von Hertzen 2009.

Methods	computer generated, central randomisation, random permutation in groups of 8, stratified by gestational age; sealed, opaque sequentially numbered envelopes
Participants	2181 women; gestational age </=63 days, inclusion criteria: haemoglobin >100g/L, willing to have surgical abortion for failure, agreed to return for follow‐up. exclusion criteria: ill health, contraindications to study medication, severe uncontrolled asthma, porphyria, valvular heart disease, smoking and another risk for CV disease, glaucoma, thromboembolism, liver disease, IUD in situ, breastfeeding, haemolytic disorders. Gestational age confirmed by ultrasound. Study conducted between 2003‐2005 at 13 departments of obstetrics and gynaecology in nine countries (China, Hungary, India, Mongolia, Romania, Slovenia, South Africa,  Viet Nam, Serbia)
Interventions	group 1: mifepristone 100 mg and 24 h later misoprostol 800 mcg vaginal group 2: mifepristone 100 mg and 48 h later misoprostol 800 mcg vaginal group 3: mifepristone 200 mg and 24 h later misoprostol 800mcg vaginal group 4: mifepristone 200 mg and 48 h later misoprostol 800mcg vaginal follow up at 2 and 6 weeks
Outcomes	complete , incomplete, missed abortion, side effects
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A‐adequate