Immersive virtual reality for older adults with mild cognitive impairment, dementia, or cognitive frailty: a systematic review and narrative synthesis (2019–2025)

Background Immersive virtual reality (VR) is increasingly used to support cognition, mobility, and emotional well‑being in older adults with mild cognitive impairment (MCI), dementia, or frailty. Evidence is expanding but fragmented across small randomized and feasibility trials. We synthesised recent studies to clarify benefits, risks, and implementation considerations. Methods Following PRISMA 2020, we searched PubMed and CINAHL from 1 January 2019 to 15 October 2025. Eligible studies enrolled adults aged ≥ 65 years with MCI, dementia, or frailty/cognitive frailty; delivered immersive or semi‑immersive VR via head‑mounted display or large‑screen projection (interactive tasks or 360° content); and reported cognitive, mobility, or emotional/behavioural outcomes in randomized, quasi‑experimental, or pre–post designs. Two reviewers independently screened and extracted data. Risk of bias was appraised with RoB 2 (randomized trials) or JBI tools (non‑randomized). Heterogeneity precluded meta‑analysis; we conducted a structured narrative synthesis. Results Seventy records were identified (PubMed 28; CINAHL 42); after removing 9 duplicates, 61 records were screened, 24 full texts were assessed, and 13 studies were included (10 randomized; 3 feasibility/mixed‑methods). The most consistent improvements occurred in executive function and processing speed among participants with MCI or cognitive frailty; several trials also reported modest gains in global cognition. Multiple trials showed better Timed Up & Go and Berg Balance outcomes and enhanced anticipatory postural adjustments versus comparators. In residential care, immersive reminiscence and group VR reduced anxiety and apathy and were well tolerated. Adverse events were uncommon and mild; adherence was high with supervised delivery. Most randomized trials had some concerns for bias; one was at overall low risk. Conclusions Immersive VR appears feasible and potentially beneficial for mentation and mobility in older adults with MCI or cognitive frailty and may reduce anxiety and apathy in institutional settings. Programmes with adequate exposure (2–3 sessions/week for 8–12 weeks; ≥15 total hours), adaptive challenge, and supervision show the most consistent benefits. Larger multicentre randomized trials with standardised outcomes and embedded implementation and economic evaluations are needed.