Does Iron Make Children Smarter? What the Evidence Says About the World's Most-Supplemented Nutrient

The question and why it matters

Iron deficiency is the most common nutritional deficiency in the world. Global burden estimates put anemia — most of it iron-related in young children — at roughly half of all children under five and a quarter of pregnant women, with the heaviest burdens in South Asia and sub-Saharan Africa. Public-health institutions treat this as an emergency, and the main tools are old and cheap: iron drops and syrups for infants, iron and folic acid tablets for schoolchildren and pregnant women, iron-fortified staple foods, double-fortified salt, micronutrient powders stirred into home food, and iron-fortified infant formula in richer settings. The World Health Organization recommends daily iron supplementation for children aged 6 to 23 months in countries where anemia prevalence in that group is 40% or higher, on the strength of the hematologic evidence.

Hematology is the easy part. Iron cures anemia, and no serious review disagrees. The harder claim — the one that carries the policy argument, drives cost-effectiveness models, and animates the headlines — is that the cure reaches the brain: that supplementing children makes them think better, learn more, and earn more. This brief asks exactly that question across the age windows and delivery modes where the evidence exists: treatment of diagnosed deficiency, universal supplementation, and population-scale fortification, from gestation through school age. The answer matters well beyond any single program. Hundreds of millions of children receive these interventions every year, several major international charities build their case on the cognitive channel, and the same policy instinct — give iron broadly, screen rarely — is now spreading to iron-replete populations where the risk-benefit balance may differ.

The question is really several questions

Three distinctions do most of the work in this literature, and collapsing them is the main source of confused headlines.

Who is being given iron. Treating a child who is genuinely iron deficient or anemic is not the same intervention as giving iron to a child whose iron stores are fine. The first restores a missing substrate to a brain that needed it; the second adds a redox-active metal that the body normally regulates tightly. Several meta-analyses find the benefit entirely in the first group, and at least one finds harm signals in the second.

When. The brain's demand for iron peaks during rapid growth — the third trimester of gestation, and the first two years of life. A supplement given at age nine is operating on a different organ than the same supplement given at nine months. Trials of treating anemia in toddlers have repeatedly failed to move development scores, which several reviewers read as a closing-window effect: the deficiency may have already done its work.

How the iron arrives. A directly observed daily tablet in a trial is not the same as salt sold in a village shop or flour milled in a factory. Field experiments in India have shown that population fortification can move hemoglobin and still fail to move any cognitive outcome, because take-up is partial and dose per serving is low. Efficacy under supervised consumption and effectiveness under real-world delivery are different claims, and the fortification literature is where the gap is widest.

Any honest answer has to be indexed by these three coordinates. "Does iron make children smarter?" without them has no single answer; it has a matrix of answers, and the matrix is what the evidence actually shows.

Best current answer

Correcting real deficiency in school-age children produces small gains on laboratory cognitive tests. Five overlapping meta-analyses of randomized trials, plus a 2026 umbrella review of 17 systematic reviews, converge on standardized mean differences of roughly 0.2 to 0.5 for general cognitive or intelligence scores among anemic or iron-deficient school-age children — on the order of two to five IQ points — with the clearest effects on attention-related measures and no consistent effect on motor development.

Iron-replete children get nothing measurable. Every meta-analysis that stratifies by baseline iron status finds the benefit concentrated in the anemic or iron-deficient subgroup. In Low and colleagues' 2013 Cochrane-grade review, the anemic subgroup improved while the non-anemic subgroup estimate was a flat zero with tight confidence intervals. A 2025 meta-analysis targeting non-anemic young people specifically found that the apparent benefit on intelligence scores vanished once iron-deficient participants were excluded.

School achievement never moves. Across every age group and every delivery mode, the most practically important outcome — whether children perform better on reading, mathematics, or school-administered tests — is consistently null. This is the most important negative finding in the literature and the one least reflected in public discussion.

