Zurich 1988.

Methods	Quasi‐randomised controlled trial.
Participants	568 women were randomised. Setting: Outpatients clinic at the Department of Obstetrics, University of Zurich, from 1983 to 1985. Inclusion criteria: women with normal and high‐risk pregnancies, at no later than 16 weeks' gestation.
Interventions	Treatment group (n = 278) Women received 15 mmol of magnesium aspartate‐hydrochloride daily from ≤ 16 weeks' gestation until delivery. Control group (n = 290) Women received a control tablet containing 13.5 mmol of aspartic acid per day. No detail of the appearance/taste of these tablets being similar to the treatment tablets. The doses were divided into 6 tablets to be taken daily.
Outcomes	Adverse effects; miscarriage; maternal weight increase; blood pressure; oedema; hospitalisation (and indication for hospitalisation); median gestation; preterm birth; duration of the first and second stage of labour; operative birth; placental weight; infant weight; low birthweight; infant length; head circumference; neonatal acid‐base values; admission to the neonatal intensive care unit; perinatal death; low Apgar score (≤ 7).
Notes	Sample size was determined by the duration of the study’s recruitment period (2 years). Groups were comparable at baseline.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Allocation was based on the women's date of birth.
Allocation concealment (selection bias)	High risk	Women with even birth date given magnesium‐aspartate‐hydrochloride, women with odd birth date given aspartic acid.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	The study was described as “double‐blind” with the use of a placebo; however allocation was based on the women's date of birth, and thus it is unclear as to whether blinding would have been successfully achieved for women and personnel.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	The study was described as “double‐blind” with the use of a placebo; however allocation was based on the women's date of birth, and thus it is unclear as to whether blinding would have been successfully achieved for outcome assessment.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Actual group values (total number per group) are not reported in the outcome data tables, and thus it is difficult to determine the exclusions and attrition and where there is missing data; quote “For various reasons such as refusal to take further tablets, delivery in other hospitals or abortion, some data were not available for analysis."
Selective reporting (reporting bias)	Unclear risk	Outcome data reported for a range of important outcomes – these however were not pre‐specified in the methods, and it is therefore difficult to assess selective reporting.
Other bias	Unclear risk	No other obvious sources of bias identified.

EPH: edema, proteinuria, hypertension
 GI: gastrointestinal
 IVF: in vitro fertilisation