Reporter brief

Systematic review of 107 randomised trials (26,886 adults with coronary heart disease) comparing exercise-based cardiac rehabilitation versus no exercise control. At short-term follow-up (6-12 months), exercise reduced myocardial infarction (RR 0.72), all-cause hospitalisation (RR 0.65), and likely reduced all-cause mortality (RR 0.86). Health-related quality of life improved across multiple domains, and eight economic studies confirmed cost-effectiveness.

Assessment pending; this report does not establish a change in evidence.

• —Limitations have not been recorded; consult the primary source.

Meta analysis; Human.

Adults with coronary heart disease (CHD), mostly post-MI or post-revascularisation; mean age 47–81 years. 78% of trials included women, who represented 17% of total participants overall.

Design

Meta analysis

Sample

26,886

Journal

The Cochrane Database of Systematic Reviews

Published

Sep 18, 2026

Short-term follow-up (6–12 months): All-cause mortality RR 0.86 (95% CI 0.74–1.00, moderate-certainty); cardiovascular mortality RR 0.87 (95% CI 0.68–1.11, moderate-certainty); MI RR 0.72 (95% CI 0.54–0.95, high-certainty); CABG RR 1.00 (95% CI 0.78–1.28, high-certainty); PCI RR 0.86 (95% CI 0.65–1.13, moderate-certainty); all-cause hospitalisation RR 0.65 (95% CI 0.51–0.82, moderate-certainty); cardiovascular hospitalisation RR 0.80 (95% CI 0.35–1.82, low-certainty). Long-term follow-up (>12 months): exercise-based CR associated with reduced cardiovascular mortality and MI. HRQoL showed clinically meaningful increases across SF-36 subscales and EQ-5D scores up to 12 months. Eight economic evaluation studies showed exercise-based CR cost-effective.

• —Assessment: Superseded.
• —Only the source material recorded on this evidence page was assessed.

Funding support from University of Glasgow, UK (for GD and RST) and Radboud University Medical Centre, NL (for FdV, DT and NS). Past NIHR Cochrane Infrastructure funding to Cochrane Heart. · Not reported.

See the Exercise timeline and recorded assessment for prior evidence.