Elwyn 2004.

Methods	Study design: cluster‐randomized trial Unit of allocation: provider (one per practice) Unit of analysis: provider Power calculation: done
Participants	Care setting: primary care; ambulatory care (usual practice and protected research clinics; urban and rural in Gwent, South Wales); UK Healthcare professionals: 21; general practitioners; fully trained Patients: 747 included in COMRADE, 352 in OPTION; non‐valvular atrial fibrillation or prostatism or menorrhagia or menopausal symptoms; male or female
Interventions	Multifaceted intervention: educational meeting (SDM skills) and audit and feedback; 5 hours Practitioners attended two workshops. During the first workshop, the background literature on SDM was outlined and participants were asked to debate its relevance to clinical practice. The skills of SDM were described and demonstrated using simulated consultations. This provided opportunities for all the participants to comment on the method, using an observational competence checklist. Simulated patients were also encouraged to comment. Participants were asked to consult with the simulated patients using pre‐prepared scenarios involving the study conditions. At the second workshop, participants were asked to consider the competences in more depth. By the end of the workshop, all participants had conducted and received feedback from at least one consultation with a simulated patient. Multifaceted intervention (control): educational meeting (risk communication skills) with audit and feedback; 5 hours A risk communication aid was presented for the four study conditions. The risk data were based on systematic reviews and presented as the best evidence available at the time of the trial. The participants were provided with treatment outcome information for the study conditions. Participants were asked to use them in simulated patient consultations. The consultations were conducted in pairs, where colleagues alternated between clinician and observer roles. This was repeated until each participant had received feedback after conducting two or three consultations using the risk communication aids across a range of conditions. A plenary group discussion, which included the patient simulators, allowed the group to share learning points and consider the application of the materials in clinical practice.
Outcomes	OPTION (continuous); SDM is assessed as the fostering by healthcare professionals of active participation of patients in the decision‐making process COMRADE (continuous); joint process between healthcare professionals and patients to make decisions
Notes	Additional information Number of approached patients (eligible): 2585 Number of patients per physician:12 or 24 patients per physician according the phase (baseline, first and second intervention)
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "All randomizations were undertaken by random number generation, and allocations by the trial statistician (KH) were concealed from those implementing the interventions or assessments." page 339
Allocation concealment (selection bias)	Low risk	Comment: unit of allocation was by provider or practice.
Blinding (performance bias and detection bias) Observer‐based outcome	Low risk	Quote: "All consultation recordings were intended to be rated by two raters and rating were undertaken blind to study group allocation of clinicians or patients." page 340
Blinding (performance bias and detection bias) Participant‐reported outcome	Low risk	Quote: "Both clinicians and patients were informed that the trial was investigating "communication skills" but were otherwise blinded to the decision‐making or risk communication focus of the intervention." page 339
Incomplete outcome data (attrition bias) Observer‐based outcome	Low risk	Comment: one summary measure for all physician.
Incomplete outcome data (attrition bias) Participant‐reported outcome	Unclear risk	Comment: follow‐up was not clear in the 2 articles.
Selective reporting (reporting bias)	Unclear risk	Comment: no evidence that outcomes were selectively reported, but no protocol.
Other bias	Low risk	Comment: no evidence of other risk of biases.
Baseline measurement? Observer‐based outcome	Unclear risk	Comment: not specified in the paper.
Baseline measurement? Participant‐reported outcome	Unclear risk	Comment: not specified in the paper.
Protection against contamination?	Low risk	Quote: "Unit of allocation is the provider. Only one practitioner per practice would be recruited." page 338
Baseline characteristics patients	Low risk	Comment: see Table 1 in Edwards et al. page 351
Baseline characteristics healthcare professionals	Unclear risk	Comment: no mention of provider characteristics in the two papers.