In infancy, the only randomized evidence with long follow-up points the other way. A Chilean trial run in the early 1990s assigned non-anemic six-month-olds to iron-fortified formula at the then-standard 12.7 mg/L or to low-iron formula at 2.3 mg/L for six months. At ages 10, 16, and 21, the iron-fortified group scored modestly worse on spatial memory, visual-motor integration, arithmetic, reading comprehension, and later educational attainment — with the harm concentrated among children whose hemoglobin had been highest at six months, and apparent benefit among those whose hemoglobin had been lowest. The effect sizes are small (fractions of a standard deviation), but the direction was consistent across three decades of follow-up in the same randomized cohort.

At population scale, blood counts improve and test scores do not. Multiple fortification experiments — double-fortified salt in rural Bihar reaching whole villages, iron-fortified school lunches in Bihar followed for four years — have produced real reductions in anemia and no detectable effect on cognitive or educational outcomes.

What iron does, and why timing could matter

The biological case for iron and the brain is strong and mechanistic, which is precisely why the mixed trial evidence is interesting rather than surprising. Iron is a cofactor for myelination, for the synthesis of dopamine and other neurotransmitters, and for the energy-metabolism enzymes that developing neurons depend on. The brain accumulates iron rapidly during late gestation and the first two years, and primate and rodent work summarized in the older reviews shows that severe deficiency during this window alters brain iron content, dopamine system function, and myelination in ways that only partly reverse with later correction. The clinical correlate is Sachdev and colleagues' 2005 finding that supplementation moved intelligence scores in children over seven but showed no convincing effect on Bayley scales in children under 27 months.

The mirror image is the harder story: iron is redox-active, and free iron catalyzes the formation of reactive oxygen species. In a developing system, more iron than needed is not automatically better. The Chilean trial is the human signature of exactly this U-shaped response — supplementation helped the infants with the lowest hemoglobin and hurt those with the highest, on the same randomized assignment and the same doses. A 2025 clinical review reaches the uncomfortable conclusion that universal prophylactic iron supplementation in infants is not supported by current evidence, and that whether non-anemic iron deficiency needs treatment at all remains genuinely uncertain.

Mechanism, of course, is not identification. The biological plausibility of both benefit and harm means the mechanism cannot arbitrate between the positive school-age meta-analyses and the null or adverse field results; only the designs can, and they are weaker than the biology.

The school-age evidence in detail

The quantitative case rests on a cluster of meta-analyses that largely pool the same trials, mostly conducted in South and Southeast Asia from the late 1980s onward. Sachdev, Gera, and Nestel's 2005 review found a pooled mental-development effect of SMD 0.30, an IQ effect of 0.41 among children aged seven and over, no effect below 27 months, and no effect on motor development. Low and colleagues' 2013 review of 32 trials in primary-school-aged children found a global cognitive SMD of 0.50 with 93% heterogeneity — but the anemic subgroup effect was 0.29 and the non-anemic effect was 0.01; IQ improved only among anemic children, by a mean of 4.5 points on an unstable confidence interval; and in the four trials at low risk of bias, even the hematologic benefits were no longer statistically detectable, a sobering detail about the review's own evidence base.

Chen and colleagues' 2022 review restricted to low- and middle-income settings found an intelligence effect of SMD 0.47 with a dose-response relationship, but nothing on attention, memory, or school performance. Gutema and colleagues' 2023 review found a similar intelligence effect — but stratified by study quality, the benefit appeared only in the high-risk-of-bias studies; the low-and-moderate-risk subgroup estimate was half the size and statistically null. Theirs is the most diagnostic analysis in the set: it suggests the headline effect is partly an artefact of weaker trials.

