Effects of Dance Interventions on Physical Function in Healthy Older Adults: A Systematic Review and 3-Level...

As global population aging accelerates, maintaining physical function in older adults has emerged as a core public health challenge. Dance, a multimodal physical activity integrating cognitive challenges and social interaction, has been proven effective in improving health in older age. However, existing evidence is largely confined to validating efficacy, often plagued by selective outcome reporting, and lacks evidence-based consensus regarding optimal intervention protocols.

This study aimed to (1) systematically evaluate the effects of dance interventions on physical function in healthy older adults, (2) provide in-depth insight into the moderating effects of participant characteristics and intervention prescription parameters (eg, session duration, intervention period, frequency) and potential dose-response relationships, and (3) offer evidence-based grounds and practical recommendations for public health policymakers and community health instructors in developing health promotion programs for older adults.

PubMed, Web of Science, and EBSCO databases were searched up to November 10, 2025. In accordance with PICO principles, randomized controlled trials and quasi-experimental studies were included if they involved healthy adults older than 60 years (P), implemented dance interventions (I), compared them against any control group (C), and reported outcomes within the International Classification of Functioning, Disability and Health (ICF) framework (O). Literature screening and data extraction were conducted independently by 2 reviewers. Methodological quality was assessed using the Cochrane RoB 2.0 tool and PEDro scale, and evidence certainty was evaluated via the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. A 3-level random-effects model was employed to handle the dependency of effect sizes, with Hedges' $g$ selected as the effect index, followed by robustness checks. In addition, moderator analyses and meta-regression were conducted to explore potential influencing factors, combined with funnel plots and a modified Egger's test to assess publication bias.

Twenty-four studies (1112 participants; 237 effect sizes) were ultimately included. These studies exhibited publication bias and low certainty of evidence. The 3-level meta-analysis indicated that dance interventions yielded significant medium-to-large improvements in physical function (Hedges' g = 0.635, P < .001), demonstrating effectiveness in both "body function" and "activities and participation" domains. Moderator analysis revealed that community-dwelling older adults derived significantly greater benefits compared with nursing home residents, and an 8-week intervention was sufficient to produce significant, cost-effective improvements. Meta-regression identified significant nonlinear relationships: (1) age followed an "inverted U-shaped" trend, with intervention effects peaking in the 70-75 age group; and (2) session duration peaked at approximately 20 minutes, with effect sizes declining beyond this threshold.

Dance serves as an effective nonpharmacological strategy for promoting healthy aging. Based on current evidence, public health policymakers should prioritize resource allocation toward the 70- to 75-year-old community-dwelling population. Regarding dance prescription design, we advocate for an 8-week baseline cycle to generate significant benefits and highlight the potential value of short-duration models (∼20 minutes) in enhancing adherence and improving function. However, given that this single-session duration falls below traditional exercise physiology recommendations, future research must further elucidate the optimal "dose-response" mechanism of dance; until its biological plausibility is established, generalizing this specific duration as a standard should be approached with caution.