Does a daily aspirin prevent a first heart attack? Three 2018 trials say rarely

For decades, a low-dose aspirin was widely treated as sensible insurance against a first heart attack. The evidence that reshaped that advice landed in 2018, when three large randomised trials tested aspirin in people who had not already had a cardiovascular event — so-called primary prevention. Their combined message is narrow but clear: in people without established heart disease, aspirin's benefit is small or absent, and it reliably raises the risk of bleeding. This is a different question from secondary prevention — aspirin after a heart attack or stroke — where the balance still favours it and where nobody should stop without medical advice.

The healthy elderly: ASPREE

ASPREE enrolled 19,114 community-dwelling adults in Australia and the United States, aged 70 or older (or 65 and older among Black and Hispanic participants in the US), none with existing cardiovascular disease, dementia or disability [s1]. They were randomly assigned to 100 mg of enteric-coated aspirin or placebo — 9525 to aspirin, 9589 to placebo [s1].

After a median 4.7 years, cardiovascular disease occurred at 10.7 events per 1000 person-years on aspirin versus 11.3 on placebo — a difference that was not statistically significant (hazard ratio 0.95, 95% CI 0.83 to 1.08) [s1]. Major haemorrhage, meanwhile, rose from 6.2 to 8.6 events per 1000 person-years (hazard ratio 1.38, 95% CI 1.18 to 1.62; P<0.001) [s1]. In healthy older adults, in other words, aspirin bought no clear cardiovascular protection while measurably increasing serious bleeding.

Moderate-risk adults: ARRIVE

ARRIVE tested the same 100 mg dose in 12,546 people judged to be at moderate cardiovascular risk — men aged 55 and older, women 60 and older, without diabetes — assigning 6270 to aspirin and 6276 to placebo [s2]. Over a median 60 months, the primary composite endpoint occurred in 4.29% of the aspirin group and 4.48% of the placebo group, again not significant (hazard ratio 0.96, 95% CI 0.81 to 1.13; p=0.6038) [s2]. Gastrointestinal bleeding events, mostly mild, roughly doubled: 0.97% versus 0.46% (hazard ratio 2.11, 95% CI 1.36 to 3.28; p=0.0007) [s2]. The observed event rate was much lower than the trial had been designed to expect, which complicates interpretation — a lower-risk population has less to gain.

People with diabetes: ASCEND

Diabetes was the strongest case for benefit, and ASCEND is where aspirin came closest to earning its place. Among 15,480 adults with diabetes and no known cardiovascular disease, serious vascular events over a mean 7.4 years occurred in 8.5% on aspirin versus 9.6% on placebo — a genuine reduction (rate ratio 0.88, 95% CI 0.79 to 0.97; P=0.01) [s3]. But major bleeding rose in near-mirror image, from 3.2% to 4.1% (rate ratio 1.29, 95% CI 1.09 to 1.52; P=0.003) [s3]. The trial's own conclusion is the key sentence: the absolute benefits were largely counterbalanced by the bleeding hazard [s3].

What the guidance now says

The pattern across all three — a small or absent reduction in events, set against a consistent rise in bleeding — is what moved official advice. In 2022 the US Preventive Services Task Force concluded that for adults aged 40 to 59 with a 10-year cardiovascular risk of 10% or greater, the decision to start low-dose aspirin should be an individual one made with a clinician (a "C" recommendation) [s4]. For adults 60 years or older, it recommended against starting low-dose aspirin for primary prevention at all (a "D" recommendation), because the bleeding risk that accrues with age erodes any benefit [s4].

What this does and does not mean

Three cautions matter. First, every one of these trials is about primary prevention — first events in people without diagnosed cardiovascular disease. None of it applies to someone taking aspirin after a heart attack, stroke or stent, where the evidence still supports it. Anyone in that situation should not stop aspirin on the strength of this coverage; stopping abruptly can itself be dangerous.

Second, these are averages over large groups. They describe how a population fares, not what is right for one person with a particular risk profile, bleeding history and set of other conditions.

Third, "no clear benefit on average" is not the same as "harmless to start or stop on a whim." Aspirin is a real drug with a real bleeding risk, which is why the decision to begin or end it belongs with a clinician who knows the individual — not with a headline, and not with this article. The shift since 2018 is simply that the default for a healthy person without heart disease is no longer a reflexive daily aspirin.