People report feeling sad and low (depression) or worried (anxiety) appear more likely to subsequently suffer a heart attack, or angina. However it is not known whether depression or anxiety actually causes heart disease. If these mental health problems and heart disease were cause and effect this has important implications for world health. Previous research on this topic has had several limitations. First, most studies have studied heart disease as if it were one thing. There is a need for studies which distinguish different types of heart disease (e.g. different types of heart attack, angina) which may be linked to mental health problems in different ways. Second, it is not clear whether symptoms of heart disease come before the depression or anxiety or the other way round? Much of the available research cannot look at this in detail because they rely on data from occasional snapshots of study populations rather than a continuous record. The investigators propose to use the linkage of the national registry of coronary events to general practice records in the GPRD, which will allow us to address these limitations. The investigators research will help us understand better whether mental health problems cause the onset of different types of coronary disease.

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).

Overall aim

To elucidate the role of depression and anxiety in the aetiology and prognosis of specific acute and chronic coronary disease phenotypes.

Background

Coronary disease and depression are major causes of mortality and morbidity globally and so the public health impact of a causal association between the two is profound1. But our meta-analysis of 22 aetiologic cohort studies (4016 events) the presence of depression was associated with a 70% increased risk of CHD events. This systematic review, and others, identified several limitations of existing research, including a) clinical phenotype resolution: broad aggregates of CHD, rather than specific coronary phenotypes, b) small size, with insufficient event numbers to compare risks in women and men, c) temporal resolution: Single time point of exposure to depression (usually prevalent); unclear temporal relation between the onset of symptoms of depression and the onset of symptoms associated with coronary disease and lack of studies assessing whether exposure to depression prior to the onset of any symptomatic coronary disease influences the prognosis (future progression) of coronary disease once established, d) incomplete and inconsistent assessment of potential confounders or mediators: unclear role of social deprivation, smoking, alcohol and other behaviours which may confound the association; unclear role of detection, treatment and control of cardiovascular risk among patients with depression and lack of consideration of co-existing psychiatric morbidities.

Study design

Observational study, using cohort and case series analyses in initially healthy populations, and among patients with specific manifestations of coronary disease.

Sample size calculations

We have ample statistical power for our main hypotheses. For example we will estimate a relative risk for the effect of depression (vs not) on STEMI, and compare this with the relative risk for the effect of depression vs not on non-STEMI. Assuming depression exposure prevalence of 10%, 500 linked cases of each type and alpha of 0.01, we will be able to distinguish a relative risk of, say 1.7.

Data analysis Our overall analytic approach involves the distinction between different coronary disease phenotypes which form the endpoints of aetiologic, and the start-points of prognostic, analyses and between different temporal patterns of evolution of risk. Furthermore we will consider the separate, and joint, effects of depression and anxiety, distinguishing between a solitary measure and the cumulative impact of serial measures, on the specific coronary disease phenotypes.

A structured Statistical Analytic Protocol detailing how we will analyse the data and to what extent analytic choices are pre-specified, can be made available on request.

Expected value of results

This provides an opportunity to investigate large scale the association of depression with specific coronary syndromes comes from the linkage of GPRD to MINAP.

Inclusion Criteria:

• Aged >18 years
• Patient in a GPRD registered practice that has consented to the linkage process
• Patients are free of any coronary syndrome at the start of follow-up

Exclusion Criteria:

• Less than 1 year of follow-up before their end-point

The study population will include all adults (18+) in GPRD registered in up-to-standard practices with at least 1 year of up to standard follow up. Analyses will focus on the ~200 practices which have consented to linkage with HES and MINAP. Essentially we will define an aetiological cohort - whole population - free of any coronary syndrome at start of follow up. Patients within this cohort are followed for the aetiologic endpoint of a first specific coronary syndrome (see below). This aetiologic endpoint is the prognostic start-point; such patients will then be followed for subsequent specific coronary syndromes and death. We will focus analyses on patients with at least a year follow up data before and their endpoint. As in a conventional cohort study we will define according to exposure history.