Multiple micronutrient supplementation for improving cognitive performance in children: systematic review of randomized controlled trials

20 RCTs (3 in children under 5, 17 in children aged 5-16); domain-level pools rest on 4-12 trials

Healthy children aged 0-18. A large share of the included trials were run in low- and middle-income countries where baseline micronutrient status was poor.

Systematic review restricted to randomized placebo-controlled trials of three or more micronutrients versus placebo, published 1970-2008, with independent double data extraction. Cognitive tests were grouped into domains (fluid intelligence, crystallized intelligence, academic performance and others) and pooled separately - exactly the outcome-class separation this database requires. Heterogeneity explored with sensitivity and meta-regression.

In children aged 5-16, multiple-micronutrient supplementation produced 0.14 SD on fluid intelligence (95% CI -0.02 to 0.29, P = 0.083 - NOT significant, 12 trials), -0.03 SD on crystallized intelligence (95% CI -0.21 to 0.15, P = 0.74, 12 trials), and 0.30 SD on academic performance (95% CI 0.01 to 0.58, P = 0.044) - but the academic estimate rests on only FOUR trials and its confidence interval nearly touches zero. No significant effects in any other cognitive domain. The authors conclude that more research is required before any public health recommendation.

None internally - placebo-controlled randomization. The transportability problem is the real issue - the pooled trials are disproportionately from deficient populations, so the estimate does not describe what supplementation would do to a well-nourished child.