. 2017 Apr 28;2017(4):CD002902. doi: 10.1002/14651858.CD002902.pub4

Burell 1996a.

*Study characteristics*
Methods	Design: multicentre RCT. Country: Sweden. Dates participants recruited: NR. Participants recruited (number of sites): hospital (14). Maximum follow‐up: 6.5 years. Follow‐up schedule: 6.5 years.
Participants	Inclusion criteria: CABG 3‐12 months prior to recruitment, non‐smokers. Exclusion criteria: diabetes mellitus, other somatic or psychiatric disease or alcoholism, and non‐Swedish speakers. Indication (% participants): CABG (100%). Psychopathology: NR. Number randomised: total: 261; intervention: 128; comparator: 133. Age (mean ± SD): total: 57.5 (SD NR) years; intervention: NR; comparator: NR. Men: total: 86%; intervention: NR; comparator: NR. Ethnicity (% white): NR.
Interventions	INTERVENTION: initial treatment (6 sessions) focused on education about CHD, surgical issues, risk factors and risk behaviours, psychological factors that influence wellbeing and Type A behaviour. From the first session, participants were given homework assignments related to observation of health behaviours. The remaining session focused on modifying Type A prone behaviours: developing and applying new reactions and behaviours that entailed less impatience, irritation, hostility, depression, and distress. Treatment targets: risk education, disease adjustment, and coronary prone behaviours (Type A behaviour, depressive reactions, anxiety). Components: risk information, guidance on behaviour change, self‐awareness/monitoring, relaxation, homework. Treatment setting (number of sites): NR (NR). Modality (group size): group (5‐9 participants). Dose: length of session: 3 hours; frequency/number of sessions: every third week/17; total duration: 51 contact hours over 1 year. Delivered by: cardiologist and nutritionist (1 session) and clinical psychologist (remainder of sessions). Follow‐up further reinforcement: 5 or 6 booster sessions in years 2 and 3. Cointerventions: access to rehabilitation programmes that were part of usual care. COMPARATOR: usual care, including access to rehabilitation programmes that were regularly offered by participating hospitals. Cointerventions: NR.
Outcomes	Total mortality. Cardiac mortality. Non‐fatal MI. Revascularisation (CABG (reoperation) and PTCA). Self‐reported Type A behaviour.
Source of funding	NR.
Conflicts of interest	NR.
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Stated participants were randomly assigned.
Allocation concealment (selection bias)	Unclear risk	Not described.
Blinding of outcome assessment (detection bias)	Unclear risk	Not described.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Insufficient information to judge, attrition appeared to be zero.
Selective reporting (reporting bias)	Unclear risk	Method did not fully specify the measures used.
Groups balanced at baseline	High risk	Control participants were significantly younger than those in the intervention group.
Intention‐to‐treat analysis	Unclear risk	Intention‐to‐treat analysis was not described, and no n values were provided in Table 2.
Groups received same cointerventions	Low risk	"Both experimental and control patients had access to rehabilitation programmes that were regularly offered by their hospitals."