In both the UK and the USA, Heart Failure (HF) is a common cause of hospitalisation, often prolongs hospital stay and is associated with an increased risk of death. Most patients who die from cardiovascular disease will first develop HF. Substantial differences in health care use and money spent on healthcare are known to exist between the United Kingdom (UK) and the United States of America (USA) but it is not clear whether this difference translates into differences in prognosis. We propose to evaluate the long outcomes of patients with HF in the UK compared to the USA. How well Implantable Cardioverter Defibrillators (ICD) and Cardiac Resynchronisation Therapy (CRT) work in HF have been established only in highly selected patients in narrowly defined inclusion criteria and broad exclusion criteria in prospective randomized controlled clinical trials. How well they work in the broader patient population with HF is unclear. Comparing practice patterns and clinical outcomes in 'real world' patients receiving ICD and CRT across different health systems with widely differing practice might help clarify benefits and improve patient management in both countries.
As a national sample of current practice, we will use linked CPRD GOLD data with HES and ONS to undertake a cohort study to provide an accurate estimate of the number of people with HF in the UK. Data from the US will be used sourced from US health care administrative database. All-cause mortality and composite of heart failure hospitalisation and mortality will be determined using Cox proportional hazard models. Overall differences in the event rates for pre-specified groups will be assessed using Kaplan- Meier survival curves, and log rank tests. Multivariable Cox proportional hazards regression analysis will be used to estimate differences after adjusting for covariates (Chronic kidney disease, chronic obstructive pulmonary disease, diabetes mellitus, hypertension, coronary artery disease, atrial fibrillation, previous stroke, malignancy, obesity and depression). Each cohort (UK and USA) will be analysed separately. On completion of the analysis, pooled estimates will be compared across the datasets. The information obtained from this study will be used to create prognostic models within each data set (and also in key subgroups of interest in each data-set).
All-cause mortality; Composite of heart failure hospitalisation and all-cause mortality; Composite of all hospitalizations (by diagnostic code and urgency) and mortality (by cause recorded on death certificates).
Jennifer Quint - Chief Investigator - Imperial College London
Varun Sundaram - Corresponding Applicant - Imperial College London
HES Accident and Emergency;HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation