Skip to content

Virtual reality rehabilitation improved balance test scores after stroke, but confidence is limited

A review of randomized trials found better scores on several balance tests, but inconsistent study results and bias limit confidence.

By 100HP editorialAbstract-based explanation checked

Based on the published abstract. The full paper may contain additional methods, results and limitations.

The 30-second takeaway

In adults with stroke, pooled randomized trials favored virtual reality rehabilitation for several measures of balance and standing control. An eyes-closed sway-speed measure showed no clear difference. These results suggest possible rehabilitation benefits, but variation and bias across studies leave uncertainty about the size and everyday importance of those benefits.

Keep in mind

This abstract-based summary reports test outcomes. The review found substantial variation between studies and high risk of bias across several outcomes; the clinical relevance and magnitude of benefits still need confirmation.

THE NUMBERS, WITH CONTEXT

What researchers found

Mean difference 3.29 (95% CI 2.76–3.83)

Berg Balance Scale score

Virtual reality versus conventional rehabilitation; assessment timing was not reported in the abstract.

Medicina (Kaunas), 2025 · Original source ↓

What was compared

Trials tested virtual reality either alone or alongside conventional therapy in adults with stroke. The review separated standing-control measurements from dynamic balance tests, rather than treating balance as a single outcome.

Where results differed

Berg Balance Scale and Timed Up-and-Go results favored virtual reality. Several standing-control measures also favored it, but the eyes-closed sway-speed result did not show a clear difference, so improvement was not uniform.

CHECK THE ORIGINAL

The original publication

Effectiveness of Virtual Reality Therapy on Static Postural Control and Dynamic Balance in Stroke Patients: Systematic Review, Meta-Analysis, and Meta-Regression of Randomized Controlled Trials.

Tian MY, Lee MH, Kim JH et al.
Medicina (Kaunas) · 2025

PubMed ID
41597377
Record checked

AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.