Heart failure is a condition where the heart does not pump enough blood to meet the needs of the body. This is due to muscular damage, poor valve function, abnormal heart rhythm or other rare causes. Acute heart failure is a common cause of admission to hospital (over 67,000 admissions in England and Wales per year) and is the leading cause of hospital admission in people 65 years or older in the UK. It is also a leading contributing cause of death. Current practice does not include the use of clinical predictive tools to aid early identification of individuals with heart failure who are at risk of death in the long-term. As a result, patients with heart failure who may be at an increased risk of death cannot be identified and managed effectively.
Several factors increase the risk of developing heart failure and may also increase the risk of death in individuals with heart failure, Obesity continues to increase in prevalence both in the UK as well across the world. Although obesity has been shown to increase the risk of heart failure, the evidence surrounding this is conflicting.
Current evidence also shows that there is an Obesity Paradox. The Obesity Paradox implies that despite increased risks of heart failure as BMI increases, once heart failure is established, those with more severe obesity survive longer.
By developing and validating a prognostic model to predict the risk of death in patients with heart failure, we aim to help clinicians better determine risk.
Background: Existing evidence indicates that there are a range of factors that predispose a patient with heart failure to mortality, with BMI being a poorly understood risk factor. Adding to this, in current practise, there are no prognostic models used to predict the long-term risk of mortality in patients with heart failure.
Aim: Develop and validate prognostic models for mortality in patients with heart failure.
Design: Retrospective open cohort study
Setting: General practices in UK providing data to the CPRD database. Cohort design.
Participants: Adult with new diagnosis of heart failure identified from practices with HES linkage, registered for at least one year before the study start date (date of heart failure diagnosis)
Outcomes: All cause-mortality
Methods: Multivariable cox regression model will be used to create a 45 year prognostic model for all-cause mortality in patients with heart failure, at any point during the study period. This will be compared with a gradient boosting machine learning model for the same outcome.
Outputs: Two prognostic models for heart failure mortality. One through statistical methods and another through data-driven methods. Determination of the effect of obesity on heart failure prognosis.
All-cause mortality, positive predictive value, negative predictive value, specificity, sensitivity, accuracy
Stephen Weng - Chief Investigator - University of Nottingham
Habibullah Muhammad-Kamal - Corresponding Applicant - University of Nottingham
Barbara Iyen - Collaborator - University of Nottingham
Joe Kai - Collaborator - University of Nottingham
Nadeem Qureshi - Collaborator - University of Nottingham
HES Admitted Patient Care;ONS Death Registration Data;Practice Level Index of Multiple Deprivation