A new review of 21 studies says children with ADHD have lower iron stores and should be screened and given iron. Its own table has two rows of 630 children with identical numbers, and eight studies where the gap all but vanishes

The idea that iron might matter in ADHD is not a fringe one. Iron is needed by the enzyme that sets the pace of dopamine production, and dopamine is the brain chemical that the main ADHD medicines act on. In 2004 a small French study found much lower ferritin, the blood marker of stored iron, in children with ADHD than in other children. Ever since, researchers in more than a dozen countries have repeated the comparison, and they have not agreed with each other.

A team at Woldia University in Ethiopia has now pooled those comparisons. Their review, published in PLOS One on 5 October 2026, searched four databases up to June 2026 and kept 21 studies that measured ferritin in children with a formal ADHD diagnosis and in children without one: 4,058 with ADHD and 4,533 without. Pooled, ferritin was 9.74 ng/ml lower in the ADHD groups, with a confidence interval of 2.98 to 16.50. The authors conclude that screening and 'enhanced iron supplementation protocols' are warranted.

We read the table of the 21 studies line by line, and the average hides more than it shows. In eight of them the two groups are less than 3 ng/ml apart, which is nothing: in several the difference is about a tenth of the normal spread between one child and the next. In one American study the children with ADHD had the higher ferritin. Then there are a few very large gaps. A study of 25 children in India reports 6 against 49. One of 40 in Egypt reports 16 against 89. One of 25 in Indonesia reports 43 against 122. Averaging findings this different produces a number, but not one that describes any of them. The paper's own measure of disagreement between studies is 99%, about as high as the scale goes.

Two rows stopped us. A 2014 study listed under Qatar and a 2015 study listed under Turkey, both with a first author named Bener (spelled Berner in one row), each have 630 children with ADHD and 630 without, and the ferritin values match to the second decimal: 36.26 against 38.14, with the same spreads. We cannot tell from the paper whether these are two reports of the same children. If they are, 630 cases and 630 controls have been counted twice. The review does not comment on it.

The breakdowns the authors ran to explain the disagreement mostly weaken the headline. Split by region, the gap is not statistically clear in Africa, in Asia or in North America, where it is 1.4 ng/ml. In the studies with 200 or more children it is not clear either. In studies from before 2015 the gap is 1.9 ng/ml; in those from 2016 on it is 19.4, and the paper has no firm explanation for the jump. One study from China supplies 1,565 of the 4,058 children with ADHD and reports a gap of 35.

Even taking the pooled number as given, it cannot carry the conclusion placed on it. Every study here measured ferritin once, in children who already had the diagnosis. That cannot show which came first. The paper itself notes that ferritin moves with inflammation and with what a child has been eating, and that whether a blood test reflects iron in the brain is 'debatable'. None of the 21 studies gave anyone iron. The authors do ask for long-term studies to work out the direction of cause, in the same paragraph that recommends supplements.

Our reading: there may well be something to the iron and ADHD question, and the small treatment trials the paper mentions in its introduction are the kind of evidence that could answer it. This review is not. What it supports is modest and already standard: if a child eats poorly, is tired or pale, or has other reasons to suspect low iron, a doctor can check with a blood test, and a real deficiency is worth treating for its own sake. It does not support giving iron to a child because of an ADHD diagnosis. Iron is one of the supplements where more is not harmless, and tablets meant for adults are dangerous to small children.