Aims The aim was to assess the effect of a telemonitoring programme vs. standard care (SC) in preventing all‐cause deaths or unplanned hospitalisations in heart failure (HF) at 18 months. Methods and results OSICAT was a randomised, multicentre, open‐label French study in 937 patients hospitalised for acute HF ≤12 months before inclusion. Patients were randomised to telemonitoring (daily body weight measurement, daily recording of HF symptoms, and personalised education) (n = 482) or to SC (n = 455). Mean ± standard deviation number of events for the primary outcome was 1.30 ± 1.85 for telemonitoring and 1.46 ± 1.98 for SC rate ratio 0.97, 95% confidence interval (CI) 0.77–1.23; P = 0.80. In New York Heart Association (NYHA) class III or IV HF, median time to all‐cause death or first unplanned hospitalisation was 82 days in the telemonitoring group and 67 days in the SC group (P = 0.03). After adjustment for known predictive factors, telemonitoring was associated with a 21% relative risk reduction in first unplanned hospitalisation for HF hazard ratio (HR) 0.79, 95% CI 0.62–0.99; P = 0.044); the relative risk reduction was 29% in patients with NYHA class III or IV HF (HR 0.71, 95% CI 0.53–0.95; P = 0.02), 38% in socially isolated patients (HR 0.62, 95% CI 0.39–0.98; P = 0.043), and 37% in patients who were ≥70% adherent to body weight measurement (HR 0.63, 95% CI 0.45–0.88; P = 0.006). Conclusion Telemonitoring did not result in a significantly lower rate of all‐cause deaths or unplanned hospitalisations in HF patients. The pre‐specified subgroup results suggest the telemonitoring approach improves clinical outcomes in selected populations but need further confirmation.