Risks of falls in subjects with multiple sclerosis

variables associated with falls. Design: Retrospective case-control study design with a 2-group sample of convenience. Setting: A hospital and home settings in Italy. Participants: A convenience sample of 50 people with MS divided into 2 groups according to their reports of falls. Interventions: Not applicable. Main Outcome Measure: Subjects were assessed with questionnaires for cognitive ability and were measured on their ability to maintain balance, to walk, and to perform daily life activities. Data regarding patients' strength, spasticity, and transfer skills impairment were also collected. Results: No statistical differences were found between groups of fallers and nonfallers using variables pertaining to years after onset, age, gender, and Mini-Mental State Examination. Near statistically significant differences were found in activities of daily living and transfer skills (P<.05). Three variables were associated with fall status: balance, ability to walk, and use of a cane (P<.01). Those variables were analyzed using a logistic regression. The model was able to predict fallers with a sensitivity of 90.9% and a specificity of 58.8%. Conclusions: Variables pertaining to balance skills, gait impairment, and use of a cane differed between fallers and nonfallers groups and the incidence of those variables can be used as a predictive model to quantify fall risk in patients suffering from MS. These findings emphasize the multifactorial nature of falls in this patient population. Assessment of different aspects of motor impairment and the accurate determination of factors contributing to falls are necessary for individual patient management and therapy and for the development of a prevention program for falls.