Incidence, prevalence, resource use and clinical outcomes of heart failure in England: a linked descriptive analysis of primary and secondary care (PULSE)

Date of Approval: 
2020-09-21 00:00:00
Lay Summary: 
Heart failure is a term used for a condition where the heart does not pump blood as well as it should. Previous studies estimate it affects around 1.6% of the UK population, and is more common in the elderly. The different types of heart failure can be categorised by a measure called “ejection fraction” which measures how well the heart is pushing blood into the body. Most research investigating the real world impact of heart failure in the UK has not considered the different types, because it can be difficult to distinguish between them without the exact ejection fraction measure. Our study aims to use data from primary and secondary care to group people with heart failure into the different types, using a simple set of rules that consider whether particular information appears in the patient record. We then aim to provide up-to-date estimates for the number of individuals in England with heart failure and how this breaks down by type, as well as providing insight into typical characteristics of these individuals (e.g. age, other conditions, types of medications used). We will use primary and secondary care data and data from the office for national statistics to investigate the outcomes of these patients in terms of hospital stays, interactions with their GP and survival. This study will provide important information regarding the impact of different types of heart failure on both patients’ lives and the health system, and may highlight potential inequalities and/or areas for improvement in patient care.
Technical Summary: 
Heart failure (HF) is associated with high levels of morbidity and mortality. It may present as impaired ventricular systolic or diastolic function, a.k.a. “reduced” (rEF) and “preserved” (pEF) ejection fraction (EF) respectively. Previous epidemiological studies of HF burden in the UK have either not been split by EF subtype or have used samples that may not be representative of the entire HF population. This study aims to 1. Estimate the prevalence and incidence of HF in the UK by subtype. 2. Describe HF patient characteristics by subtype. 3. Describe aspects of the burden of HF by subtype in terms of clinical and health resource outcomes. We will use linked data from CPRD and Hospital Episode Statistics (HES) to establish a cohort of patients with HF between January 2015 and June 2019. Coding systems for UK primary care allow specification of subtype, and EF may be entered for some patients. By combining these ways to distinguish subtype, we will group HF patients into reduced, preserved and unknown EF. Using additional data on the overall CPRD population, we will estimate yearly prevalence and incidence of HF by subtype, using 95% confidence intervals to quantify uncertainty. Patient characteristics by subtype will be presented descriptively. Primary outcomes of interest during the study period (HF cohort only) include hospitalization for heart failure (obtained from linked HES), HF related GP consultations, prescriptions for HF medication, and mortality (obtained via linked office for national statistics data). Recurrent event outcomes will be analysed using negative-binomial models including subtype as a covariate. Cumulative incidence functions will be used to estimate survival and other time to event outcomes overall and by subtype. Potential differences between subtypes will be quantified using the Fine and Grey model to allow for competing risks due to other cause mortality where relevant.
Health Outcomes to be Measured: 
Primary outcomes: Prevalent heart failure (HF); Incident HF; HF related hospital admission (HHF); HF related primary care consultation; Number of prescriptions issued for HF medications; cardiovascular mortality; all-cause mortality Secondary outcomes: Referral source of HF hospital admission; time to first hospitalisation for HF; time to second hospitalisation for HF; time to third hospitalisation for HF, number of GFR/serum creatinine tests in primary care; change in eGFR.
Application Number: 
20_000051
Collaborators: 

Ruth Farmer - Chief Investigator - Boehringer-Ingelheim Pharmaceuticals, Inc
Ruth Farmer - Corresponding Applicant - Boehringer-Ingelheim Pharmaceuticals, Inc
Andrew Ternouth - Collaborator - Boehringer-Ingelheim Pharmaceuticals, Inc
Gary Solomons - Collaborator - Boehringer Ingelheim Limited
Nicholas Gollop - Collaborator - Boehringer-Ingelheim - UK

Linkages: 
HES Admitted Patient Care;ONS Death Registration Data