The risk of cardiovascular related death following cytomegalovirus (CMV) infection in England: an open cohort study using data from the Clinical Practice Research Datalink (CPRD)

Date of ISAC Approval: 
Lay Summary: 
It is thought that over half of the population has been infected with cytomegalovirus (CMV). Most people do not realise they have been infected and in the majority of people the virus causes a mild flu like illness which gets better by itself. After the illness the infection may persist for life. It is thought that it is managed by a person's immune system and does not cause noticeable symptoms. However, there is some evidence that being infected with the virus may increase the risk of death. In particular there may be a link between CMV infection and death related to cardiovascular disease (heart attacks or strokes for example); however the evidence is not clear cut. This research aims to understand if CMV infection is associated with deaths related to cardiovascular disease. We will use GP records to identify one group of patients who have had a positive test result for CMV infection and a second group of people who have negative test results or who have never been tested for CMV. We will then look at the past medical records and death certificates of these people and compare the risk of death from cardiovascular disease in the two groups.
Technical Summary: 
Cytomegalovirus (CMV) infection is highly prevalent (estimated seroprevalence 66-70%). There is mixed evidence of an association between CMV infection and death related to cardiovascular disease although the exact biological mechanisms remain unclear. This study will investigate this association using a large data set from the Clinical Practice Research Datalink (CPRD). A retrospective cohort study will be conducted with a population aged over 45 years, registered with a CPRD practice between 2003 and 2013 (minimum one year of CPRD follow up and link to mortality data). CMV exposure will be defined as participants with a record of seropositivity or a CMV infection. Patients with negative CMV test results will form the unexposed group. This group may be supplemented with a random sample of people with no record of a CMV test following descriptive analysis. Office of National Statistics mortality data will be used to identify the deaths related to CVD, stroke or coronary heart disease. Competing risks survival analysis will be used to quantify the risk of CVD, stroke or CHD death associated with CMV infection. Analyses will adjust for the potential confounding effects of age, sex, GP practice, geographical area, social deprivation, BMI >30, smoking status, comorbidities and inflammation.
Health Outcomes to be Measured: 
Death from cardiovascular disease Death from stroke and coronary heart disease

Puja Myles - Chief Investigator - CPRD
John Mair-Jenkins - Corresponding Applicant - Public Health England
Professor Jonathan Van-Tam - Collaborator - University Of Nottingham
Dr Richard Puleston - Collaborator - University Of Nottingham

HES Admitted;ONS;Patient IMD;Practice IMD (Standard)