QUESTIONS: In people with Parkinson's disease, does home-based prescribed exercise improve balance-related activities and quality of life compared with no intervention? Are the effects of home-based exercise similar to those of equivalent centre-based exercise? DESIGN: Systematic review and meta-analysis of randomised and quasi-randomised controlled trials. PARTICIPANTS: Adults diagnosed with idiopathic Parkinson's disease. INTERVENTION: Predominantly home-based prescribed exercise (defined as a minimum of two-thirds of the exercise being completed at home). The intervention had to primarily involve physical practice of exercises targeting gait and/or standing balance compared with either control (ie, usual care only, a sham intervention or no physiotherapy) or equivalent predominantly centre-based exercise. OUTCOME MEASURES: The primary outcome was balance-related activities and the secondary outcomes were gait speed, Berg Balance Scale, Functional Reach test, and quality of life. RESULTS: Sixteen trials met the inclusion criteria and all contributed to the meta-analyses. Twelve trials compared home-based prescribed exercise with control, and four trials compared home-based prescribed exercise with equivalent centre-based exercise. Home-based prescribed exercise improved balance-related activities (SMD 0.21, 95% CI 0.10 to 0.32) and gait speed (SMD 0.30, 95% CI 0.12 to 0.49), but not quality of life (SMD 0.11, 95% CI -0.01 to 0.23) compared with control. Home-based and centre-based exercise had similar effects on balance-related activities (SMD -0.04, 95% CI -0.36 to 0.27) and quality of life (SMD -0.08, 95% CI -0.41 to 0.24). CONCLUSION: Home-based prescribed exercise improves balance-related activities and gait speed in people with Parkinson's disease, and these improvements are similar to improvements with equivalent centre-based exercise.