Lopez Cabezas 2006.

Study characteristics
Methods	Aim of study: to assess the efficacy of multi‐factorial educational intervention carried out by a pharmacist in patients with heart failure Study design: RCT (2 hospitals; allocation: individual) Number of arms/groups: 2
Participants	Description: patient/consumer Geographic location: Spain Setting: hospital discharge Inclusion criteria: patients admitted for definite heart failure (Framingham criteria ‐ 2 major or 1 major + 2 minor criteria met simultaneously) Falces 2008 describes subgroup > 70 years Exclusion criteria: living out of the area of influence of the hospital, living in old people's home, moved to a social‐health centre or other centre for acute patients, suffering any type of dementia or psychiatric disease, refusing participation Number of participants randomised: 134 (70 and 64), subgroup > 70 years: 103 (53 vs 50) Number of participants included in analysis: 134 (70 and 64), except adherence 63 (40 vs 23), subgroup > 70 years: 82 (45 vs 37) Age: 75.3 ± 8.4 vs 76.1 ± 9.4, subgroup > 70 years: 79.0 ± 4.9 vs 80.1 ± 5.5 Gender: female: 41 (58.6%) vs 34 (53.1%), subgroup > 70 years: 60.4% vs 56% Ethnicity: not specified Number of medications: type not specified, 7.1 ± 3.0 vs 7.1 ± 2.5, subgroup > 70 years: 7.5 ± 3.1 intervention vs 7.0 ± 2.1 control Frailty/Functional impairment: New York Heart Association Functional Classification: I to II: 58 (84.1%) vs 54 (87.1%) Cognitive impairment: dementia excluded Comorbidities: total not reported
Interventions	Group 1 ‐ Active information programme: active information programme (run by a pharmacist from the research team) consisted of a personal interview at the time of discharge and subsequent telephone reinforcement. Intervention included information about the disease, diet education, and information about the medications. Simple language was used, adapted to the cultural level of patients, with support of audiovisual and written didactic materials. Monthly during first 6 months of follow‐up, and subsequently every 2 months, patients were called to reinforce the intervention and to solve doubts or problems that may have arisen Group 2 ‐ Usual care Co‐intervention: N/A Provider: pharmacist Where: in person at discharge; telephone to home When and how often: discharge, monthly follow‐up for 6 months, then 2‐monthly follow‐up for 6 months Intervention personalised: yes
Outcomes	Timing of outcome assessment: 12 months Medication adherence (objective) : pill count/tablet accountability: % of reliable patients (95% to 100% compliance) Satisfaction with intervention (subjective): Catalan Health Department satisfaction survey, asking patients about care and information received and asking them to provide a global score from 0 to 10 Health‐related quality of life (subjective): EuroQol, validated in Spanish and in Catalan Adverse clinical health outcomes (objective): hospital re‐admissions, percentage of patients with re‐admission, subgroup > 70 years: mortality Adverse clinical health outcomes (objective): number of deaths Cost‐effectiveness (objective): financial evaluation: hospitalisation costs calculated for both groups, with intervention direct costs, delivered materials, and time spent by the pharmacist added in
Notes	Trial registration: N/A Consumer involvement: not specified Funding source: Health Research Fund and European Regional Development Fund Dropout: 12 months of data for adherence only available for 40 and 23 patients, subgroup > 70 years: 20 died (7 vs 13) Language translation: yes ‐ Falces 2008 was translated to English
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Generation of the randomisation sequence was the responsibility of the clinical epidemiology unit. Randomisation lists were generated by software and in blocks of 4
Allocation concealment (selection bias)	Unclear risk	Allocation controlled by admissions department; recruitment carried out by cardiology department
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not blinded. Neither the physician nor the nurse responsible for the patient knew the allocation until the educational intervention on the day of discharge
Blinding of outcome assessment (detection bias) All outcomes	High risk	Compliance assessed by pharmacist. Pharmacist responsible for active info programme knew allocation; this could have generated contamination problems
Incomplete outcome data (attrition bias) All outcomes	High risk	No mention of reason for attrition; only 47% completed 12‐month compliance Falces paper: compliance data available for only 49 patients (59%). Methods stated that those who did not attend follow‐up were to be considered non‐compliant, but results were not presented this way. P value figure not listed
Selective reporting (reporting bias)	High risk	Reported only 'reliable patients' ‐ but compliance had 3 levels: reliable, partially reliable, not reliable. Financial evaluation not a planned outcome Falces paper: no reasons given for chosen variables in hazard ratio calculation
Other bias	Unclear risk	Sample size calculation: 67/group not reached; 3 years from final data to publication