Dementia is a major neurodegenerative condition with limited treatment options, making non-pharmacological interventions increasingly important. Reminiscence therapy (RT) has shown benefits in dementia care, and its digital formats - delivered via virtual reality, mobile applications, or multimedia - may offer greater personalization, accessibility, and flexibility in clinical practice. However, evidence of their effectiveness remains fragmented and requires systematic synthesis. We conducted a systematic review and meta-analysis to evaluate the efficacy of digital RT in dementia and mild cognitive impairment (MCI), with literature searches conducted up to 1 May 2025. Eligible studies included randomized controlled trials (RCTs) and one-arm pre-post studies. Outcomes assessed were cognition, mood, behavioral and psychological symptoms of dementia (BPSD), quality of life (QOL), communication and engagement. RCTs were pooled in meta-analyses, while pre-post studies were narratively synthesized. Meta-analysis of RCTs suggested a possible improvement in BPSD, although the evidence was based on few studies of limited size and mixed comparators. Personalized and socially engaging formats appeared most associated with favorable outcomes. No significant effects were found for cognition, mood, QOL or communication and engagement, and overall results remained inconclusive due to variability in intervention design, delivery methods, and outcome measures. Narrative synthesis of pre-post studies revealed mixed results across domains, reflecting small samples, diverse digital formats and lack of control groups. Digital RT may provide limited benefits for managing BPSD, particularly when incorporating personalized and socially engaging elements. However, the overall evidence remains inconclusive and inconsistent across other outcome domains. Larger, multicenter randomized controlled trials using standardized protocols, validated outcome measures, and direct comparisons with both traditional RT and usual care are needed to clarify its therapeutic value and identify the most effective digital formats and target populations. The review protocol number was registered on PROSPERO (CRD420251039295).
Tran et al. (Wed,) studied this question.