Schnipper 2010.

Study characteristics
Methods	Cluster RCT
Participants	Primary care clinics within a regional academic medical network, USA (Partners HealthCare System) 7009 patients, 239 providers, 10 practices
Interventions	Documentation‐based reminder for patients with coronary artery disease or diabetes that provided decision support with tailored recommendations for care
Outcomes	Process adherence (prescribing, testing, documentation)
Co‐Interventions	Educational: Single educational session for providers in intervention group Beyond Clinician Education: Local opinion leaders’ endorsement and audit and feedback in intervention group
CDSS Features ‐ Acknowledgement of CDSS Required	No
CDSS Features ‐ Other	Conveyed patient‐specific information, developed by study investigators, makes care recommendation, 'pull' mode of delivery, targeted underuse
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Allocation concealment (selection bias)	Low risk
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	N/A
Baseline characteristics similar?	Low risk
Unit of analysis error	Low risk