Home-Based Virtual Reality for Upper Limb Stroke Rehabilitation: Supervised vs. Unsupervised Approaches – A Systematic Review with Meta-Analysis of Unsupervised Interventions

This systematic review and meta-analysis aims to evaluate the effectiveness of **unsupervised home-based virtual reality (VR)** interventions compared to **active home exercise programs** on upper limb functional recovery in adult stroke survivors (≥18 years).

**Purpose**:
Upper limb impairment affects 55–85% of stroke survivors and significantly limits activities of daily living. Home-based VR has been promoted as a cost-effective, accessible, and motivating rehabilitation tool to increase therapy dose post-discharge. However, existing evidence is conflicting, particularly when VR is compared to dose-matched active controls without therapist supervision. This review addresses a critical gap by focusing exclusively on **unsupervised** VR interventions to determine whether the technology itself confers additional benefit beyond structured exercise.

**Methods**:
A comprehensive search was conducted in PubMed, Scopus, and Web of Science from inception to October 2025. Only randomized controlled trials (RCTs) involving home-based VR (with or without exergaming), adult stroke survivors, and quantitative upper limb outcomes (e.g., Fugl-Meyer Assessment–Upper Extremity, Action Research Arm Test) were included. Two independent reviewers performed study selection, data extraction, and risk of bias assessment using the Cochrane RoB 2 tool. A random-effects meta-analysis was conducted on homogeneous studies (n=2) using standardized mean difference (SMD). Heterogeneity was assessed with I², and certainty of evidence was rated using GRADE. A narrative synthesis was performed for clinically heterogeneous studies.

**Results (Completed Review)**:
Seven RCTs (n=414 participants) were included. The meta-analysis of two large, homogeneous trials (n=291) showed no significant difference between unsupervised home-based VR and active home exercise (SMD = -0.06; 95% CI: -0.29 to 0.17; P = 0.62; I² = 0%). Narrative synthesis revealed positive effects in studies incorporating remote therapist supervision (telerehabilitation), but not in unsupervised settings. Risk of bias was high across all studies due to lack of blinding. GRADE certainty was rated **Very Low**.

**Expected/Observed Outcomes**:
- No superiority of unsupervised home-based VR over active home exercise.
- Therapist supervision appears to be a key determinant of efficacy, not the VR technology per se.
- Future high-quality RCTs should compare supervised vs. unsupervised models and assess long-term adherence and cost-effectiveness.

**Registration Note**:
This protocol was registered **retrospectively** in OSF Registries due to initial oversight as a novice researcher. A detailed a priori protocol was followed throughout to minimize bias. The full review has been completed and is under peer review.

**Keywords**: stroke, virtual reality, home-based rehabilitation, upper extremity, systematic review, meta-analysis