<p>GRADE certainty of evidence.</p>

This study aimed to evaluate the effects of exercise interventions on the Freezing of Gait Questionnaire (FOGQ), Unified Parkinson’s Disease Rating Scale Part III (UPDRS-III), Berg Balance Scale (BBS), and Mini-Balance Evaluation Systems Test (Mini-BESTest), and to explore potential sources of heterogeneity.</p> <p>Methods</p><p>Standardized mean differences (SMDs; Hedges’ g) with 95% confidence intervals (CIs) were pooled using random-effects restricted maximum likelihood models.

Exploratory subgroup analyses examined age, disease duration, intervention duration, training frequency, single-session duration, and exercise modality as potential sources of heterogeneity. Risk of bias was assessed using the Cochrane Risk of Bias 2 tool, and certainty of evidence was assessed using Grading of Recommendations Assessment, Development and Evaluation. This review was registered in PROSPERO (CRD420261281718).</p> <p>Results</p><p>The meta-analysis included 20 randomized controlled trials published between 2009 and 2024, involving 714 patients with PD, including 372 participants in the exercise intervention groups and 342 in the control groups.