Fawzi 2007

Methods	This was a double‐blind trial in Dar es Salam, Tanzania. Pregnant women who attended antenatal clinics between August 2001 and July 2004 were included.
Participants	Pregnant women who attended antenatal clinics, had a negative test for HIV infection, planned to stay in the city until delivery and for 1 year thereafter with gestational age between 12 and 27 weeks according to LMP were included. The study groups were similar with respect to baseline characteristics.
Interventions	Intervention group (n = 4214) received vitamin B1 20 mg, B2 20 mg, B6 25 mg, B12 50 μg, C 500 mg, E 30 mg niacin 100 mg, folic acid 0.8 mg. Control group (n = 4214) received placebo. Women were randomly assigned to receive either MM or control from the time of enrolment until 6 weeks after delivery.
Outcomes	LBW (< 2500 g), preterm delivery (< 37 weeks of gestation), very LBW (< 2000 g), extremely preterm delivery (< 34 weeks of gestation), SGA (< 10th percentile for gestational age), fetal death, death in first 6 weeks, length, head circumference, placental weight, risk of caesarean section, maternal mortality, haematological status (Hb < 11 g/dL and < 8.5 g/dL, immune status (CD4 count < 775 per cubic mm, CD8 count < 480 per cubic mm and CD3 count < 1350 per cubic mm).
Notes	All women irrespective of group received iron 60 mg and folic acid 0.25 mg. Malaria prophylaxis (sulphadoxine‐pyrimethamine tablets) at 20 and 30 weeks of gestation was given to all. The study groups were similar with respect to baseline characteristics.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "A list was prepared according to a randomization sequence in blocks of 20; at enrolment, each eligible women was assigned to the next numbered bottle" and computerised random number generator was used (personal communication). Comment: probably done.
Allocation concealment (selection bias)	Low risk	Quote: "Each eligible women was assigned to the next numbered bottle". Comment: probably done.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Exclusion (0.5%) and attrition (5.4%) were reported with reasons in each arm.
Selective reporting (reporting bias)	Low risk	Comment: all outcomes mentioned in the methods section were presented in the paper.
Other bias	Low risk	Comment: no other bias was identified.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Active tablets and placebo were similar in shape, size and color and were packaged in identical coded bottles" and "Each eligible women was assigned to the next numbered bottle". Comment: participants and caregivers were blinded to the treatment assignment.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "research assistants who assessed the study outcome were unaware of the intervention group". Comment: outcome assessors were blinded.