We aim to investigate the prognosis of patients diagnosed with AF, particularly in relation to the development of subsequent stroke, heart failure, and myocardial infarction. We will explore the relationship between these outcomes and a range of risk factors.

The development of stroke in AF patients continues to be an area of substantial research focus. However, comparatively little research has investigated the extent to which HF and MI also make a substantial contribution to morbidity and mortality in this patient group, and whether there is overlap in the prognostic factors associated development of stroke, HF, and MI.

Conen et al. demonstrated that mortality risk in AF patients is partly mediated by the development of non-fatal stroke, HF, and MI. However, they did not investigate differences in the cumulative incidence of these conditions between different patient groups (e.g. men and women), or the relationship between potential prognostic factors and the development of these conditions. Sets of prognostic factors for stroke and HF in AF patients have been defined through the development of prognostic models, but these models were developed specifically for each condition so it is unclear whether these prognostic factors are associated with increased risk of a particular condition, or simply any major adverse cardiovascular event. Additionally, some potentially important prognostic factors were not evaluated in these studies (e.g. anaemia and kidney failure).

Thus we chose to conduct an exploratory study of prognostic factors for HF, MI, and stroke in patients diagnosed with AF. We selected our candidate factors from those that have previously been associated with stroke, HF, or MI (in AF patients or the general population). Identification of prognostic factors for stroke, HF, and MI in those diagnosed with AF is a first step toward understanding both the development of these conditions, and the scope for targeting preventive treatments to improve prognosis.

This study will be undertaken using linked electronic health record data for primary and secondary care from CALIBER. This data set contains a broad range of clinically relevant, clinically conducted measurements of potential prognostic factors, and also provides a very large baseline sample from which we can draw a sufficient number of incident AF cases to investigate our three endpoints.

The study has two aims. First, to determine the cumulative incidence of fatal and non-fatal heart failure (HF), myocardial infarction (MI) and stroke (ischaemic, haemorrhagic, and NOS) in patients diagnosed with atrial fibrillation (AF). Differences between clinically relevant groups (e.g. men and women) will be explored. Second, to compare the direction and magnitude of associations between prognostic factors and the development of these conditions (HF, stroke, MI) in patients with AF. The following panels of prognostic factors will be investigated: sociodemographic; anthropomorphic and haemodynamic; behavioural; co-existing conditions (cardiovascular and non-cardiovascular); blood biomarkers; secondary preventive drugs.

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. The central theme of the CALIBER research is linkage of the Myocardial Ischaemia National Audit Project (MINAP) with primary care (GPRD) and other resources. The overarching aim of CALIBER is to better understand the aetiology and prognosis of specific coronary phenotypes across a range of causal domains, particularly where electronic records provide a contribution beyond traditional studies. 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 atrial fibrillation in their primary or secondary care record.
• Patients in GPRD practices which are deemed "up to standard" by GPRD criteria
• Patients whose records are deemed "acceptable" by GPRD criteria and contain at least one year of data
• Patients whose age and sex, as recorded in GPRD is the same as that recorded in HES.

Exclusion Criteria:

• A diagnosis of heart failure, stroke, or myocardial infarction occurring before diagnosis of atrial fibrillation

Adults aged 30 years and over, registered with a CPRD practice that is up-to-standard, and having a minimum of one year of validated follow-up data. The index date is 1st January 1998 with follow-up ending on 26th March 2010, which corresponds to the administrative censoring date of the CPRD component of the CALIBER dataset.