Proton pump inhibitors (PPIs) are widely used to treat gastric acid conditions in older people and can be co-prescribed with aspirin - a drug reducing the risk of heart attack - and other medicines to prevent gastric side effects. PPIs work well and have few adverse effects when used for short periods. However, when used for a longer time, PPIs might increase the risk of heart attack and other severe health conditions such as pneumonia and fractures. It has been suggested that these adverse effects may be more severe in later life. A randomised clinical trial (RCT) would be the best way to understand PPIs effects in older vulnerable people. However, in this population, RCTs would raise ethical concerns, and could be practically very challenging. Therefore, we plan to investigate the effect of PPIs on the risk of cardiovascular events, community acquired pneumonia and fractures in primary care older patients using observational data collected by GPs as part of their daily routines. The challenges posed by the use of observational data will be addressed as much as possible with up-to-date statistical methods. If the epidemiological reports of increased risk with PPIs’ longer-term use are confirmed, this should lead to substantial changes in practice.
The primary objective is to evaluate the risk of incident cardiovascular events (myocardial infarction, MI), as well as that of community acquired pneumonia and fragility fractures in patients aged 60 and over exposed to PPIs for long periods of time. We aim to understand whether age and burden of disease increase the risk of these potential adverse reactions and costs associated with PPI use. To do this, a retrospective parallel cohorts study will be constructed to understand risks in older patients. PPIs (class) receipt will be the exposure and MI, pneumonia and fragility fractures the primary endpoints. Main hypotheses will be tested using Cox regressions including age, burden of disease as potential effect modifiers, and other available covariates in the Cox models to obtain maximally truthful estimates; known confounding will be addressed with regression adjustment by covariates and by propensity scoring, and unknown confounding by the Prior Event Rate Ratio approach. Costs will be derived by extracting health-care utilisation measures from CPRD and Healthcare Resource Group categories from Hospital Episode Statistics. This work is funded by NIHR Research for Patient Benefit (PB-PG-0214-33099), and is closely linked to the NIHR School for Public Health Research “Ageing Well” programme work in Exeter on risks and benefit of cardiovascular prevention strategies in the elderly.
David Melzer - Chief Investigator - University of Exeter
Jan Zirk-Sadowski - Corresponding Applicant - University of Exeter
Alessandro Ble - Collaborator - University of Exeter
Jane Masoli - Collaborator - University of Exeter
Joao Delgado - Collaborator - University of Exeter
Peter Hughes - Collaborator - University of Exeter
Ruben Mujica Mota - Collaborator - University of Exeter
William David Strain - Collaborator - University of Exeter
William Hamilton - Collaborator - University of Exeter
William Henley - Collaborator - University of Exeter
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation