The influence of beta‐blocker therapy (bisoprolol or carvedilol) (bB) on the prognosis of heart failure (HF) patients with diabetes mellitus (DM) is uncertain.

To assess the effect of bB on the prognosis of HF patients with new‐onset DM treated with a contemporary medical regime.

Prospective study of 5314 HF patients with previously unknown DM. Mean age was 71.8 ± 7.9 years, 53.0% were women, and 50.2% had HF with preserved ejection fraction (HFpEF). During a median follow‐up of 56.9 ± 18.2 months, 68.9% of the patients died, 88.6% were hospitalised for HF, and 1519 (27.3%) developed DM (62.3% of them received bB, 947 patients). We propensity‐matched 572 HF patients with DM on bB, with 572 HF patients with DM non‐treated with bB.

Beta‐blocker therapy was associated with a decreased hazard risk (HR) of all‐cause death [HR: 0.68, CI 95% (0.61–0.75)], mainly because of a reduced risk of death from cardiovascular causes [HR: 0.70 (0.64–0.77)] (p < 0.001). Similarly, bB was associated with a decreased HR of hospitalisation [HR: 0.82 (0.72–0.92)] (p < 0.001). Nevertheless, the 30‐day re‐admission rate and the number of visits were not significantly associated with bB. These relationships of bB with prognosis were maintained, independently of the gender, the type of HF (HFpEF ot HFdEF), the comorbidities and the medication used (p < 0.01).

Therapy with bB, bisoprolol or carvedilol, is associated with a reduced mortality and morbidity of HF patients with new‐onset DM, not only in men but also in women, as well as in patients with HFpEF or HFdEF.