The main objectives of this study are i) to assess how heart failure was captured accross different linked electronic health record sources within the CALIBER program and the overlap between primary care, hospital admissions and/or the national mortality register, and ii) to assess risk factors, heart failure treatment and survival in patients, stratified by EHR source.

Heart failure (HF) is one of the leading causes of hospital admissions and mortality in modern healthcare systems. It can be viewed as a collective clinical syndrome of many signs and symptoms and is frequently the common endpoint of various heart diseases. Often, it is not diagnosed until it has reached a level whereby quality of life is significantly, and often irreversibly, impaired. Even though vast quantities of National Health Service (NHS) data concerning patients with heart failure are recorded, there are limited 'real world' longitudinal insights about the prognosis and consequences of HF. Although linked electronic health records cohorts such as the CALIBER program become increasingly available, for heart failure the overlap, risk factors and subsequent mortality have not been compared. Previous studies on heart failure using EHR sources have used ICD-9 or 10 codes for the identification of heart failure cases and the prevalence estimates of risk factors and comorbidity. Furthermore, the assessment of supporting information for heart failure present in electronic healthcare registries remains largely unknown. Currently, heart failure is typically inferred based on previous reports or the prescription of heart failure related medication. To strengthen heart failure case ascertainment in large electronic healthcare registries, linkages with primary care data such as what is done in CALIBER could allow more detailed insight in medical history, clinical diagnoses, anthropometric measures, health behaviour, laboratory tests, medical procedures and prescriptions.

In this study, the investigators assessed the distribution of recording, supportive medical information for heart failure diagnosis, risk factors and subsequent mortality of heart failure patients captured in linked EHR data from primary care, hospital admissions and/or death registry.

This study is part of the CALIBER (Cardiovascular disease research using linked bespoke studies and electronic records) programme funded over 5 years from the NIHR and Wellcome Trust. CALIBER has received both Ethics approval (ref 09/H0810/16) and ECC approval (ref ECC 2-06(b)/2009 CALIBER dataset).

Inclusion Criteria:

• Patients with a coded diagnosis for heart failure in primary care or hospital admission
• Patients with a coded diagnosis for heart failure as a cause of death in the national mortality register
• Patients in CPRD practices which are deemed "up to standard" for more than 1 year by CPRD criteria.
• Patients with at least one year of follow-up in the CPRD practice CPRD.

Exclusion Criteria:

• past medical history of heart failure

From 1 January 1997, all patients aged ≥18 years old, registered in CPRD practices in England consenting to data linkage, with at least one year of up-to-standard pre-study follow-up are potentially eligible.