The primary aim was to evaluate the effect of dapagliflozin according to QRS duration across the spectrum of left ventricular ejection fraction (LVEF), given that prolongation of QRS duration is associated with less favourable ventricular remodelling with pharmacological therapy and worse outcomes. A pooled analysis of the DAPA-HF and DELIVER trials, excluding patients with a paced rhythm and cardiac resynchronization therapy. Overall, 4008 patients had heart failure (HF) with reduced ejection fraction (HFrEF), and 5816 had HF with mildly reduced/preserved ejection fraction (HFmrEF/HFpEF). QRS duration was <120 ms in 7039 patients (71.7%), 120-149 ms in 1725 (17.6%), and ≥150 ms in 1060 patients (10.8%). The median follow-up time was 23 months. The rate of the primary composite outcome of cardiovascular death or worsening HF was 9.2 (95% confidence interval [CI] 8.7-9.7), 14.3 (13.0-15.7), and 15.9 (14.1-17.9) per 100 patient-years in the <120, 120-149, and ≥150 ms groups, respectively. This gradient in event rates was observed both in HFrEF and HFmrEF/HFpEF. Dapagliflozin, compared with placebo, reduced the risk of the primary outcome consistently across the QRS duration subgroups (hazard ratio [95% CI] 0.75 [0.67-0.85], 0.79 [0.65-0.96], and 0.89 [0.70-1.13] in the <120, 120-149, and ≥150 ms groups, respectively; p for interaction = 0.28). The effect of dapagliflozin on the primary outcome was consistent across the QRS duration regardless of HF phenotype that is, HFrEF or HFmrEF/HFpEF. Prolongation of QRS duration is associated with worse outcomes irrespective of HF phenotype. Dapagliflozin reduced the risk of the primary outcome, regardless of QRS duration, in DAPA-HF and DELIVER.