Universal iron with malaria chemoprevention did not improve infant cognition in Malawi

Design

Four-arm randomised clinical trial

Population

2,168 six-month-old infants in rural, malaria-endemic Malawi, 71% anaemic

Primary outcome

Bayley-III cognitive composite score after six months of intervention

Effect

Iron syrup −0.52 (−2.52 to 1.48); micronutrient powder −0.85 (−2.80 to 1.09) vs chemoprevention alone

IRMA randomised 2,168 six-month-old infants in rural, malaria-endemic southern Malawi to six months of iron syrup with malaria chemoprevention, iron-containing micronutrient powder with chemoprevention, chemoprevention alone, or placebo. Seventy-one per cent were anaemic at entry.

Neither iron strategy improved the primary outcome, the Bayley-III cognitive score at one year, compared with chemoprevention alone (mean differences −0.52 and −0.85, both intervals crossing zero). Both raised ferritin, but neither reduced the prevalence of anaemia. Reassuringly, neither increased malaria. Chemoprevention reduced malaria compared with placebo.

The WHO recommends universal preventive iron for young children where anaemia is common, on the assumption that it benefits development. This trial did not find that benefit in this setting, though it was not designed to show harm and did not find any. Whether the result applies to Indian children, where the causes of anaemia and malaria exposure differ, is uncertain.

• Iron supplementation raised ferritin but did not lower anaemia rates when given with malaria chemoprevention.
• No increase in malaria was seen with iron in this trial, an important safety finding for endemic areas.
• Malaria chemoprevention itself reduced malaria incidence against placebo.
• Look for other causes of infant anaemia — infection, inflammation, haemoglobinopathy — rather than assuming iron deficiency.

Why it matters

It questions whether universal infant iron delivers the developmental benefit it is recommended for.

Don't overread it

The guideline on preventive iron has not changed, and the setting differs from most Indian practice.

The statistics, in plain English

Mean differences of −0.52 and −0.85 points on a scale with a mean around 100 are tiny, and their confidence intervals cross zero, so the trial found no detectable effect. It was large enough to rule out a meaningful benefit of a few points.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.