Walker 2014.

Methods	Paired cluster‐randomised trial
Participants	Who: Interprofessional teams at 24 study hospitals Number: 450 physicians and nurses Proportion of eligible staff participating: 6.4 to 31.6% of eligible staff trained, mean participation rate 20.5%. 3228 were eligible.
Interventions	Intervention description: PRONTO obstetric emergency training, which is largely interactive using a low‐technology hybrid simulator Control: No intervention Location: Unclear Delivered by: PRONTO trainers Length: 24 hours of training; 2 days followed by 1 day Duration: Single intervention
Outcomes	Outcomes: Perinatal mortality, neonatal mortality, achievement of strategic planning goals Follow‐up: 3 years
Population studied	Description: Hospital‐based neonatal and maternal morbidity and mortality at the study sites Number: 58,837 deliveries and 641 births observed, deliveries at study sites.
Funding Source	Mexican National Institute of Women (INMUJERES), the State Secretary for Women in the states of Chiapas and Mexico, Bill and Melinda Gates Foundation, and the Laerdel Foundation
Study Setting	24 community hospitals in Mexico
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	No description of how randomised, and 4 hospitals not randomised.
Allocation concealment (selection bias)	Unclear risk	Not discussed
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not discussed, but not possible to blind participants
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not discussed
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Not discussed
Selective reporting (reporting bias)	High risk	Original primary and secondary outcomes could not all be reported due to rarity. This was made clear and justified.
Other bias	High risk	In 2 of the pairs, the local ministry of health selected the intervention hospitals rather than them being randomised. 11 of the original hospitals dropped out for a variety of reasons and were replaced by 11 similar sites.