Butler 1967.

Methods	RCT, 3‐arm trial with individual randomisation.
Participants	200 women before 20th week of gestation and Hb above 100 g/L attending antenatal clinic at the Maternity Hospital in Glossop Terrace, Cardiff, United Kingdom were studied. Exclusion criteria included urinary infection and threatened miscarriage, confusion over therapy, intercurrent illness and difficult veins, intolerant to the iron form, premature labor.
Interventions	Participants were randomly allocated to 1 of 3 groups: group 1: received 122 mg of elemental iron (as ferrous sulphate) daily; group 2: received 122 mg of elemental iron (as ferrous sulphate) + 3400 μg (3.4 mg) of folic acid daily; group 3: received no intervention. A group 4 was formed as some participants (n = 38) from group 3 received iron supplements for treatment of anaemia in the course of the intervention. They are excluded from the analysis. Women were supplemented from week 20 to 40 of gestation. Setting and health worker cadre: the intervention was performed by obstetricians and hematologists at the antenatal clinic, Cardiff Maternity Hospital in Cardiff, United Kingdom.
Outcomes	Maternal: Hb concentrations, blood and plasma volume, HCT (not reported), MCV, albumin and globulin fractions at weeks 20, 28, 36 and 40 of gestation and at the first postanal visit, oedema, intrapartum haemorrhage.
Notes	Unsupervised. 154 women were followed through to the postnatal visit. Only 16 women (30%) in the no‐treatment group remained untreated. Compliance not reported. Gestational age at start of supplementation: early gestational age (supplementation started before 20 weeks' gestation). Anaemic status at start of supplementation: mixed anaemia status (Hb above 100 g/L). Daily iron dose: higher daily dose (60 mg of elemental iron or more) (122 mg elemental iron). Iron release formulation: normal release iron supplement/not specified. Iron compound: ferrous sulphate. 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	Randomised list stratified by age, parity and initial Hb level.
Allocation concealment (selection bias)	Low risk	The code was not opened for the iron and iron + folic acid group until the end of the investigation, thus clinical staff could not anticipate the randomisation sequence. There was no treatment for 1 group.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participant and provider were blinded to treatment for groups 1 and 2. The control group received no treatment and did not get a placebo.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	It was not clear whether those assessing outcomes were aware of allocation, but it is unlikely that this possible lack of blinding affected the laboratory outcomes reported.
Incomplete outcome data (attrition bias) All outcomes	High risk	More than 20% were lost to follow‐up to the postnatal visit. 154 women were randomised and for many outcomes there were missing data. 70% of the 54 women initially allocated to the no treatment group received iron supplements for anaemia (as there was no placebo, staff would be aware that women were not receiving supplements). Results for those women treated or not treated in the control group were reported separately. Results are therefore difficult to interpret.
Selective reporting (reporting bias)	Low risk	Authors provided the full database for this review.
Other bias	Low risk	No baseline imbalance apparent.