Contemporary epidemiology of hospitalised heart failure with reduced versus preserved ejection fraction: a study of whole-population electronic health records in England

Abstract

ABSTRACT Background Heart failure (HF) is common and complex condition often associated with numerous co-existing chronic medical conditions and a high mortality. The contemporary epidemiology of hospitalised HF, particularly how it might have changed since the COVID-19 pandemic, remains incompletely characterised. Methods Using whole-population electronic health records on 57 million individuals in England, we identified patients hospitalised with HF with reduced ejection fraction (HFrEF) or preserved ejection fraction (HFpEF) from Jan 1, 2019, to Dec 31, 2022. For patients with new-onset HF, we assessed trends in all-cause and cause-specific re-hospitalisation and mortality, and dispensed guideline-recommended therapies. We estimated adjusted hazard ratios (HR) to compare HFrEF versus HFpEF for re-hospitalisation and mortality outcomes, including by co-existing chronic medical conditions. We computed population-attributable fractions to estimate percentages of re-hospitalisations and deaths attributable to specific co-existing chronic medical conditions. Findings Among 233,320 patients with HF, 101,320 (43·4%) had HFrEF, 71,910 (30·8%) had HFpEF, and 60,090 (25·8%) had unknown classification. Over the study period, there were modest declines in 30-day and one-year all-cause re-hospitalisations, most pronounced for 30-day re-hospitalisation (annual change in incidence between 2019 and 2022, -6·2% [95% CI -10·5% to -1·6%] for HFrEF; -4·8% [95% CI -9·2% to -0·2%] for HFpEF). There were no overall changes in all-cause mortality. Compared with HFrEF, patients with HFpEF had higher rates of re-hospitalisations for any cause (HR 1·20 [95% CI 1·18–1·22]), and higher overall mortality (HR 1·07 [95% CI 1·05–1·09]) driven by non-cardiovascular causes (HR 1·25 [95% CI 1·21–1·29]). Rates of re-hospitalisation and mortality were highest in patients with chronic kidney disease (CKD), chronic obstructive pulmonary disease, dementia, and liver disease. Overall, 5·8% (95% CI 5·1% to 6·4%) of re-hospitalisations and 13·5% (95% CI 12·3% to 14·7%) of deaths were attributable to CKD, double that of any other condition. There was swift implementation of newer guideline-recommended therapies, but markedly lower dispensing of these medications in patients with co-existing CKD. Interpretation Rates of re-hospitalisation in HF patients in England have decreased in recent years. Further population health improvements could be achieved through enhanced implementation of guideline-recommended therapies, particularly in patients with co-existing CKD, who, despite their high risk, remain undertreated.