WHO 2001 GP1pv.

Methods	computer generated sequence of random numbers in block of ten, identical placebo tablets
Participants	multicentre trial, 10 centres: Chandigarh, Edinburgh, Havana, Hong Kong, Ljubljana, Shanghai, Stockholm, Szeged, Tbilisi, Tianjin 896 women, at 57 to 63 days of gestation with regular menstrual cycles, pregnancy confirmed clinically or by ultrasound exclusion criteria: contraindication to the study drugs, chronic respiratory, digestive, endocrine, genito‐urinary, neurological or cardio‐vascular disease, severe liver disease, history of thrombo‐embolism, IUCD in situ, breastfeeding
Interventions	group 1: mifepristone 200 mg group 2: mifepristone 600 mg and gemeprost 1 mg after 48 hours (all)
Outcomes	complete, incomplete, missed abortion, time to onset of bleeding, duration of bleeding, time to return to menses, bleeding before gemeprost, time of expulsion
Notes	power calculation ( 80% power at a significant level of 0.05 ) intention ‐to ‐treat analysis 2 women received blood transfusion, not mentioned which group
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate