JMIR Publications

ABSTRACT

Background:

Virtual reality (VR) interventions have been increasingly applied in health care, but their multifaceted impacts on physical, mental, and social health outcomes remain unclear. Existing systematic reviews and meta-analyses have reported inconsistent findings, highlighting the need for an umbrella review to synthesize the highest level of evidence.

Objective:

This study employed an umbrella review approach to systematically integrate current evidence from systematic reviews and meta-analyses on the impact of virtual reality (VR) interventions on multidimensional health outcomes. It aims to comprehensively evaluate the effectiveness, strength of evidence, and limitations of VR interventions across different populations and outcomes, thereby providing a basis for clinical practice and future research.

Methods:

Strictly adhering to PRISMA-P guidelines, systematic searches were conducted in databases such as Embase, Medline, Cochrane Library, and Web of Science from their inception to August 2025. Systematic reviews and meta-analyses evaluating the impact of VR interventions on human health outcomes were included. Two researchers independently performed literature screening and data extraction, while the AMSTAR2 2 and GRADE tools were used to assess methodological quality and evidence quality, respectively.

Results:

Multiple meta-analyses were integrated. The findings revealed that the effects of VR interventions exhibited significant "population-outcome" specificity. VR interventions demonstrated positive effects in improving upper limb function in stroke patients, motor abilities in children with cerebral palsy, aerobic capacity in patients with cardiovascular diseases, exercise capacity in patients with chronic obstructive pulmonary disease, as well as cognitive function, musculoskeletal function, balance, and pain (e.g., knee osteoarthritis, fibromyalgia) across various populations (SMD/MD were statistically significant). Additionally, VR interventions enhanced the quality of life in some populations. However, VR did not show significant associations with improvements in activities of daily living in dementia patients, long-term (3-6 months) pain and functional improvements in patients with chronic low back pain, or certain balance functions in frail older adults. The evidence exhibited high heterogeneity, and the GRADE quality of evidence for most outcomes was rated as "low" or "very low."

Conclusions:

VR interventions hold broad application prospects in the health field, but their effects are not universal and must adhere to the principle of "precision adaptation." The current evidence system is limited by factors such as high heterogeneity and uneven methodological quality. Future research should focus on standardizing intervention protocols, validating long-term effects, and exploring underlying mechanisms to enhance evidence quality and promote scientific clinical translation.