A September 12 Bayesian network meta-analysis compared 11 categories of non-pharmacological interventions across 50 randomized trials involving 2,856 people with mild-to-moderate Alzheimer’s disease.
What the analysis found Dual-task training ranked highest for MMSE and MoCA outcomes, with tDCS also ranking favorably. For ADAS-Cog, rTMS ranked highest and was the only intervention reported as significantly better than usual care.
How to read the result A ranking probability is not the same as proof that one approach is best for every patient. Network meta-analysis combines direct and indirect comparisons, and results can depend on the quality, size, protocols and outcome measures of the included trials. The checked abstract does not provide the individual rTMS protocols, durability, adverse-event synthesis or detailed certainty ratings. The authors describe the rTMS finding as requiring cautious interpretation.
Why it matters The study helps define which comparisons and cognitive measures deserve closer examination, while supporting shared decisions that account for a person’s diagnosis, goals and medical care.
Source and disclosure Peer-reviewed systematic review in the Journal of Alzheimer’s Disease; authoritative abstract checked. Funding and conflicts were unavailable in the reviewed record. Prepared with AI assistance and source-checked September 15, 2026.