006 Ewert 2006.

Methods	Double‐blinded randomised multicenter study. Randomisation and allocation concealment were computer generated.
Participants	Inclusion criteria: pregnant women were aged 18 years or older, 37 to 42 weeks of gestation, requiring cervical ripening and labour induction and at least 1 previous delivery with gestation of 37 weeks or more. The current pregnancy had to be a singleton fetus with cephalic presentation and the women had to be classified as a pregnant uncomplicated with Bishop score less than 6. Exclusion criteria: women with plus than 4 terms full deliveries and those with caesarean delivery. Spontaneous labour, tocolytic agents used within 7 days before induction, the use of any other cervical‐ripening or labour‐inducing agent before enrolment, suspected cephalopelvic disproportion, fetal distress, use of non steroidal anti‐inflammatory drugs 4 hours before the drug studied treatment, pyrexia, unexplained genital bleeding after 24 weeks of this pregnancy, pelvic inflammatory disease, placenta previa and known or suspected allergy to misoprostol or other prostaglandins.
Interventions	A single misoprostol vaginal insert was administered high into the posterior fornix and positioned transversally behind the cervix. Each insert was loaded with 1 of the 4 dose reservoirs of misoprostol being investigated: 25, 50, 100 and 200 mcg. The vaginal insert was designed to release misoprostol at a rate of approximately 1, 2, 4 and 8 mcg per hour respectively.
Outcomes	The median time to vaginal delivery was 27.5, 19.1, 13.1 and 10.6 hours for the 25‐, 50‐, 100‐ and 200‐mcg doses, respectively. The percentage of women who delivered within 12 hours was 9%, 14%, 47% and 53% (P < .001 using the 25 mcg group as the comparator) and within 24 hours was 42%, 79%, 81% and 70%(P = .003). Uterine hyperstimulation syndrome occurred in 1 women who received the 25 mcg, 2 women who received the 100 mcg, and 3 women who received 200 mcg dose reservoirs.
Notes	This trial was conducted at 6 sites in the United Kingdom. At this review the comparison of doses were made grouping lower 2 doses vs higher 2 doses.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Adequate
Allocation concealment (selection bias)	Low risk	Adequate
Blinding (performance bias and detection bias) All outcomes	High risk	Unblinded only for the outcome assessors
Incomplete outcome data (attrition bias) All outcomes	High risk	Adequate
Selective reporting (reporting bias)	Low risk	Adequate
Other bias	Low risk	Adequate