To undertake an individual patient data (IPD) meta-analysis to assess the impact of exercise-based cardiac rehabilitation (ExCR) in patients with heart failure (HF) on mortality and hospitalisation, and differential effects of ExCR according to patient characteristics: age, sex, ethnicity, New York Heart Association functional class, ischaemic aetiology, ejection fraction, and exercise capacity. Randomised trials of exercise training for at least 3 weeks compared with no exercise control with 6-month follow-up or longer, providing IPD time to event on mortality or hospitalisation (all-cause or HF-specific). IPD were combined into a single dataset. We used Cox proportional hazards models to investigate the effect of ExCR and the interactions between ExCR and participant characteristics. We used both two-stage random effects and one-stage fixed effect models. IPD were obtained from 18 trials including 3912 patients with HF with reduced ejection fraction. Compared to control, there was no statistically significant difference in pooled time to event estimates in favour of ExCR although confidence intervals (CIs) were wide [all-cause mortality: hazard ratio (HR) 0.83, 95% CI 0.67-1.04; HF-specific mortality: HR 0.84, 95% CI 0.49-1.46; all-cause hospitalisation: HR 0.90, 95% CI 0.76-1.06; and HF-specific hospitalisation: HR 0.98, 95% CI 0.72-1.35]. No strong evidence was found of differential intervention effects across patient characteristics. Exercise-based cardiac rehabilitation did not have a significant effect on the risk of mortality and hospitalisation in HF with reduced ejection fraction. However, uncertainty around effect estimates precludes drawing definitive conclusions.