Aspirin is a cheap medication that is taken commonly by many patients to reduce the risk of heart attacks and strokes. There is preliminary evidence that it might also be able to reduce early cancer changes and therefore reduce the risk of cancer. However, there are potentially bad effects of aspirin, for example causing ulcers and bleeding in the stomach. A recent study looking at the risks and benefits controversially suggested that widespread use of aspirin in elderly patients would actually increase their risk of dying.
Purpose:
This study will look what the current risks and benefits of taking aspirin are in a large study in the English population. We will follow all patients who have evidence of taking aspirin in their GP records through their adult life, and compare the association of the cumulative dose of aspirin with both the potential benefits, including reducing cancer and heart problems, balanced against the potential harms like increased bleeding from the stomach and gut.
Importance:
This will benefit patients by providing contemporary evidence on to support either widespread guidance advising patients who do not have established disease to take aspirin as a preventative medication, or more restrictive advice for specific patient groups if there is an harm for some patients who do not have established disease.
Background: That aspirin can reduce the risk of colorectal and possibly other cancers has recently been challenged by the finding of an increased mortality risk in the elderly from the ASPREE study. This suggested that widespread unselected aspirin use in the elderly was harmful, and therefore chemo-preventative effects might be outweighed by aspirin harms. Therefore we will examine what the current risks and benefits of aspirin use are in the general population.
Study Aim: The proposed study will measure the contemporary risks and benefits of aspirin in a large population based cohort
Primary exposure: Cumulative aspirin dose
Outcomes identified in cancer registry, primary care, ONS death registry and secondary care:
1) Cancer incidence categorised by site from linked cancer registry data
2) All cause mortality fron ONS death registry
3) Cardiovascular events from linked HES, ONS death registry, and primary care CPRD data
4) Cerebrovascular events (both bleeding and ischaemic) from linked HES, ONS death registry, and primary care CPRD data
5) Gastrointestinal bleeding from linked HES, ONS death registry, and primary care CPRD data
Study design: This will include all patients with prescribed aspirin use plus an equal sized randomly sampled control group of non aspirin users. The study design will be an open cohort with patients entering and leaving the cohort depending on when they are observed in the CPRD dataset, with follow up time counted from their 18th birthday.
Analysis plan: The cohort will be expanded and split so that time is at a yearly resolution. Cumulative aspirin use will be calculated as a time varying covariate at this one year resolution. A Cox proportional hazards model will then be used to predict the hazard of cancer diagnosis from the time varying cumulative dose of aspirin. This will be adjusted for sex, other relevant medications and co-morbidity.
1. Cancer diagnosis coded in cancer registry categorised by site (ICD 10 coding and site coding) ;
2. Cancer stage and grade categorised by site coded in cancer registry categorised by site (ICD 10 coding, behaviour, morphology, TMN staging, diagnosis at death, and also including cancer specific staging for colon, prostate, breast, uterine and melanoma, and using treatment data to stage if not formal staging not available);
3. Post cancer survival in ONS death registry categorised by site in coded in cancer registry;
4. All cause mortality in ONS death registry;
5. Cardiovascular events from linked HES, ONS death registry, and primary care CPRD GOLD and AURUM data;
6. Gastrointestinal bleeding from linked HES, ONS death registry, and primary care CPRD GOLD and AURUM data;
7. Cerebrovascular events (both bleeding and ischaemic) from linked HES, ONS death registry, and primary care CPRD GOLD and AURUM data;
Colin Crooks - Chief Investigator - University of Nottingham
Colin Crooks - Corresponding Applicant - University of Nottingham
David Humes - Collaborator - University of Nottingham
Joe West - Collaborator - University of Nottingham
Matthew Grainge - Collaborator - University of Nottingham
Timothy Card - Collaborator - University of Nottingham
HES Admitted Patient Care;NCRAS Cancer Registration Data;No additional NCRAS data required;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;CPRD Aurum Ethnicity Record;CPRD GOLD Ethnicity Record