Choudhury 2012.

Methods	Cluster‐randomised non‐inferiority trial. 2‐arm study.
Participants	478 pregnant women (gestational age 14–22 weeks) attending 42 antenatal care centres (ANC) in the rural sub district of Kaliganj in central Bangladesh. Women with haemoglobin lower than 70 g/L or greater than 140 g/L, with more than 22 weeks of gestation, or who were already taking iron supplements prior to the start of the study were excluded.
Interventions	Participants were randomised, at the cluster level (ANC), to 1 of 2 interventions: group 1 (21 clusters, n = 243) received micronutrient powders containing (60 mg of elemental iron as ferrous fumarate), 400 μg of folic acid, 30 mg of vitamin C, and 5 mg of zinc) daily or group 2 (21 clusters, n = 235) received tablets containing 60 mg elemental iron and 400 μg (0.4 mg) folic acid daily. Participants from group 1 were instructed to sprinkle the full content of a package of the powders into any semi‐solid or semi‐liquid food.
Outcomes	Haemoglobin concentrations, mild and moderate anaemia at baseline, 24, 28 and 32 weeks of gestation, adherence to intervention.
Notes	Baseline nutritional status of women at the start of intervention mixed group 1: 50% (n = 121) and group 2: (39.1%, n = 92). Malaria status in the study area: absent/unknown/unreported malaria at the time of the trial. Dosage of elemental iron in the micronutrient powder (per portion added or individual sachet): more than 30 mg. Provision scheme of the micronutrient powder intervention: daily. Source of funding: HJ Heinz Company Foundation. Clustering effect accounted in the original analysis.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Method of randomisation not reported.
Allocation concealment (selection bias)	Low risk	As this is a cluster‐randomised trial, the risk of selection bias at the individual level is low.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants were not blinded to the interventions.
Blinding of outcome assessment (detection bias) All outcomes	High risk	Outcome assessors were not blinded to the interventions.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Overall attrition of 15%, similar in both groups.
Selective reporting (reporting bias)	Unclear risk	It is unclear if outcomes other than those reported were assessed and purposely not reported.
Other bias	High risk	Higher prevalence of baseline anaemia among participants from group 1 receiving micronutrient powders (50.0%) compared to the participants in group 2 receiving iron and folic acid tablets (39.1%).