Two newer results sharpen the picture further. A 2024 randomized trial in southern Ethiopia, where iron deficiency was uncommon, found that intermittent iron produced no cognitive benefit and a small decrement on a visual-search task, while vitamin A — not iron — improved digit span. And the 2026 umbrella review, appraising 17 systematic reviews with AMSTAR-2, concluded that small benefits on intelligence, memory, and attention hold specifically in anemic children; effects in non-deficient populations are negligible or uncertain; and no consistent improvement appears in mental development or school achievement. That review's bottom line — target the deficient, do not supplement the healthy — is also what a plain reading of the stratified meta-analyses would have produced a decade earlier.

Infancy: the trial that should give universal programs pause

The Santiago Chile trial remains unique: the only randomized comparison of standard-level versus low iron in non-anemic infants ever followed into adulthood. Randomization was double-blind, formula was provided and quantified by weekly home visits, and attrition across 25 years has been remarkably low. At 10 years, the iron-fortified group scored lower on spatial memory and visual-motor integration, with suggestive deficits in IQ and arithmetic; at 16 years, lower on eight of nine tests, significantly so for visual memory, arithmetic, and reading comprehension; at 21 years, lower on neurocognitive, emotional-regulation, and educational-attainment outcomes, with dose-response relationships for formula volume consumed. A companion analysis comparing the supplemented children against a never-supplemented arm found small but significant decrements in visual-motor integration, quantitative reasoning, and task errors, with the same baseline-hemoglobin interaction: harm concentrated in infants whose six-month hemoglobin exceeded roughly 125 g/L.

Three qualifications keep this from being a simple verdict. The trial was run in 1991–1994, in a population that was no longer badly iron-deficient; the comparison is between two doses, not iron versus nothing; and no other trial has attempted the same comparison with long follow-up, so the finding is one randomized lineage, however well-instrumented. The counterargument from the same data is the 12% to 35% iron-deficiency rate at 12 months in the low-iron arm — real deficiency was traded for real protection against it. But the cognitive ledger, unusually, came out negative on net for the fortified arm. Any program contemplating universal iron for iron-sufficient infants is making a bet against that ledger.

The prenatal window

The strongest long-run positive signal in the entire literature does not come from child supplementation at all. In rural Nepal, children whose mothers received iron and folic acid during pregnancy scored about 2.4 points higher on a nonverbal intelligence test at ages seven to nine, with better inhibitory control, working memory, and fine motor speed, against a vitamin-A-only control — one of the few nutrition follow-ups to survive its own multiplicity adjustments. But the intervention was iron plus folic acid, the effect was modest, and the design cannot separate the two nutrients. The follow-up of a separate Nepali trial comparing multiple micronutrients against iron-folic-acid alone found no additional cognitive advantage of the richer package at age 12, which is consistent with iron being the active ingredient but does not prove it. Preschool iron given after prenatal exposure added nothing in the same cohort — again the window story.

The honest reading is that the prenatal period is where iron intervention looks most promising for lasting cognition, based on one well-conducted trial with a bundled treatment; it is not yet a settled causal result, and several other antenatal micronutrient follow-ups have been null.

Fortification in the field: blood counts improve, test scores don't

Two field experiments bracket the delivery problem. Banerjee, Barnhardt, and Duflo's Bihar double-fortified-salt experiment randomized 400 villages and found, after two years, no economically meaningful effect of selling or giving away fortified salt on hemoglobin, anemia, cognition or mental health — with adolescent instrumental-variable estimates showing a small hematologic response and still no cognitive signal. Von Grafenstein and colleagues' 2023 school-lunch trial in Bihar, by contrast, achieved real biological delivery: four years of iron-fortified salt in school meals raised hemoglobin by 0.245 g/dL and cut anemia by 13.4 percentage points relative to a late-start control group — and moved no cognitive or educational outcome at all. When delivery works and blood counts rise for years, test scores still do not follow.

