PREDICTing the net benefit of aspirin for primary prevention
EBM Focus - Volume 14, Issue 20
Reference: Ann Intern Med 2019 Sep 17 early online
Three trials published in the last year (ASPREE, ASCEND, and ARRIVE) have brought into question the net benefit of aspirin therapy for primary prevention of cardiovascular disease (CVD). Risk stratification tools could help personalize the discussion about the benefits and harms of aspirin therapy with patients.
A recent study out of New Zealand utilized a previously validated risk calculator (PREDICT) linked to electronic primary care practice management systems in New Zealand to estimate the absolute CVD and bleeding event risks for over 245,000 adults aged 30-79 years. Participants with another indication for aspirin therapy, chronic kidney disease, diabetic nephropathy, atrial fibrillation, or antithrombotic therapy use in the past 6 months were excluded. After calculating the CVD risk, the authors combined this with the relative effect of aspirin therapy as predicted from the updated systematic review to estimate the benefit of aspirin therapy over a 5-year period (RR 0.89 for CVD events) and harms (HR 1.43 for major bleeding events). The difference in the final estimate of harm was then subtracted from the estimate of benefit, yielding a net effect. In an analysis assuming one CVD event (death or hospitalization from CVD event) was equivalent to one major bleeding event (death or hospitalization from bleeding), 2.5% of women and 12.1% of men were classified as benefiting from aspirin therapy over 5 years. Importantly, death or hospitalization due to hemorrhagic stroke was included in both benefit and harm analyses. Substantial benefit, defined as a difference in net effect of 5 events per 1000 persons treated for 5 years, would be gained by 0.2% of women and 2.0% of men. Participants with diabetes, tobacco use, or antihypertensive use were more likely to benefit from aspirin, while those with alcohol use or cancer were more likely to be harmed. Net benefit also differed among different ethnicities and socioeconomic strata.
This study lays out a personalized approach to aspirin for primary prevention. Not only does the model include patient-specific data, but the risk calculator can take into account patient preferences—perhaps some patients are more fearful of major bleeding than of a major CVD event. One limitation of this analysis was the failure to analyze mortality outcomes separately, given that death from either cardiovascular causes or bleeding should count more than other events that someone may recover from. These adjustments are easily made if the risk calculator is designed properly. While clinicians and patients often view the electronic medical record as a barrier to care, putting these big datasets to use may help us provide more personalized medicine.
For more information, see the topic Aspirin for Primary Prevention of Cardiovascular Disease in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Carina Brown, MD, Assistant Professor at Cone Health Family Medicine Residency. Edited by Alan Ehrlich, MD, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School and Katharine DeGeorge, MD, MS, Associate Professor in Family Medicine at the University of Virginia and Clinical Editor at DynaMed.