Skip to content

Dance shows cognitive promise, but the review evidence is weak

An umbrella review favoured dance on cognitive tests in older adults with impairment, while finding serious limitations in the underlying reviews.

By 100HP editorialAbstract-based explanation checked

Based on the published abstract. The full paper may contain additional methods, results and limitations.

The 30-second takeaway

Dance interventions were associated with better cognitive test performance than control conditions in older adults with cognitive impairment. However, all included reviews were rated low or critically low quality, and relatively few focused on dementia. The findings suggest a possible cognitive benefit, without establishing reliable effects across these different patient groups.

Keep in mind

The evidence quality was weak, and results on the MoCA test varied substantially across analyses. The abstract does not report intervention lengths or show whether any cognitive gains lasted after dancing stopped.

THE NUMBERS, WITH CONTEXT

What researchers found

SMD 0.61 (95% CI 0.30 to 0.91)

MoCA global cognition score

Dance versus no intervention or other interventions; follow-up lengths were not reported. The effect uses standardized units, not raw test points.

BMC Geriatr, 2025 · Original source ↓

A review of reviews

This was an analysis of existing reviews, rather than a new dance trial. It brought together earlier research syntheses on dance for older adults with cognitive impairment, including comparisons with no intervention or other interventions.

What the score means

The effect is expressed in standardized units, rather than points on the cognitive test. Higher scores favoured dance, but this result does not establish slower dementia progression or improved daily independence.

CHECK THE ORIGINAL

The original publication

Effects of dance interventions on brain health for older adults with cognitive impairment: an umbrella review.

Prates RACS, Kaczmarska M, Senger D et al.
BMC Geriatr · 2025

PubMed ID
41194042
Record checked

AI-assisted research and writing. This explains one selected publication; it is not a complete review of everything known. Our approach.