Roman 1983.

Methods	Study design: Single centre RCT Country: Chile Dates patients recruited: June 1973 to June 1981 Maximum follow up: 9 years
Participants	Inclusion criteria: Patients with transmural AMI. Exclusion criteria: Severe arrhythmias persisting after the acute phase of AMI (frequent ventricular premature beats, grade iii‐iv of the Lown classification, atrial flutter, partial or complete A_V block);great left‐ventricular enlargement; left ventricular aneurysm; persistent cardiac failure; severe diastolic hypertension post‐myocardial infarction angina. N Randomised: total: 193; intervention: 93; comparator: 100 Diagnosis (% of pts): Transmural AMI: 100% Anterior wall infarction: 55% Posteroinferior infarction: 45% Age (mean ± SD): intervention: 56.2 ± 10.3; comparator: 59.1 ± 8.8 Percentage male: intervention: 93.6%; comparator: 87% Ethnicity: NR
Interventions	Intervention: Supervised physical training programme according to the guidelines reported by Zohman and Tobias. It was started with combined ergometric, calisthenic and walk‐jogging exercise lasting 30 min, three times a week. The intensity of the training was graded according to the target heart rate threshold, defined as 70% of maximal heart rate achieved by the patient in the former ergometric work test. Components: exercise only. Setting: centre. Aerobic exercise: Modality: combined ergometric and walk‐jogging exercise. Length of session: 30 min. Frequency: three times a week. Intensity: 70% of maximal heart rate. Resistance training included? Calisthenics. Total duration: average 42 months (range 6 to 108 months). Co‐interventions: none described. Comparator: Control patients were medically treated according to the guidelines commonly used, namely, short‐ and long‐lasting nitrites, ß‐ blockers or Ca antagonists (nifedipine). Co‐interventions: A small number (8 patients) were also treated with oral anticoagulants.
Outcomes	Mortality, MI and revascularisations.
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	“Patients were randomly allocated…”
Allocation concealment (selection bias)	Unclear risk	Allocation concealment not described.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding not described.
Incomplete outcome data (attrition bias) All outcomes	Low risk	18/93 (19.4%) and 18/100 (18%) withdrew or dropped out from intervention and control groups over the 9‐year period.
Selective reporting (reporting bias)	Low risk	Mortality, morbidity and complications were recorded over the duration of the study and are presented as rates.
Groups balanced at baseline	Low risk	“as can be observed, both groups were matched in all the characteristics that could eventually alter the late prognosis of the disease”
Intention‐to‐treat analysis conducted	Low risk	ITT not described, and no details of how missing data is handled are given, but groups appeared to be analysed according to original allocation.
Groups received same treatment (apart from the intervention)	Unclear risk	Control patients were medically treated throughout, but it appears that CR patients were only prescribed medication on the appearance of unstable angina or electrocardiographic ischemia.