Dose-Response Relationship of Virtual Reality Training in Parkinson's Disease: A Meta-Analysis of Overall Motor Function, Functional Mobility, and Balance Confidence.
Lin Xie, Jing Dai
✦ AI-curated · Sources linked
In 30 seconds
This meta-analysis of 31 randomized controlled trials involving 1,110 people with Parkinson’s disease examined whether virtual reality training (VRT) effects on motor function, mobility, and balance confidence varied with training dose. Overall, VRT improved Timed Up and Go (TUG) scores (weighted mean difference [WMD] = -3.69) and Activities-specific Balance Confidence scores (WMD = 11.80), with both changes meeting the minimal clinically important difference. Overall UPDRS-III improvement was not significant. TUG benefits were associated with sessions of 21–40 minutes, 4–7 times weekly, for 6–8 weeks; combining VRT with physical therapy was associated with clinically important effects.
Key findings
- Across trials, VRT improved TUG scores (WMD = -3.69), meeting the minimal clinically important difference.
- Balance confidence improved with VRT (ABC WMD = 11.80), also meeting the minimal clinically important difference.
- VRT alone improved TUG but did not reach the minimal clinically important difference; combined VRT and physical therapy showed clinically important benefit.
- No significant overall UPDRS-III improvement was observed, although high-frequency training (≥4 sessions/wk) was associated with potential benefit.
Why it matters
The findings suggest that VRT’s effects in Parkinson’s disease may differ by outcome and training dose: mobility and balance confidence improved overall, whereas overall motor-function scores did not. The reported dose parameters and the apparent added value of combining VRT with physical therapy may help frame rehabilitation planning, while the evidence does not establish a universally optimal protocol.
What to keep in mind
The authors caution that the finding of no statistical heterogeneity among patients aged ≥70 years is based on few studies and may reflect limited statistical power. Dose-response estimates are drawn from meta-analysis and meta-regression of trial data.