Kerr 1958.

Methods	RCT, 4 arms with individual randomisation.
Participants	430 apparently healthy women with 24‐25 weeks of singleton pregnancy and Hb equal or above 104 g/L attending antenatal clinic at Simpson Memorial Maternity Pavillion, Edinburgh, United Kingdom.
Interventions	Participants were randomly allocated to 1 of 4 groups: group 1 received 35 mg of elemental iron (as ferrous sulphate) 3 times a day; group 2 received 35 mg of elemental iron (as ferrous gluconate) 3 times a day; group 3 received 35 mg of elemental iron (as ferrous gluconate) with 25 mg of ascorbic acid, 3 times a day; group 4 received placebo. Supplementation started at 24‐25th week of gestation until term. Setting and health worker cadre: the intervention was performed by physicians at the Simpson Memorial Maternity Pavilion in Edinburgh, United Kingdom.
Outcomes	Maternal: Hb, red cell count, HCT at baseline and at 37th week.
Notes	Unsupervised. Groups 1 and 2 were merged for analysis. Group 3 was not used in this review. Compliance not measured. Gestational age at start of supplementation: late gestational age (supplementation started after 20 weeks' gestation). Anaemic status at start of supplementation: unspecified/mixed anaemia status (no severe anaemia, all had Hb equal or above 104 g/L). Daily iron dose: higher iron dose (all treatment groups received more than 60 mg of elemental iron daily (105 mg)). Iron release formulation: normal release preparation/unspecified. Iron compound: ferrous gluconate. Malaria setting: non‐malarial setting. As of 2011: Malaria: no risk.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	By cards shuffle.
Allocation concealment (selection bias)	Unclear risk	Not described.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Partial blinding. Placebo and all supplements described as identical. Women were blinded, but medical staff were aware of which was the control group.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Low risk for laboratory outcomes. Possible risk of bias for dietary survey and reporting of side effects.
Incomplete outcome data (attrition bias) All outcomes	High risk	23% of participants were lost to follow‐up.
Selective reporting (reporting bias)	Unclear risk	There is insufficient information to permit judgement.
Other bias	Unclear risk	No other bias apparent.