Gülmezoglu 2006.

Methods	Cluster RCT Follow up: providers: DONE patients: N/A Blinded assessment: DONE Baseline: DONE Reliable outcomes: DONE Protection against contamination: DONE Overall quality: HIGH
Participants	Doctors, midwives, interns and students in obstetric practices in 22 hospitals in Mexico City and 18 in Thailand Country: Mexico and Thailand Proportion of eligible providers who participated: 65% Hospital, inpatients Academic/Teaching setting: NON‐TTEACHING Type of targeted behaviour: GENERAL MANAGEMENT OF A PROBLEM (OBSTETRIC PRACTICES) Complexity of targeted behaviour: HIGH
Interventions	1. CME: series of 3 workshops at time 0, after 6 weeks and after 6 months + Meeting with hospital directors and department heads + Provision of the database, computers and printers + coordinator from each hospital 2. No intervention control
Outcomes	Professional practice: % change in practice rates in 6 provider behaviours for obstetric care Patient: none Seriousness of outcome: HIGH
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate