Development and implementation of a heart failure dashboard for England: nationwide analysis of 1.4 million primary and secondary health care records

Abstract Introduction: Heart failure (HF) represents a major health and economic burden to the UK National Health Service (NHS), with patients frequently presenting in the secondary care setting for index diagnosis and treatment. Purpose: To use nationwide health records data from primary and secondary care to provide interactive summaries and visualisation within a standardised dashboard that would enable higher resolution local and national level insights for care and outcomes planning. Methods: Hospital Episode Statistics (HES) data were presented alongside with Quality and Outcomes Framework (QOF) data from NHS England. Elective and non-elective inpatient admissions for patients without a record of age less than 18, where HF was a recorded diagnosis were interrogated over the four most recent data years. Outpatient appointments were counted if the patient had any diagnosis of HF in the data. A HF HES diagnosis was defined by ICD-10 codes (HF: I50.0, I50.1, I50.9, I11.0, I13.0, I13.2 and cardiomyopathy: I25.5, I42.0-I42.9, I43.0-I43.2, I43.8, I51.5, O90.3). An interactive dashboard was developed to present admission rates, length of stay (LOS), cost, readmissions, in-hospital mortality and the patient’s deprivation scores from the Indices of Multiple Deprivation (IMD) from HES. QOF data was included for HF prevalence and quality metrics. Results: Data were drawn from 672,409 patients identified in primary care and 750,230 patients in the hospital setting with HF between 01/04/23 to 31/03/24. The ratio of primary care to hospital HF identification was 89.6%. Acute HF admissions cost the NHS over £500million in 2023/24, and 19.4% of patients hospitalised had ≥1 prior emergency HF admission in the previous 365 days. Of the comorbidities analysed, the most common were hypertension (79% of patients), atrial fibrillation (55%), and ischaemic heart disease (53%). Analysis by IMD found that deprivation impacted acute HF admission numbers, with numerical differences seen between the 2 most and least deprived deciles (21% of acute HF admissions vs. 18% respectively). Subnational analysis demonstrated geographic variation. ICB acute admission rates ranged from 9 to 26 admissions per 100 QOF patients, and the average LOS ranged from 7.7 to 12.3 days per admission (Figure 1). The dashboard was operationalised to enable users to query the data based on a range of variables including geography, admission type, diagnosis position (primary or any) and patient demographics (Figure 2). Conclusions: The number of HF cases in primary and secondary care in England is high, with evidence for under-recording in primary care. Comorbidities were frequent, highlighting the complex health needs of the population. Variation at subnational level suggests patient needs and healthcare delivery is impacted by geography and demographics. Utilising data to understand the local HF population could help healthcare organisations optimise care and improve outcomes for individuals with HF.Figure 1:ICB Regional Variation Figure 2:Patient demographics