One counterexample deserves its own weight. Wenger and colleagues' randomized feeding study of iron-biofortified pearl millet in Indian adolescents — six months, measured consumption, iron-deficient participants — improved ferritin, behavioral cognition measures on ten of twenty-one outcomes, and event-related-potential markers of brain function, with the cognitive change statistically mediated by the brain electrophysiology change. That is the mechanism chain, measured end to end, in humans. It is also an efficacy trial under controlled consumption in a deficient population: the exact conditions under which the meta-analyses predict benefit, and the exact opposite of what village salt distribution provides.

Competing explanations for the positive school-age meta-analyses

The pooled school-age effects are consistent with a real, modest, deficiency-specific benefit — but at least four other explanations carry nontrivial weight, and the data cannot fully separate them.

First, small-study and publication bias: the funnel plots in a literature of fewer than ten contributing trials per outcome have little power, and Gutema's risk-of-bias stratification shows the effect shrinking to nullity exactly where trials are stronger. Second, co-administered nutrients: several influential trials delivered iron with folic acid or vitamin C, and the umbrella review notes that the intelligence effect weakens when iron-alone trials are isolated. Third, measurement: the "global cognitive" endpoint is a different instrument in almost every trial, and I² values near 90% indicate the pooled estimate averages over genuinely different outcomes. Fourth, regression to the mean and expectancy: anemic children are the subgroup with most room to move, and trials from this era rarely blinded outcome assessors to allocation against a hemoglobin-drawn attention effect. None of these explanations requires discarding the biology; all of them require shrinking the headline effect toward the low end of the 0.2–0.5 range.

Evidence quality and coverage limits

The randomized base is old where it is largest: the trials underlying most meta-analytic weight were run in India, Thailand, and Indonesia in the 1980s and 1990s, with sample sizes in the dozens to hundreds, allocation concealment often unclear, and cognitive batteries that would not pass today's measurement standards. Heterogeneity is extreme — I² values of 85–98% are typical — so pooled means should be read as centres of distributions, not as the effect any given program should expect. The 2-to-5-year age band is nearly evidence-free for cognition: the dedicated 2013 review found only one trial reporting cognitive outcomes, and none at low risk of bias. Under-threes with diagnosed deficiency have eight small trials behind the Cochrane conclusion, two of which conflict on longer-term follow-up. And the cognitive follow-ups of the Chile trial, though randomized, analyze one cohort in one country, with the subgroup interactions (baseline hemoglobin, formula volume) coming from secondary analyses.

What the literature measures well is hematology; what it measures poorly is everything downstream. If the iron-cognition story has a decisive gap, it is the absence of any large, modern, individually randomized, iron-status-stratified trial with a standardized cognitive battery and school-administered achievement outcomes at multiple years — the trial that would let the meta-analyses retire their 1990s-era inputs.

Counterevidence

Against the view that iron supplementation clearly helps children's cognition: the consistently null school-achievement results across all syntheses; the risk-of-bias stratification showing the intelligence effect concentrated in weaker trials; the Ethiopian trial's null and slightly adverse results in a low-prevalence setting; the 2025 non-anemic meta-analysis whose effect disappears without iron-deficient participants; and two four-year field experiments that moved anemia without moving any test score.

Against the view that iron supplementation is broadly harmless: the Chilean follow-ups across 10, 16, and 21 years, the baseline-hemoglobin interaction, the dose-response on formula volume, and the 2025 review's conclusion that universal infant prophylaxis lacks support. Against a strong reading of the Chile harm, however, stands the absence of any replication attempt, the low-iron arm's own deficiency burden, the era and population specificity of the trial, and the enormous hematologic benefit of preventing iron-deficiency anemia in genuinely deficient infants — a benefit this brief does not question.

