Abstract
BACKGROUND: Heart failure is often diagnosed during unplanned admission to hospital, which is associated with poor outcomes. The aim of this study was to examine the potential for early heart failure identification in primary care, trends in diagnostic practices, sociodemographic inequalities, and the association between diagnostic pathways and outcomes. METHODS: We conducted a retrospective cohort study using the Clinical Practice Research Datalink. Adults aged 18 years or older with newly diagnosed heart failure were identified from linked primary care and hospital records in England between Jan 1, 2000, and March 31, 2021. From primary care records, we analysed heart failure indicators (ie, breathlessness, ankle swelling, and loop diuretic use) recorded up to 5 years before diagnosis and diagnostic investigations (ie, natriuretic peptide tests, echocardiography, and specialist review) recorded within the previous 6 months. Trends over time and differences among sociodemographic groups are reported. Associations between diagnosis timing, investigation use, location (inpatient vs outpatient) and the primary outcome of 1-year survival were assessed, with analyses restricted to Jan 1, 2015, to Dec 31, 2019. Associations were adjusted for age, sex, ethnicity, socioeconomic status, year of diagnosis, systolic blood pressure, BMI, cholesterol, smoking, comorbidities, and prescribed drugs at diagnosis. FINDINGS: Among 412 173 new heart failure diagnoses (median age 78·0 years [IQR 69·0-85·0]), 194 175 (47·1%) were women and 217 998 (52·9%) were men. In 407 622 participants with ethnicity data, ethnicity was recorded as White for 375 808 (92·2%), south Asian for 11 644 (2·9%), Black for 6994 (1·7%), other or mixed for 3622 (0·9%), and unknown for 9554 (2·3%). Although 274 228 (66·5%) of 412 173 patients had previous heart failure indicators, diagnostic timing worsened over the study period (2000-04 to 2015-19), with a median lag time increasing from 16·4 months (IQR 1·6-45·7) to 35·4 months (6·4-54·7) and the proportion of inpatient diagnoses rising from 30 560 (33·9%) of 90 136 patients to 55 905 (46·8%) of 119 355 patients. Among 80 824 individuals with previous indicators suggestive of heart failure diagnosed between Jan 1, 2015, and Dec 31, 2019, only 10 079 (12·5%) patients underwent natriuretic peptide testing, 15 986 (19·8%) had echocardiography, 27 804 (34·4%) were referred to specialists, and 42 302 (52·3%) had no diagnostic investigations recorded in primary care. Women, individuals living in deprived quintiles, and those with multiple long-term conditions had up to five times longer delays, lower investigation rates, and a higher likelihood of hospital diagnosis. Delays (adjusted hazard ratio [HR] 1·15 [95% CI 1·10-1·20]), absence of investigations (1·89 [1·83-1·95]), and inpatient diagnosis (2·58 [2·50-2·66]) were all associated with higher mortality. Mortality was lowest among outpatients who underwent primary care investigations (571 [5·5%] of 10 469 patients) and highest among inpatients with long-term loop diuretic use but no previous investigation (5429 [33·0%] of 16 458 patients; adjusted HR 5·29 [95% CI 4·83-5·79]). INTERPRETATION: Most patients with heart failure show early signs in primary care, yet few receive timely diagnostic evaluation. Delays, missed investigations, and inpatient diagnoses are associated with poor outcomes and care inequities. FUNDING: British Heart Foundation and National Institute for Health Research.