Efficacy of perturbation balance training on reducing falls risk – an umbrella review

4650

Catherine Marx1, Grant Trewartha1, Laura Cordova-Rivera2, Hollie Rice1, Lotte Janssens3, Samantha L Harrison1

1 School of Health & Life Sciences, Teesside University UK; 2 Physiotherapy, Faculty of Health & Wellbeing, School of Sport, Exercise & Rehabilitation, Northumbria University, UK; 3 REVAL Rehabilitation Research Center, Hasselt University, Belgium

Scientific Research

SP - Falls

Falls

View DOI

Abstract

Introduction

Avoiding a fall after an unexpected slip or a trip requires rapid reactive balance responses, which may be trained via perturbation balance training (PBT). However, PBT protocols vary and their effectiveness is unclear. This umbrella review explored whether PBT, compared to no or conventional balance training (CBT), is effective in reducing falls risk.

Methods

Systematic reviews with or without meta-analyses were searched in CINAHL, MEDLINE, AMED, EMBASE, PubMed, PEDro, Cochrane database, and Google Scholar and screened by two reviewers. AMSTAR2 was used for quality assessment. Overlap was addressed through the Corrected Cover Area Index. Narrative synthesis was applied to the results.

Results

Seven reviews were eligible. Populations included healthy older adults (HOAs, n=2), people with stroke (n=2); multiple sclerosis (MS, n=1); rheumatoid arthritis (n=1), and Parkinson’s disease (PD, n=2). Reviews compared PBT with usual treatment, or no training. Quality was rated as ‘critically low’ (n=6) and ‘low’ (n=1). A 2.6% overlap was observed.

Two reviews showed marginal improvements for falls (primary outcome) post-PBT. Fall risk factors were addressed secondarily, including balance (n=7), Posturography (n=4), gait (n=4), and balance confidence (n=3). Meta-analyses (n=2) showed moderate and no effects of PBT on balance respectively (SMD=0.60 [95% CI 0.15, 1.06], p=0.01; SMD=0.09 [95% CI −0.40, 0.58], p=0.720). Postural sway improved in people with PD (MD= −1.74 [95% CI −3.18, −0.29]; p=0.02). Onset latency improved in both people with MS (MD [ms] 36.11 [95% CI 17.61, 54.60]; p<0.001) and healthy controls (MD [ms] 33.25 [95% CI 12.31, 54.19]; p=0.002). Meta-analyses (n=2) on gait and balance confidence found no improvements (SMD 0.16 [95% CI -0.18, 0.49], p=0.36) (SMD 0.11 [95% CI −0.24, 0.45]; p=0.55); however, narrative syntheses found significant improvements in stepping characteristics.

Conclusions

Efficacy of PBT on fall risk-related outcomes is inconclusive, which may be due to heterogeneity of protocols and outcome measures.