Skip to content

B vitamin review found a very small cognitive benefit after study exclusions

The pooled estimate became much smaller after excluding outlier and single-blind studies, making the analysis choices central to interpreting the result.

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

The review found better global cognitive scores with B vitamin supplementation, but the estimated benefit depended strongly on which studies were included. Removing statistical outliers and single-blind studies left a very small effect that the authors rated as high certainty. This is a finding about cognitive test performance, not direct evidence of dementia prevention.

Keep in mind

The full pooled analysis had very low certainty and substantial variation between studies. The higher-certainty estimate excludes selected studies, and the abstract does not establish an improvement in everyday functioning.

THE NUMBERS, WITH CONTEXT

What researchers found

Hedges' g = 0.110 (95% CI 0.034–0.186)

Global cognitive function after study exclusions

B vitamins versus placebo or usual dementia care after at least 26 weeks; standardized effect after removing statistical outliers and single-blind studies.

Nutr Rev, 2025 · Original source ↓

Why the estimate changed

The initial pooled estimate was larger, but statistical outliers and single-blind studies contributed to both the estimated benefit and variation between trials. Excluding those studies substantially reduced that variation and the estimated effect.

What the score means

The reported effect is standardized across cognitive measures, so it is not a percentage improvement in memory. The subgroup analysis did not detect different effects across cognitive-status categories, which does not establish identical benefits.

CHECK THE ORIGINAL

The original publication

Efficacy of B Vitamin Supplementation on Global Cognitive Function in Older Adults: A Systematic Review and Meta-analysis.

Berg J, Grant R, Siervo M et al.
Nutr Rev · 2025

PubMed ID
40966571
Record checked

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