Rowlands 2003.

Methods	Cluster RCT Follow up: providers: DONE patients: N/A Blinded assessment: DONE Baseline: NOT CLEAR Reliable outcomes: DONE Protection against contamination: DONE Overall quality: MODERATE
Participants	26 practices with four or more partners Country: UK Proportion of eligible providers who participated: 9% General practice Academic/Teaching setting: NOT CLEAR Type of targeted behaviour: other (referrals to specialist services) Complexity of targeted behaviour: LOW
Interventions	1. CME: several within‐practice educational meetings, total of 5 hrs on average for each practice 2. No intervention control
Outcomes	Professional practice: referral rates Patient: none Seriousness of outcome: LOW
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate