Low-Dose Aspirin Usage for Primary Prevention of Cardiovascular Disease Has Fallen by More Than Half Since 2018

4 min read Original article ↗

Low-dose aspirin (typically 81 mg taken daily) was long recommended to help prevent a first heart attack or stroke in adults at elevated cardiovascular risk. Based on earlier trial evidence, the 2016 U.S. Preventive Services Task Force (USPSTF) endorsed daily aspirin for adults aged 50–69 with at least a 10% ten-year risk of cardiovascular disease (CVD) who were not at increased risk for bleeding.1 However, three large trials published in 2018 (ARRIVE2, ASCEND3, and ASPREE4) found that the cardiovascular benefits of aspirin for primary prevention were smaller than previously observed and were largely offset by an increased risk of serious bleeding. These findings prompted major guideline revisions: the 2019 ACC/AHA guideline recommended against routine primary-prevention aspirin and limited consideration to select adults aged 40–70 at higher CVD risk,5 and the 2022 USPSTF update recommended against initiating aspirin in adults 60 and older altogether.6 Despite these shifts, it is unclear how quickly prescribing practices have changed, and while aspirin’s bleeding risk is well-established in clinical trials, less is known about which real-world patient characteristics most strongly predict serious bleeding in primary prevention populations.

For the prescribing trends analysis, we examined 279 million primary care encounters that occurred between 2015 and 2025 among adults aged 40 and older. Patients with conditions that would indicate aspirin use for secondary prevention (such as coronary artery disease, prior stroke, or peripheral artery disease) as well as those for whom aspirin was contraindicated due to allergy or pregnancy were excluded. 

The share of visits where low-dose aspirin appeared on the medication list fell from a peak of 7.4% in mid-2018 to 3.2% by the end of 2025, a reduction of more than half. The decline has been steady since 2018, and the downward trend was consistent across all demographic subgroups. Notably, adults aged 80 and older, the group current guidelines most strongly recommend against starting on aspirin, still had the highest prevalence at 5.7% in late 2025, down from a peak of 10.9%.

Figure 1

Quarterly Rate of Low-Dose Aspirin per Primary Care Encounter

Figure 1. Low-dose aspirin prevalence rates among adults aged 40+ without a secondary prevention indication. See interactive web version for additional demographic group breakouts.

For the bleeding risk analysis, we studied 625,742 patients aged 40 and older who received their first prescription for daily low-dose aspirin between 2017 and 2025, excluding those with a prior bleeding diagnosis, secondary prevention indications, or pregnancy. To isolate the effect of aspirin itself on bleeding, we needed a comparison group that was similar in health profile but not taking aspirin. We matched aspirin-prescribed patients with those who had an allergy to aspirin documented, a group unlikely to be using aspirin. Matching was based on demographics, start year, comorbidities, and ulcer medication use.

Aspirin’s association with major bleeding was concentrated in adults 75 and older, consistent with findings from earlier clinical trials.2,3,4 Patients aged 75–79 who used aspirin had a 33% higher risk of major bleeding compared to same-aged patients with a documented aspirin allergy, and those aged 80 and above had a 37% higher risk. For adults under 75, there was no significant difference in bleeding risk between aspirin users and those with an allergy to aspirin.

Figure 2

Likelihood of a Major Bleeding Event by Low-Dose Aspirin Use

Figure 2. The likelihood of a patient experiencing a major bleeding event on low-dose aspirin compared to those with an aspirin allergy.

These data come from Cosmos, a dataset created in collaboration with a community of Epic health systems representing more than 300 million patient records from 2,000 hospitals and more than 47,000 clinics from all 50 U.S. states, Canada, Lebanon, and Saudi Arabia. This study was completed by two teams that worked independently, each composed of a clinician and research scientist. The two teams came to similar conclusions. Graphics by Brian Olson.