Effects of virtual reality on the balance performance of older adults: a systematic review and meta-analysis.

High heterogeneity across studies (I² 66% to 80%) because the VR interventions and control conditions varied widely in type, dose, and duration. 80% of included studies had selection bias in how they generated their random sequence, and 52% failed to conceal allocation. Most studies did not blind participants or therapists, and only about 32% used a blinded assessor. Only three of the included studies scored 8 or higher on the PEDro quality scale (out of 10). Only four studies measured the ABC scale, so the null confidence finding rests on a very small evidence base. Non-English publications were excluded, which may omit relevant trials. The authors note that whether VR can replace traditional exercise therapy remains unclear and that more high-quality samples are needed.

Funded by Kaohsiung Veterans General Hospital (Project VGHKS109-131). The authors declared no potential conflicts of interest.

Do not read the Discussion's claim that VR increases balance confidence as supported by the data. The ABC scale result was not significant (p=0.30), and only four studies measured it. The authors' inference that VR reduces fall risk through improved confidence is not backed by their own pooled result.