The primary aim of this study is to examine if long-term patterns of alcohol consumption are associated with time-to-onset for incident coronary heart disease (fatal and non-fatal), using data from multiple cohorts.

The relationship between alcohol consumption and coronary heart disease (CHD) remains an issue of debate. By capturing drinking trajectories over time, we may be equipped to obtain new insights into this relationship. Studies have shown that such trajectories have differential associations for intermediate traits (carotid intima-media thickness, pulse wave velocity and inflammatory markers), but no studies exist that link stability of drinking to actual CHD. The current study will employ a longitudinal cohort design to evaluate the association between long-term alcohol consumption trajectories and time-to-event for CHD. Data will be drawn from six cohorts (five British, one French). The combined participant pool comprises 64,926 individuals (58% male, individual cohort sizes ranging from 1,444 to 25,636 participants); those with a baseline history of CHD will be excluded. Repeat alcohol intake measurements across a 10-year interval will be the exposure, with participants' intake trajectory defined according to their alcohol consumption volume and its consistency over time. To account for heterogeneity across cohorts, individual participant data meta-analysis methods will be employed in determining CHD diagnosis rates and hazard ratios for the different intake trajectories, with adjustment for relevant demographic and clinical characteristics. Results from the modelling work illustrating the form and magnitude of the association between the alcohol intake categories and CHD will be presented.

This work will help further understanding of the role that alcohol intake and its stability over time play in subsequent CHD risk, and will have implications for our understanding of alcohol's relationship to cardiovascular health in the general population.

Inclusion Criteria:

• Participated in cohort studies being investigated
• Provided alcohol intake data at least one time point
• Was included in coronary heart disease onset tracking

Exclusion Criteria:

• Left cohort study prior to completion of 10 year exposure window
• Had history of coronary heart disease before end of 10 year exposure window

The participants will be drawn from birth, regional and occupational cohort studies in England, Wales, Scotland and France.

Researchers can request access to anonymised data. The data are already available to bona fide researchers via application: European Prospective Investigation of Cancer-Norfolk (EPIC-Norfolk; http://www.srl.cam.ac.uk/epic/contact/), Gaz et Electricité (GAZEL; http://www.gazel.inserm.fr/en/projects/submitting-a-project.html), Medical Research Council National Survey of Health and Development 1946 (NSHD; http://www.nshd.mrc.ac.uk/data/), West of Scotland Twenty-07 Study (T-07-1930s and T-07-1950s; http://2007study.sphsu.mrc.ac.uk/Information-on-data-sharing.html), and Whitehall II (WII; http://www.ucl.ac.uk/whitehallII/data-sharing).