Volume 2001.

Study characteristics
Methods	Aim of study: to describe changes in patients' adherence to therapy regimens, patients' expectations of the care they receive from their pharmacist, patients' satisfaction with pharmacy services, and patients' HRQoL after provision of pharmaceutical care Study design: cluster‐RCT (16 pharmacies; cluster unit of allocation: pharmacy) Number of arms/groups: 2
Participants	Description: patient/consumer Geographic location: Canada Setting: community pharmacy Inclusion criteria: coverage of medications under Alberta Health & Wellness senior drug benefit plan (age ≥ 65 years), ≥ 3 medications according to dispensing records, able to complete telephone interviews, residing in Alberta for 12 of 15 study months, agree to get prescriptions from study pharmacy (Kassam 2001) Exclusion criteria: communication and language barriers, terminal disease, unable to provide informed consent Number of participants randomised: 16 pharmacies (8 vs 8), 363 patients (159 vs 204) Number of participants included in analysis: 292 completed Age: 73.89 ± 6.09 vs 73.18 ± 6.11 Gender: female: 63.5% vs 69.6% Ethnicity: not specified Number of medications: prescription: 4.67 ± 2.82 vs 3.90 ± 2.49. Non‐prescription: 0.63 ± 0.92 vs 0.73 ± 1.17 Frailty/Functional impairment: not specified Cognitive impairment: not specified Comorbidities: not specified
Interventions	Group 1 ‐ Pharmaceutical care research and education project (PREP): comprehensive pharmaceutical care. Treatment pharmacists (enrolled in an intensive education programme) used the Pharmacist's Management of Drug‐Related Problems (PMDRP) instrument to summarise information collected during patient interview and used SOAP (subjective, objective, assessment, and plan) record to document actions and follow‐ups Group 2 ‐ Usual care: control pharmacists not told which patients had or had not agreed to participate Co‐intervention: N/A Provider: pharmacists (community) Where: community pharmacies When and how often: 1 interview with frequent follow‐up Intervention personalised: yes ‐ based on interview and required pharmaceutical care
Outcomes	Timing of outcome assessment: baseline and 12 to 13 months Medication adherence (subjective) : Morisky self‐reported adherence: adherence to medication regimens assessed using a 4‐item self‐report Morisky (validated) measure. Summing numerical values for each answer provides a summary adherence score ranging from 0 to 4, with lower scores indicating better adherence Satisfaction with intervention (subjective): patient satisfaction with pharmacy services using 34‐item instrument based on work by MacKeidan and Larson and Johnson et al; 7‐point Likert scale used, with 1 indicating highest satisfaction. Only general satisfaction extracted Health‐related quality of life (subjective): SF‐36 health survey, validated
Notes	Trial registration: N/A Consumer involvement: not specified Funding source: Alberta Ministry of Health, Alberta Pharmaceutical Association, University of Alberta Central Research Fund, Merck Frosst Canada, Hoechst Marion Roussel, Alberta Health‐Health Services Research Innovation Fund; drug references for pharmacists were provided by Bristol‐Myers Squibb Dropout: unclear: 5 treatment and 7 control pharmacies supplied patient data; 3 treatment and 1 control never recruited patients ICC not reported, numbers of participants in intervention and control groups at follow‐up unclear. Trial authors contacted for more information but no response. Thus unit of analysis error exists
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Pharmacies paired based on Statistics Canada median income for first 3 digits of pharmacy's postal code. Study statistician did not know the identity of pharmacies and randomly assigned pharmacies to treatment or control within the pair. One pair were very closely located, thus assigned to same treatment/control to minimise contamination. Unclear why groups still balanced (8 and 8)
Allocation concealment (selection bias)	Unclear risk	Statistician did not know identity of pharmacies but took steps to minimise contamination and match characteristics
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not possible to blind patients to intervention; control pharmacists not told which patients had or had not agreed to participate
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Telephone survey by Population Research Lab at the University ‐ unclear whether blinded
Incomplete outcome data (attrition bias) All outcomes	High risk	Only patients with data at all 3 time points were included in the analysis; T1 = 363, T2 = 292
Selective reporting (reporting bias)	Low risk	Appears to be presented as per methods
Other bias	Unclear risk	"We intended that each pharmacy would identify enough to produce sample of 50 participants" ‐ not reached, sample size based on HRQoL Recruitment bias (selective recruitment of cluster participants): high risk, patient recruitment occurred after allocation of clusters, resulting in potential for recruitment bias; "it is possible that pharmacists selected patients whom they believed would benefit from the intervention or who had a more positive attitude toward pharmaceutical care"