Reviews/Iron Status/Untargeted iron raises infection risk
Func · IRON-SAFE-01 · v2
The question is whether giving iron to everyone, without first checking who needs it, makes children in areas where infections like malaria are common more likely to get sick. Some strong trials found exactly that: a large study in a high-malaria setting was stopped early because children given iron were more likely to be hospitalised or die [1], and a trial in the Peruvian Amazon found more malaria and diarrhea in older children given iron [8]. But other equally careful trials that also handed out bed nets or malaria-prevention medicine found no extra malaria at all, and in one case slightly less [3][10]. So the harm looks real, but mainly when there is no active effort to control malaria alongside the iron. The panel judged the plain, unconditional statement only partly supported because the corpus itself points both ways.
Two direct trials support harm: the halted Pemba trial [1], where children on iron and folic acid were 12% more likely to die or need hospital care (p=0.02), and the Peruvian Amazon trial [8], where iron raised P. vivax and diarrhea morbidity in children over five. Subgroup analyses of a Ghanaian trial add conditioned harm signals, with more malaria episodes in certain haemoglobinopathy and blood-group carriers on iron [4] and elevated severe-malaria risk in HLA-G variant carriers [5]. Against this, the Ghana MNP trial found malaria incidence significantly lower in the iron group (RR 0.87) when bed nets and treatment were provided [3], and the Malawi trial found neither iron arm increased malaria risk when given with chemoprevention [10]. The countervailing trials all supplied active malaria control, so the harm signal is present but tied to the surrounding conditions. The contested point: The corpus divides on whether the risk holds unconditionally: [1] and [8] show harm in settings without active malaria control, while [3] and [10] show no increased malaria precisely where bed nets or chemoprevention were provided. The dispute is therefore not the existence of a signal but whether the bare claim can stand without naming the absence of malaria control and iron-replete status as conditions.
All 10 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Iron Status, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
Co-intervention confounding: every trial finding no increased risk ([3], [10]) also supplied bed nets or malaria chemoprevention, so the claim is only true where malaria control is absent — a condition the bare statement omits.
Direct corpus contradiction: [3] (RR 0.87, lower malaria) and [10] (no increased risk) are large RCTs opposing the harm signal from [1] and [8]; the corpus cannot support an unconditional claim.
Where it remains uncertain. Two rigorous trials that added bed nets or chemoprevention found no increased malaria, with lower incidence in one [3] and no increased risk in the other [10].
Aleatory
The harm signals in [4] and [5] arise from many subgroup comparisons across haemoglobinopathies, blood groups and a genetic variant, where some positive findings are expected by chance.
Epistemic
Infection here is largely malaria; only [8] adds diarrhea, and [9] reports haematology with its infection safety results still underway, so bacterial-infection risk is not established.
Inferential
Only [1]'s interpretation localises harm to those not iron deficient; no trial randomised targeted versus untargeted allocation, so the tie to the word 'untargeted' is inferred rather than measured.
The load-bearing papers from a corpus of 10, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Large halted RCT: universal iron increased severe illness/hospitalisation in a high-malaria population — primary support.
“Routine supplementation with iron and folic acid in preschool children in a population with high rates of malaria can result in an increased risk of severe illness and death.”
Panel auditInternal validity: high · scope of inference: Children aged 1-35 months in a high malaria transmission setting (Pemba, Zanzibar) receiving universal iron plus folic acid.Key limitations: stopped early · deaths alone not significant (15%, -7 to 41, p=0.19) · combined iron+folic acid intervention, not iron isolated
Subgroup support: harm concentrated in specific haemoglobin/blood-group strata.
“subjects with HbAS and HbAC and blood groups A and O in the iron group had a significantly increased number of malaria episodes at EL than at BL (p < 0.05)”
Panel auditInternal validity: low · scope of inference: Ghanaian children 6-35 months stratified by haemoglobinopathy genotype and ABO blood group.Key limitations: within-arm EL vs BL contrast, not vs control · seasonal confounding · multiple subgroup testing · mixed direction across genotypes
Independent RCT in a different endemic setting extends the harm signal to malaria and diarrhea in older children.
“In older children (more than five years of age), iron supplementation increased morbidity due to Plasmodium vivax and diarrhea.”
Panel auditInternal validity: medium · scope of inference: Older children (>5 years) in the Peruvian Amazon, a P. vivax malaria-endemic area.Key limitations: subgroup (older children) effect · no numeric effect size or CI given · wide age range 0.5-15 years · opposite/protective signal in younger children with iron plus zinc
Counter-evidence: under bed nets and available treatment, no increased malaria; bounds the claim's conditionality.
“daily use of a MNP with iron did not result in an increased incidence of malaria among young children”
Panel auditInternal validity: high · scope of inference: Children 6-35 months in rural Ghana with insecticide-treated nets and active malaria treatment available.Key limitations: difference lost significance after adjusting for baseline iron/anaemia (adjusted RR 0.87, 0.75-1.01) · co-intervention with bed nets and treatment · severe anaemia excluded
4 of 10 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
A real, well-documented safety signal from strong RCTs is offset by equally strong RCTs showing no risk under malaria control, leaving the unconditioned claim only partially supported.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Targeted versus untargeted allocation not randomised. The claim turns on 'untargeted', yet no trial compared iron restricted to iron-deficient children against blanket supplementation in the same setting. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A factorial RCT in an endemic setting randomising iron-deficiency-screened targeting versus universal supplementation, with severe illness and infection as endpoints.
Open question: Does the excess risk arise specifically from supplementing children who are not iron deficient? Why it is still open: [1] interprets harm as concentrated in those not iron deficient, but no trial randomised targeted against untargeted allocation to test it directly.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 18 Sep 2026 · protocol v0.19