Eskeland 1997.

Methods	RCT, 3‐arm trial with individual randomisation.
Participants	90 healthy non‐anaemic pregnant women with singleton pregnancy of less than 13 weeks, attending an inner city maternity centre in Bergen, Norway and willing to participate. Exclusion criteria: uncertain gestational age according to menstrual history, Hb concentration < 110 g/L, chronic disease or pregnancy complications (hypertension, diabetes, bleeding), multiple pregnancy, liver enzymes out of normal range and logistic difficulties foreseen at baseline (moving out of area).
Interventions	Participants were randomly allocated to 1 of the following: group 1: 3 tablets containing 1.2 mg heme iron from porcine blood and 9 mg of elemental iron (as ferrous fumarate) (Hemofer®) and 1 placebo tablet (total 27 mg elemental iron a day); group 2: 1 tablet containing 27 mg elemental iron (as iron fumarate) with 100 mg vitamin C (Collet®) and 3 placebo tablets; group 3: 4 placebo tablets. Supplementation started at 20th week until 38‐40th week. Setting and health worker cadre: the intervention was performed by midwives and physicians at an inner city maternity centre in Bergen, Norway.
Outcomes	Maternal: Hb, RBC count, HCT, MCV, MCH, MCHC, reticulocytes, serum iron, total iron binding capacity, serum transferrin, erythrocyte protoporphyrin at baseline and at 20, 28, 38 weeks, 8 weeks postpartum, and 6 months postpartum; pregnancy complications: hypertension, pre‐eclampsia, forceps, postpartum haemorrhage, maternal well being and breastfeeding duration. Infant: birthweight and length.
Notes	Unsupervised. Only groups 1 and 3 (placebo) were included in this review. Compliance was 81% and 82% in groups 1 and 3 respectively. Gestational age at start of supplementation: late gestational age (supplementation started at or after 20 weeks' gestation). Anaemic status at start of supplementation: non‐anaemic. Daily iron dose: lower daily dose (less than 30 mg elemental iron daily). Iron release formulation: normal release preparation/not specified. Iron compound: iron fumarate. 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	Computer‐generated.
Allocation concealment (selection bias)	Low risk	This was a placebo‐controlled trial.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Described as a double‐blind trial (placebo controlled). It was stated that staff providing care were not aware of treatment allocation and were only given information about Hb.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Double‐blind trial and laboratory outcomes.
Incomplete outcome data (attrition bias) All outcomes	High risk	23% and 21% in groups included.
Selective reporting (reporting bias)	Unclear risk	There is insufficient information to permit judgement.
Other bias	Unclear risk	No other bias apparent.