Group exercise for balance in people with Parkinson's disease: a systematic review with meta-analysis

High-quality evidence shows that exercise helps people with Parkinson's disease improve functional abilities including balance. However, few studies have investigated whether the setting and format through which balance-focused exercise programs are provided matters. This systematic review investigated group exercise compared to individual exercise, and to no-exercise control (CTL), on clinical measures of balance for people with Parkinson's disease.

MEDLINE, EMBASE, CINAHL and Cochrane CENTRAL databases were searched through March 24, 2024, on terms for Parkinson's disease; exercise or physical activity; community-based or group classes; balance or postural control. Citations, abstracts and full-text articles were independently reviewed, and included studies were rated on risk of bias by two authors.

Twenty-seven randomized controlled trials (30 reports) with 1200 participants met criteria. Meta-analysis using mean difference (MD) compared group exercise to CTL on seven clinical measures of balance. Three yielded significant differences favoring group exercise: Timed Up and Go = -2.29 (MD), -3.56 to -1.02 (95% Confidence interval) (95% CI); Mini-BEST = 2.72 (MD), 1.88 to 3.57 (95% CI); Berg Balance Scale = 4.31 (MD), 1.33 to 7.29 (95% CI). Meta-analyses were also conducted on six clinical measures of balance, comparing group exercise to individual exercise, yielding no significant differences.

For people with Parkinson's disease, group exercise may be more effective than CTL on some clinical measures of balance and it yields similar results to individual exercise. People with Parkinson's disease are encouraged to participate regularly in group or individual exercise based on preference and accessibility.