A systematic review of factors associated with accidental falls in people with multiple sclerosis: a meta-analytic approach

Objective: To determine whether there are demographic, clinical, and instrumental variables useful to detect fall status of patients with multiple sclerosis. Data sources: PubMed and the Cochrane Library. Review methods: Eligible studies were identified by two independent investigators. Only studies having a clear distinction between fallers and non-fallers were included and meta-analysed. Odds ratios (ORs) and standard mean differences (SMDs) were calculated and pooled using fixed effect models. Results: Among 115 screened articles, 15 fulfilled criteria for meta-analyses, with a total of 2425 patients included. Proportion of fallers may vary from 30% to 63% in a time frame from 1 to 12 months. No significant publication bias was found, even though 12/15 studies relied on retrospective reports of falls, thus introducing recall biases. Risk factors for falls varied across studies, owing to heterogeneity of populations included and clinical instruments used. The meta-analytic approach found that, compared with non-fallers, fallers had longer disease duration (SMD = 0.14, p = 0.02), progressive course of disease (OR = 2.02, p < 0.0001), assistive device for walking (OR = 3.16, p < 0.0001), greater overall disability level (SMD = 0.74, p < 0.0001), slower walking speed (SMD = 0.45, p = 0.0005), and worse performances in balance tests (Berg Balance Scale: SMD = -0.48, p = 0.002; Timed up-and-go test, SMD = 0.31, p = 0.04), and force-platform measures (postural sway) with eyes opened (SMD = 0.71, p = 0.006) and closed (SMD = 0.83, p = 0.01), respectively. Conclusion: Elucidations regarding risk factors for accidental falls in patients with multiple sclerosis (PwMs) are provided here, with worse disability score, progressive course, use of walking aid, and poorer performances in static and dynamic balance tests strongly associated with fall status.