The association between alcohol consumption and cardiovascular disease (CVD) has mostly been examined using broad endpoints or cause-specific mortality. The purpose of our study is to compare the effect of alcohol consumption in the aetiology of a range of cardiovascular disease phenotypes.

The relationship between cardiovascular diseases (CVDs) and alcohol consumption is complex. Moderate alcohol consumption has consistently been found to be associated with a reduced risk of commonly reported aggregates of CVD (e.g. coronary heart disease [CHD]) while heavy drinking (and abstaining from alcohol) is believed to be associated with an augmented risk. However, much remains to be clarified. For example, previous studies have typically been too small in size to adequately assess the impact of alcohol consumption on a range of major, pathologically distinct CVDs. Understanding the role of alcohol consumption in specific CVDs may provide novel insights into the mechanisms which give rise to the observed beneficial effects of moderate consumption for these aggregates.

As the focus of our research question is the effect of alcohol consumption on specific cardiovascular phenotypes we will not derive a single primary outcome. Utilising the strengths of the CALIBER data platform we will define multiple cardiovascular disease outcomes including: chronic stable angina (SA), unstable angina (UA), coronary heart disease not otherwise specified (CHD, NOS), acute myocardial infarction (MI), heart failure (HF), ventricular arrhythmias including cardiac arrest, abdominal aortic aneurysm (AAA), peripheral arterial disease (PAD), ischaemic and haemorrhagic (subarachnoid and intracerebral) strokes, transient ischaemic attack (TIA), and sudden cardiac death (SCD) or unheralded coronary death (UCD). Additionally non-CVD mortality will be defined as a competing outcome. These outcomes are defined/validated using multiple sources; including a combination of symptoms, diagnoses (including the use of additional information from ECG findings and troponin values) and prescriptions.

For comparison purposes (to major studies/existing consortia) we will also estimate models for aggregated CHD (MI and unheralded coronary death), CVD (CHD and stroke of any type) and fatal CVD (combination of fatal CHD and fatal CVD) endpoints.

Secondary outcomes: To examine whether the association between drinking categories and specific MI phenotypes differ we will rerun our models in patients who remained in the cohort from January 2003, when information from MINAP became available. This will allow for the composite MI category to be decomposed into ST elevation myocardial infarction (STEMI), non-ST elevation myocardial infarction (nSTEMI) and myocardial infarction not otherwise specified (MI NOS).

In light of current debates on the U/J-shaped relationship observed between alcohol consumption and aggregated CVD outcomes five drinking categories have been defined including: (1) Never drinkers ("tee-total" and "non-drinkers"), former drinkers (those with codes for "stopped drinking alcohol" and/or "ex-drinker"), occasional drinkers (those with codes for "drinks rarely" and/or "drinks occasionally"), current moderate drinkers (those who had a code for current alcohol consumer and an indicator of whether they drank within daily [4/3 UK units of alcohol for men and women respectively] and weekly [21/14 UK units for men and women respectively] recommended sensible drinking limits) and current heavy drinkers (defined as those who exceeded daily or weekly sensible drinking limits). Unfortunately information on binge drinking was only available for a select minority of the cohort (~100 people) therefore a separate category for this drinking behaviour was not defined (but these patients were coded as heavy drinkers). Our referent category will be current moderate drinkers.

Competing-risk Cox proportional hazard models will be used to examine the association between drinking category and different cardiovascular disease phenotypes. Competing-risk regression takes into account the instantaneous failure rates of specific competing-causes (i.e. different cardiovascular phenotypes and death from non-coronary causes) - not doing so may overestimate the association between drinking categories and outcome. A detailed analytic protocol is available upon request.

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), secondary care (HES) 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:

• Aged ≥ 30 years.
• Patient in a GPRD registered practice that has consented to the linkage process (who also met data quality standards).

Exclusion Criteria:

• A recorded history of any cardiovascular disease phenotype prior to entering the study.
• Cause of death unknown.

The cohort used in the present study was drawn from patients registered with GPRD general practices in England that consented to data linkage (approx. 5% of the UK population). We used an open cohort design, where participants joined the cohort when they met the inclusion criteria at any point between 1st January 1997 and 25th March 2010 (the last GPRD data submission). Patients were included in cohort if they were aged ≥ 30 years, had at least one year of electronic health record data which met GPRD data quality standards, and had no record indicating any cardiovascular disease prior to study entry. Patients were followed up until the date of an initial presentation of one of our cardiovascular endpoints or were censored on the date of leaving the practice/last data submission from their practice. Patients who died before 1st January 2001 were excluded as cause-specific mortality data was not available for them (this approach is adopted in the Prospective Studies Collaboration).