| Methods |
Double‐blind randomised‐controlled trial. Tablets prepared by pharmacy. Randomisation based on permutation blocks of 4. |
| Participants |
120 women with indication for induction of labour between 37 and 41 weeks. Bishop score < 4. Exclusion: medical conditions, non‐vertex presentation, > 1 previous caesarean section, multiple pregnancy, premature rupture of membranes. 112 women included in analyses: see Notes. |
| Interventions |
Mifepristone 200 mg orally for 2 days, or placebo. All women observed daily for 4 days; if not then in labour, induction by vaginal prostaglandins (cervix still unfavourable) or artificial rupture of membranes and oxytocin (favourable cervix). |
| Outcomes |
Duration of labour, obstetric and neonatal outcome, PGE2 tablets, oxytocin dose, neonatal glucose and blood pressure on days 1 and 2. |
| Notes |
8 women required delivery by CS at < 12 hours after initial treatment because of fetal distress or severe hypertension (3 mifepristone; 5 placebo). They were not included in analyses. Preliminary results on 62 women published in Lancet letter. Full results published in French in 1993. |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Adequate sequence generation? |
Low risk |
Quote: "women were randomly allocated". Comment: "detailed information included in paper". |
| Allocation concealment? |
Low risk |
Adequate. "Used a balanced randomisation list obtained by premutation block". |
| Blinding?
All outcomes |
Low risk |
Adequate, as placebo and mifepristone tablets were visually identical and externally produced. |
| Incomplete outcome data addressed?
All outcomes |
High risk |
8 patients excluded (3 in mifepristone group and 5 in placebo group) as they required caesareans for medical reasons within 12 hours. |
| Free of selective reporting? |
Low risk |
Adequate. All listed outcomes were reported on |
| Free of other bias? |
Low risk |
|
