Hofmeyr 2001

Methods	2‐arm placebo‐controlled randomised trial.
Participants	Between dates unspecified, 600 parturients were randomised in a hospital setting in South Africa. The population comprised women of unspecified parity, unspecified whether singleton or multiple pregnancy, at unspecified for PPH, who delivered by vaginal delivery. Exclusion criteria were not specified.
Interventions	600 mcg of misoprostol administered orally (n = 300) versus placebo or control (n = 300).
Outcomes	The study recorded the following outcomes: PPH at 1000, additional uterotonics, transfusion, manual removal of placenta, nausea, vomiting, fever, shivering, abdominal pain.
Notes	Contact with study authors for additional information: no. Additional data from authors: no.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random assignments were generated by computer in blocks of 18.
Allocation concealment (selection bias)	Low risk	Investigators used sequentially‐numbered, opaque test tubes.
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Misoprostol and placebo were similar in size and colour but not shape
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Assessor blinding was not reported.
Objective assessment of blood loss	Low risk	Within a minute of delivery, investigators removed any linen soiled with amniotic fluid, and placed a fresh, disposable absorbent linen‐saver sheet with plastic backing, and a low wedge‐shaped plastic "fracture" bedpan under the mother. "This was found to be a comfortable and efficient way of collecting the great majority of blood lost after delivery, and could be left in place without discomfort even during perineal suturing. When active bleeding had stopped, any blood clots were expressed from the uterus, the bedpan was removed and a sanitary towel was applied. The [volume of] blood in the bedpan was measured in a measuring jug. An hour after delivery, any bloodstained linen‐savers and sanitary towels were placed in a plastic bag and weighed in g." After subtracting the known dry weights of these materials, the bloodstained weights were added to the volume of blood collected in the bedpan to ascertain the total blood loss in the first hour after delivery".
Incomplete outcome data (attrition bias) All outcomes	Low risk	"There were no withdrawals after randomisation and all outcomes were analysed in the allocated group." However the primary outcome data of 1 study participant in the placebo group were unavailable.
Selective reporting (reporting bias)	Unclear risk	The protocol of the study was unavailable for verification.
Intention to treat analysis	Unclear risk	All those who were enrolled and randomly allocated to treatment were included in the analysis, in the groups to which they were randomised.
Funding source	Low risk	The study was supported by funding from the South African Medical Research Council (public funding) and University of the Witwatersrand (the institution of the authors).