Atrial fibrillation (AF) is a condition that causes an irregular and often abnormally fast heartbeat. It is common and can cause further problems and development of other diseases including heart failure.
Whilst AF is common, it is often diagnosed too late – only after complications and hospitalisation has occurred. There are still uncertainties about who gets AF and why, and which health conditions people with AF are likely to develop later in life.
In this research we will use hospital and primary care medical records to see if we are able to predict who is likely to get AF, and to look at what happens to patients diagnosed with AF in the long term. We will study patient’s information such as age and gender, as well as their medical history to determine how these may be associated with developing AF. We will also study the pharmaceutical and surgical treatments used to treat AF and compare how different treatments can impact long term risks of further disease. The research will use traditional statistical methods as well as artificial intelligence to try and answer these questions. The findings of this research will help understand how AF is linked to other conditions and show the benefit of diagnosing and treating AF to stop further disease development.
Atrial fibrillation (AF) is a common, chronic condition that incurs substantial health-care expenditure.
Many patients with AF are diagnosed too late and once stroke has occurred. There is a fundamental knowledge gap in prediction of for whom and when new onset AF will occur; if filled this could transform the outcomes of patients with AF. Equally, little is known about the full health burden of AF – beyond that of hypothesis-driven clinical outcomes. Oral anticoagulation (OAC) can be given to patients with AF for stroke prevention, however uptake is limited. Surgical intervention can alleviate AF and associated outcomes but longitudinal studies with long-term follow up are scarce and the comparative effects of OAC are unclear.
There is a dearth in largescale population-based studies that provide high-resolution insights into the effects of AF treatment, and the risk factors and disease trajectories of patients with AF. Using CPRD GOLD and Aurum primary care data linked with ONS mortality, HES-APC and index of multiple deprivation, this study aims to investigate the clinical pathways of AF patients across three objectives:
1. To assess the risk factors of AF using cox proportional hazard regression model and survival random forest analysis, and identify multimorbidity clusters for patients with AF using hypergraphs network analysis.
2. To determine disease trajectories following the diagnosis of AF using process mining, and quantify the disease pathways amongst AF patients compared with the general population using flexible parametric relative survival models.
3. To investigate the use of OAC and surgical intervention for stroke prophylaxis in patients with AF, and quantify their association with major cardiovascular and non-cardiovascular outcomes using multivariable survival models.
This study will help target therapeutic strategies to specific patient groups and detect the high-risk individuals for AF prevention.
Cardiovascular and non-cardiovascular related hospital admission; atrial fibrillation incidence; all-cause and cause specific mortality; surgical procedures; oral anticoagulation prescriptions
Major cardiovascular outcomes including stroke; heart attack; ischemic heart disease; heart failure and all-cause mortality.
Non-cardiovascular outcomes including pneumonia; renal failure; liver disease; cancer; dementia; cause-specific mortality.
Jianhua Wu - Chief Investigator - Queen Mary University of London
Harriet Larvin - Corresponding Applicant - Queen Mary University of London
Chris Gale - Collaborator - University of Leeds
Lan Mu - Collaborator - Queen Mary University of London
Paris Baptiste - Collaborator - Queen Mary University of London
Ramesh Nadarajah - Collaborator - University of Leeds
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation