Abstract Aims Cardiac resynchronization therapy (CRT) improves outcomes in heart failure (HF) but may be underutilized. The reasons are unknown. Methods and results We linked the Swedish Heart Failure Registry to national registries with ICD-10 (International Classification of Diseases-10th Revision) co-morbidity diagnoses and demographic and socio-economic data. In patients with EF ≤39% and NYHA II–IV, we assessed prevalence of CRT indication and CRT use. In those with CRT indication, we assessed the association between 37 potential baseline covariates and CRT non-use using multivariable generalized estimating equation (GEE) models. Of 12 807 patients (mean age 71 ± 12 years, 28% female), 841 (7%) had CRT, 3094 (24%) had an indication for but non-use of CRT, and 8872 (69%) had no indication. Important variables independently associated with CRT non-use were: HF duration <6 months [risk ratio (RR) 1.21, 95% confidence interval (CI) 1.17–1.24]; non-cardiology planned follow-up (RR 1.14, 95% CI 1.09–1.18); age >75 years (RR 1.13, 95% CI 1.09–1.18); non-cardiology care at baseline (RR 1.10, 95% CI 1.07–1.14); small-town non-university centre (RR 1.08, 95% CI 1.05–1.12); female sex (RR 1.07 95% CI 1.03–1.10) (all P < 0.05); as was absence of AF, living alone; psychiatric diagnosis; smoking; and non-use of HF drugs. Education, income, cancer, or HF characteristics were not independently associated with CRT non-use. Conclusion In this population-wide HF registry, CRT was underutilized. Non-use was associated mostly with demographic and organizational, but not clinical or socio-economic factors. This calls for programmes to raise awareness of CRT indications and improve access and referrals to cardiology specialists.