Controlled clinical trials that evaluated physical activity or exercise interventions aimed at the treatment of idiopathic Parkinson's disease were eligible for inclusion. To be included, trials had to report one of the following outcomes: postural instability, deficits in balance demanding activities or health-related quality of life. Trials that evaluated the immediate effects of sensory cueing strategies or the immediate effect of interventions such as whole body vibration only were excluded.
Interventions varied widely between trials and included muscle strengthening, range of motion, balance training, walking, home based exercise, treadmill training, occupational therapy, relaxation, dance lessons, Qigong, and flexibility training. There was also a wide range of frequency and duration of interventions between trials. Control groups included visit to Parkinson's disease nurse, aerobic exercise, medical therapy only, treadmill, other active intervention or no intervention. The majority of participants were male; mean age ranged from 62 years to 77 years. Severity of Parkinson's disease varied between trials (details reported in review); mean duration of Parkinson's disease ranged from 3.3 years to 10.6 years, where reported. Outcomes included postural instability, balance task performance and health-related quality of life, which were measured by a wide variety of different instruments (details reported in the review).
Three reviewers independently selected studies for inclusion.