Brook 2003.

Methods	Randomised trial
Participants	Patients with depression: 135 (intervention 64; control 71) Health professional (delivering intervention): 19 Practice: not clear Community pharmacy The Netherlands Year of study: April 2000 to April 2001.
Interventions	Pharmacist coaching patients and take‐home video, vs usual care Length of the intervention: not clear Number of interventions: 3 during 6 months
Outcomes	Disease control assessed by self‐rating 90‐item (Hopkins) Symptom Checklist (SCL‐90)
Notes	Required 75 patients in arm to detect 13% difference in depression at significance level of 0.05. No useable quantitative data. Funding source: Organon unconditionally sponsors International Health Foundation. The study received an unconditional grant from GlaxoSmithKline Conflict of interest: The study was carried out without interference of either of the companies.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomisation occurred at patient level and with a 1:1 ratio, using block randomisation to ensure equal numbers of intervention and control patients by pharmacy.
Allocation concealment (selection bias)	Low risk	Randomisation used "block randomization". The whole sample was randomised before delivery to the pharmacies. These forms were precoded and delivered in sealed envelopes.
Blinding of participants and personnel (performance bias) All Outcomes/Outcome 1	High risk	Quote: "Neither patients, nor pharmacists were blinded for group assignment" Unclear if this influenced intervention. Same pharmacists delivered both arms, therefore potential for contamination
Blinding of outcome assessment (detection bias) All Outcomes/Outcome 1	High risk	Subjective outcome in an unblinded trial
Incomplete outcome data (attrition bias) All outcomes	Low risk	Between group attrition < 10%.
Selective reporting (reporting bias)	Low risk	1 outcome, appropriately reported
Other bias	Low risk	None