Methods	RCT parallel group
Participants	n = 90 pts in Home-based CR group; n = 90 pts in Centre-based CR group; 100% MI; mean age 69 (SD 1.6); 71% male Inclusion: age >45 y, MI Exclusion: Severe cognitive impairment; physical disability; left ventricular ejection fraction <35%; contraindications to vigorous exercise; eligibility for myocardial revascularisation, living too far from CR unit
Interventions	Home-based CR (intervention): Exercise: Overall duration; 8-wks; Frequency: 3 days/wk; Duration: 1-hour; Intensity 7085% peak HR; Modality: cycle ergometer. Physical therapist home visits every other week Other: monthly family-oriented support groups Centre-based CR (control): Exercise: Overall duration; 12-week programme; Frequency: 3 days/wk; Duration: not reported; Intensity 70-85% peak HR; Modality: cycle ergometer. Trans-telephonic ECG monitoring during exercise Other: risk factor management counselling; support group meetings
Outcomes	Primary: Total work capacity (TWC) Secondary: Health-related quality of life (Sickness Impact Profile, SIP), mortality, morbidity (cardiovascular events), healthcare utilisation (medical visits, rehospitalisations), costs & adherence (number of completed training sessions)
Follow up	2, 8 & 14 months post randomisation
Subgroup analyses	Subgroup analysis in age groups (middle-aged 45-65 years, old 65-75 years, very old >75 years)
Country & settings	Italy, single hospital centre
Notes	Data presented separately for 3 age groups. Follow up data on charts only; authors contacted for numerical data at follow up & these have been supplied for TWC & SIP separately for 3 groups; data pooled across age groups by reviewers
Risk of bias
Bias	Authors’ judgement	Support for judgement
Adequate sequence generation?	Unclear risk	Not reported
Allocation concealment?	Unclear risk	Not reported
Blinding? All outcomes	Low risk	“Testing personnel were blinded to patient assignment.”
Incomplete outcome data addressed? All outcomes	Low risk	“we performed a sensitivity analysis comparing results obtained with and without replacement of missing data with data obtained with the expectation-maximization imputation method. Because the 2 analyses provided similar results, which were also similar with missing data substituted with data estimated in a worst-case scenario, only the data from patients who completed the study are presented
Free of selective reporting?	Low risk	All outcomes described in the methods section are reported in the results
Groups balanced at baseline?	Low risk	“baseline sociodemographic and clinical characteristics were similar across the 3 arms of the trial” Baseline characteristics by home and hospital group allocation not reported in tabular format
Intention to treat analysis?	Low risk	Although term ‘intention to treat analysis’ was not stated directly, on the basis of CONSORT diagram presented, the groups did appear to be analysed according to original random allocation
Groups received same intervention?	Low risk