Czeizel 1994.

Methods	RCT 2‐arm parallel‐group design.
Participants	5502 pregnant women (out of the 7905 participants) attending prenatal care at the Hungarian Family Planning Program (HFPP) in Hungary. Inclusion criteria were: (I) no delayed conception or infertility (i.e. no conception after more than 12 months of sexual activity without contraception), (II) not currently pregnant, (III) voluntary participation and a promise of compliance. In the first 4 years of the HFPP, there were 2 other criteria: age under 35 (women over 35 were referred to a genetic counselling clinic) and no previous wanted pregnancy. Of the 5502 women, 49 were lost to follow‐up and 95 had false‐positive pregnancy results ("chemical pregnancies"). Thus, 5358 women were included in the meta‐analysis.
Interventions	Women were randomly assigned to 1 of 2 groups: Group 1 (n = 2738 women): women received a supplement (Elevit Pronatal®) containing 800 µg (0.8 mg) folic acid, in addition to 6000 IU (until the end of 1989) and 4000 IU (in 1990 to 1991) vitamin A; 1.6 mg vitamin B1; 1.8 mg vitamin B2; 19 mg nicotinamide; 2.6 mg vitamin B6; 10 mg pantothenic acid (as calcium pantothenate); 0.2 mg biotin; 4.0 pg vitamin B12; 100 mg vitamin C; 500 IU vitamin D; 15 mg vitamin E (as alpha‐tocopherol‐tocopherol acetate); 125 mg calcium; 125 mg phosphorus; 100 mg magnesium; 60 mg elemental iron; 1 mg copper; 1 mg manganese; and 7.5 mg zinc. Group 2 (control; n = 2620 women): women received a supplement containing 1 mg copper; 1 mg manganese; 7.5 mg zinc; 7.5 mg (calcium ascorbate) vitamin C and lactose 736.27 mg. A single tablet in either group was to be taken daily for 1 month before planned conception until 12 weeks of pregnancy (confirmed by sensitive pregnancy test after the first missed menstrual period and by ultrasonography within 2 weeks). Women were asked to ingest the tablet each day before recording the basal body temperature and to leave unused tablets in the box.
Outcomes	Maternal: weight gain, increased appetite, lack of appetite, nausea and vomiting, vertigo, heartburn, constipation, diarrhoea, ectopic pregnancy, miscarriage, spontaneous abortion, multiple birth, twin birth. Infant: NTDs and other congenital anomalies (congenital limb deficiency, cardiovascular congenital abnormalities, congenital pyloric stenosis, cleft lip, cleft palate) stillbirths, infant mortality, prenatally terminated fetuses, birthweight, low birthweight, gestational age, weight, length, and head circumference at 8 to 16 months of age, functional development tests at 8 to 16 months.
Notes	Exclusion criteria not clear. The final database included 5358 women with evaluated pregnancy. Summary characteristics of the study: by scheme: daily supplementation; by daily dose: more than 400 µg folic acid per day (800 µg or 0.8 mg of folic acid was provided per day); by timing of supplementation: periconceptional period (supplements were provided at least 28 days before conception for at least until the second missed menstrual period); by history of a pregnancy affected by a NTD (recurrence): no; by folate compound: folic acid; by micronutrient composition: folate + multiple micronutrients; by malaria setting at the time of the study: unreported. Source of funding: Hoffmann‐La Roche supplied the supplements.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Women were asked about whether they agreed to their allocation on the basis of a randomisation table.
Allocation concealment (selection bias)	Unclear risk	Not described. It was not clear whether staff carrying out recruitment were aware of randomisation group.
Blinding (performance bias and detection bias) Women	Low risk	Women were informed about the "blind' use of 1 of 2 kinds of tablets.
Blinding (performance bias and detection bias) Clinical staff	Unclear risk	Not clear if staff were blinded.
Blinding (performance bias and detection bias) Outcome assessors	Unclear risk	Certificates were filled in by mothers and signed by physicians. If certificates were not sent back, one of the co‐workers visited the participants at home.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Of 7905 women randomised 69.6% had confirmed pregnancies. Loss to follow‐up for women with confirmed pregnancies 1%.
Selective reporting (reporting bias)	Unclear risk	Not apparent.
Other bias	Unclear risk	In a comment in Czeizel 1992 it is stated that randomisation was broken twice; first to evaluate teratogenic effect of vitamin A and the second at the end of the trial in 1991.