Baker · Alzheimer's & dementia : the journal of the Alzheimer's Association 2023 · Randomized 2x2 factorial placebo-controlled trial · n=2262

Effects of cocoa extract and a multivitamin on cognitive function: A randomized clinical trial.

Level 2 - randomized trial

Individual randomized controlled trial

PubMed 36102337 · doi:10.1002/alz.12767 · record verified 2026-08-26

What was done

The COSMOS-Mind trial was a 3-year, 2-by-2 factorial randomized controlled trial testing daily cocoa extract (containing 500 mg/day flavanols) versus placebo and a daily commercial multivitamin-mineral (MVM) versus placebo. Cognition was evaluated by telephone at baseline and annually in 2,262 older adults (mean age 73 years; 60% women; 89% non-Hispanic White; 92% completed baseline and at least one follow-up). The primary endpoint was change in a global cognition composite z-score with 3 years of cocoa extract; the pre-specified secondary endpoint was change in global cognition with MVM. Memory and executive function composites were also evaluated.

What was found

Cocoa extract had no significant effect on global cognition (mean z-score = 0.03, 95% CI: -0.02 to 0.08; P = .28). Daily MVM supplementation significantly improved global cognition compared to placebo (mean z = 0.07, 95% CI: 0.02 to 0.12; P = .007), with observed benefits in episodic memory and executive function. The MVM benefit was larger in participants with a history of cardiovascular disease (z = 0.14, 95% CI: -0.02 to 0.31 vs. no history z = 0.06, 95% CI: 0.01 to 0.11; nominal interaction P = .01). The interaction between cocoa extract and MVM was not statistically significant.

Why it matters

This trial provides large-scale randomized evidence that daily multivitamin-mineral supplementation can modestly improve cognitive function in older adults over 3 years, while finding no cognitive benefit from cocoa flavanols.

Limits

The study sample was 89% non-Hispanic White, limiting generalizability to more racially and ethnically diverse populations. Cognitive testing was pragmatic and administered via telephone rather than comprehensive in-person neuropsychological testing, and the trial was not designed to isolate specific micronutrient mechanisms within the multivitamin formula.