Worrall 1999.

Methods	Cluster RCT Follow up: providers: DONE patients: NOT CLEAR Blinded assessment: NOT CLEAR Baseline: DONE Reliable outcomes: NOT CLEAR Protection against contamination: DONE Overall quality: MODERATE
Participants	42 family physicians Country: Canada Proportion of eligible providers who participated: 41% General practice Academic/Teaching setting: NOT CLEAR Type of targeted behaviour: GENERAL MANAGEMENT OF A PROBLEM (management of depression) Complexity of targeted behaviour: MEDIUM
Interventions	1. CME: 3‐hr small group educational session: teaching and case‐based discussion + psychiatrist help service available once a week + clinical guidelines 2. Control: mailed clinical guidelines on the management of depression
Outcomes	Professional practice: % of correct diagnoses of depression Patient: mean patient score on the Centre for Clinical Epidemiological Studies Depression scale Seriousness of outcome: MODERATE
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk	A ‐ Adequate