Calibrated confidence and reversal conditions

• ~95%: iron supplementation and fortification reduce anemia and iron deficiency in children, when actually consumed. The hematologic evidence is as settled as nutrition trials get.
• ~80%: correcting iron deficiency in anemic or iron-deficient school-age children yields a small improvement (about 0.2–0.5 SD) on laboratory cognitive tests. The sign is robust across syntheses; the magnitude is not.
• ~85%: iron-replete school-age children receive no measurable cognitive benefit, and no iron intervention tested in any age group improves school achievement on average.
• ~65%: standard-level iron fortification in infancy carries a small risk of lasting cognitive harm for iron-sufficient children, concentrated in those with high baseline hemoglobin. One randomized lineage, three decades of consistent follow-up, mechanistic plausibility, no replication.
• ~55%: prenatal iron-folic-acid supplementation in deficient populations produces small long-run cognitive benefits in offspring. One clean trial with a bundled intervention; several null antenatal follow-ups of different packages.
• ~40%: any population-scale staple fortification program — salt, flour, school meals — improves children's cognition at scale under real delivery conditions. Every direct test to date has failed on that outcome.

The school-age benefit estimate would fall sharply if an individually-randomized, iron-status-stratified trial with modern batteries found nothing; it would rise if a preregistered multi-site trial in confirmed-deficient children found effects at the top of the historical range. The infant-harm estimate would collapse if a second randomized comparison of standard versus low iron in iron-sufficient infants, with long follow-up, found no difference — a trial that has not been attempted since Chile. The fortification estimate would move if a trial delivered fortificant with verified individual uptake (biofortified staples, supervised meals) and still found null cognition; the Indian biofortification work is the closest existing test and went the other way.

The provisional bottom line

Iron is not a cognitive enhancer. It is a substrate whose absence damages the developing brain and whose restoration, if done before the damage consolidates, buys back a little of what was lost — small gains on laboratory tests in children who were genuinely deficient, and no measurable gain in children who were not. The programs that scale best are the ones with the weakest cognitive returns, because population delivery cannot easily find and treat the deficient without also reaching the replete — for whom the best available randomized evidence, limited as it is, suggests the correct dose may be zero. That tension, between a real biological need and an imprecise instrument, is the honest state of the field.

References

1. World Health Organization. Guideline: Daily iron supplementation in infants and children. Geneva: WHO; 2016. ISBN 9789241549523. https://www.who.int/publications/i/item/9789241549523
2. Sachdev HS, Gera T, Nestel P. Effect of iron supplementation on mental and motor development in children: systematic review of randomised controlled trials. Public Health Nutrition. 2005;8(5):502-514. doi:10.1079/phn2004677
3. Low MSY, Farrell A, Biggs BA, Pasricha SR. Effects of daily iron supplementation in primary-school-aged children: systematic review and meta-analysis of randomized controlled trials. CMAJ. 2013;185(17):E791-E802. doi:10.1503/cmaj.130628
4. Thompson J, Biggs BA, Pasricha SR. Effects of daily iron supplementation in 2- to 5-year-old children: systematic review and meta-analysis. Pediatrics. 2013;131(4):739-753. doi:10.1542/peds.2012-2256
5. Pasricha SR, Hayes E, Kalumba K, Biggs BA. Effect of daily iron supplementation on health in children aged 4-23 months: a systematic review and meta-analysis of randomised controlled trials. Lancet Global Health. 2013;1(2):e77-e86. doi:10.1016/S2214-109X(13)70046-9
6. Wang M. Iron therapy for improving psychomotor development and cognitive function in children under the age of three with iron deficiency anaemia. Cochrane Database of Systematic Reviews. 2013;(2):CD001444. doi:10.1002/14651858.CD001444.pub2
7. Falkingham M, Abdelhamid A, Curtis P, Fairweather-Tait S, Dye L. The effects of oral iron supplementation on cognition in older children and adults: a systematic review and meta-analysis. Nutrition Journal. 2010;9:4. doi:10.1186/1475-2891-9-4
8. Chen Z, Yang H, Wang D, Sudfeld CR, Zhao A. Effect of oral iron supplementation on cognitive function among children and adolescents in low- and middle-income countries: a systematic review and meta-analysis. Nutrients. 2022;14(24):5332. doi:10.3390/nu14245332
9. Gutema BT, Sorrie MB, Degefa N, et al. Effects of iron supplementation on cognitive development in school-age children: systematic review and meta-analysis. PLoS ONE. 2023;18(6):e0287703. doi:10.1371/journal.pone.0287703
10. Caballero-Apaza L, Arbildo-Vega HI, Mamani-Cori V, et al. Effect of iron supplements on cognitive development in children: an umbrella review. Frontiers in Nutrition. 2026;13:1718507. doi:10.3389/fnut.2026.1718507
11. Fiani D, Chahine S, Zaboube M, Solmi M, Powers JM, Calarge C. Psychiatric and cognitive outcomes of iron supplementation in non-anemic children, adolescents, and menstruating adults: a meta-analysis and systematic review. Neuroscience & Biobehavioral Reviews. 2025;176:106372. doi:10.1016/j.neubiorev.2025.106372
12. Chmielewska A, Domellöf M. Iron deficiency in infants and children — the current research challenges. Current Opinion in Clinical Nutrition & Metabolic Care. 2025;28(3):270-276. doi:10.1097/mco.0000000000001120
13. Gahagan S, Delker E, Blanco E, Burrows R, Lozoff B. Randomized controlled trial of iron-fortified versus low-iron infant formula: developmental outcomes at 16 years. Journal of Pediatrics. 2019;212:124-130. doi:10.1016/j.jpeds.2019.05.030
14. Delker E, Blanco E, Burrows R, Lozoff B. Young adult outcomes associated with lower cognitive functioning in childhood related to iron-fortified formula in infancy. Nutritional Neuroscience. 2021;24(10):771-780. doi:10.1080/1028415X.2020.1804099
15. East PL, Reid B, Blanco E, Burrows R, Lozoff B. Iron supplementation given to nonanemic infants: neurocognitive functioning at 16 years. Nutritional Neuroscience. 2022;25(10):2012-2021. doi:10.1080/1028415X.2021.2013399
16. Christian P, Murray-Kolb LE, Khatry SK, et al. Prenatal micronutrient supplementation and intellectual and motor function in early school-aged children in Nepal. JAMA. 2010;304(24):2716-2723. doi:10.1001/jama.2010.1861
17. Dulal S, Liégeois F, Osrin D, et al. Does antenatal micronutrient supplementation improve children's cognitive function? Evidence from the follow-up of a double-blind randomised controlled trial in Nepal. BMJ Global Health. 2018;3(1):e000527. doi:10.1136/bmjgh-2017-000527
18. Banerjee A, Barnhardt S, Duflo E. Can iron-fortified salt control anemia? Evidence from two experiments in rural Bihar. NBER Working Paper 22121. Cambridge, MA: National Bureau of Economic Research; 2016. https://www.nber.org/papers/w22121
19. von Grafenstein L, Kumar A, Kumar S, Vollmer S. Medium-run impacts of iron-fortified school lunch on anaemia, cognition, and learning outcomes in India. Obesity Reviews. 2023;24(7):e12659. doi:10.1111/obes.12559
20. Wenger MJ, Murray-Kolb LE, Scott SR, Boy E. Modeling relationships between iron status, behavior, and brain electrophysiology: evidence from a randomized study involving a biofortified grain in Indian adolescents. BMC Public Health. 2022;22:1347. doi:10.1186/s12889-022-13612-z
21. Gutema BT, Levecke B, Sorrie MB, Megersa ND, Zewdie TH. Effectiveness of intermittent iron and high-dose vitamin A supplementation on cognitive development of school children in southern Ethiopia: a randomized placebo-controlled trial. American Journal of Clinical Nutrition. 2024;119(2):470-484. doi:10.1016/j.ajcnut.2023.11.005
22. World Health Organization. Guideline on haemoglobin cutoffs to define anaemia in individuals and populations. Geneva: WHO; 2021. https://www.who.int/publications/i/item/9789240030923