risk prediction model for people with MS (PwMS) who use a wheelchair or scooter full-time.

Methods: This prospective cohort study is a secondary data analysis that recruited PwMS from the community across the US. Forty-eight adults with MS who use a wheelchair or scooter as their main form of mobility were included. Dependent variable was fall incidence over 3-month recorded through diaries. Dependent variable was categorized as fallers (≥ 1 fall) and non-fallers (0 fall). Predictors were demographics, history of falls in the past 6 months, measures of fear of falling, Spinal Cord Injury- Falls Concern Scale, California Verbal Learning Test II, and Multiple Sclerosis Quality of Life- mental health. Multivariable logistic regression analyses were conducted to identity predictors of future falls.

Results: In total, 63 falls (range 0 - 8) were reported over the 3-month period from a total of 26 fallers (54%). Multivariate logistic regression analyses indicated that the risk factor with the best predictive ability of future falls included history of falls in the past 6 months (sensitivity 77%, specificity 54%, and area under the receiving operating curve statistic = 0.76, 95% CI 0.62 to 0.89).

Conclusions: Findings highlight the importance of asking full-time wheelchair and scooter users with MS if they have fallen in the past 6 months to quickly identify those who are at increased fall risk and in need of follow up assessment and intervention to identify and address modifiable risk factors. More attention to fall risks among full-time wheelchair or scooter users with MS is suggested to increase the understanding among clinicians and researchers of modifiable risk factors.