Aims Heart failure (HF) is an important clinical problem. Expert consensus has defined HF as a primary care-sensitive condition for which the risk of unplanned admissions may be reduced by high quality primary care, but there is little supporting evidence. We analysed time trends in HF admission rates in England and risk and protective factors for admission.Methods and results We used Hospital Episodes Statistics to produce indirectly standardized HF admission counts by general practice for 2004–2011. Clustered negative binomial regression analysis produced admission risk ratios and assessed the significance of potential explanatory covariates. These included population factors (deprivation; HF, coronary heart disease, and smoking prevalence), primary care resourcing [access; general practitioner (GP) supply], and primary care quality (‘Quality and Outcomes Framework’ indicator.) There were 327 756 HF admissions of patients registered with 8405 practices over the study period. There was a significant reduction in admissions over time, from 6.96/100 000 in 2004 to 5.60/100 000 in 2010 (P < 0.001). Deprivation and HF prevalence were risk factors for admission. GP supply and access protected against admission. However, these effects were small and did not explain the large and highly significant annual trend in falling admission rates.Conclusions The observed fall in admissions over time cannot be explained by the primary care covariates we included. This analysis suggests that the potential for further significant reduction in emergency HF admissions by improving clinical quality of primary care (as currently measured) may be limited. Further work is required to identify the reasons for the reduction in admissions.