Should You Take Aspirin Every Day? The Evidence Changed.

The daily aspirin recommendation was once as standard as telling a man to eat less salt. For two decades, physicians handed it out to middle-aged men with a few cardiovascular risk factors as routine prevention, and patients accepted it without much scrutiny. The evidence that built that habit has since been contradicted by larger, better-designed trials, and the major guidelines have responded accordingly.

The Mechanism

Aspirin works by irreversibly blocking cyclooxygenase-1 (COX-1), an enzyme that platelets use to synthesize thromboxane A2. Thromboxane A2 is a potent promoter of platelet aggregation and vasoconstriction. When aspirin acetylates COX-1, that platelet cannot produce thromboxane A2 for the rest of its functional life, which is roughly seven to ten days. Because platelets are anucleate, they cannot synthesize new COX-1 to replace what aspirin inactivated. The antiplatelet effect therefore persists until the bone marrow replaces the inhibited platelet population.

This is why aspirin made intuitive sense for heart attack prevention: atherosclerotic plaques rupture, platelets pile on at the rupture site, and a clot forms inside a coronary artery. Block platelet aggregation upstream, and you reduce clot formation. That reasoning is pharmacologically sound and clinically validated in secondary prevention, where a patient already has atherosclerotic plaque that has proven it can cause events. The problem is that the same COX-1 inhibition that slows platelet aggregation in arteries also blunts the protective platelet response in the gastrointestinal mucosa, where ongoing microscopic injuries depend on prompt platelet plugging. Aspirin tips that balance, and in a person without established coronary disease, the bleeding side of the ledger often outweighs the clot-prevention side.

The dose matters here. At 81 mg daily, aspirin produces near-complete COX-1 inhibition in platelets with less systemic exposure than 325 mg. Both doses suppress thromboxane A2 by more than 95 percent at steady state, so the antiplatelet effect is essentially equivalent. The higher dose adds more COX-2 inhibition and more prostaglandin-mediated GI mucosal damage without proportionally more cardiovascular protection. This is why guidelines recommend 81 mg when aspirin is indicated, not 325 mg.

What the Evidence Shows

The aspirin-for-everyone habit was not invented from nothing. The Physicians’ Health Study (Steering Committee, 1989) randomized 22,071 male US physicians to aspirin 325 mg every other day versus placebo. It found a 44 percent reduction in first myocardial infarction in the aspirin group. That finding was real, and it was substantial. What got lost in translation was the baseline: these were physicians, a population with lower cardiovascular event rates than the general population, and the absolute risk reduction was modest even though the relative risk reduction looked impressive. The Women’s Health Study (Ridker et al., NEJM 2005) randomized 39,876 healthy women to aspirin 100 mg every other day versus placebo. It found no significant reduction in MI, though there was a reduction in stroke in women over 65. Neither trial is the reason guidelines changed. They are the reason guidelines existed in the first place, and they need to be understood in that context.

What actually changed the calculus was three trials published in 2018, against a background of dramatically improved cardiovascular care. 5 / Solid

ASPREE (McNeil et al., NEJM 2018, doi.org/10.1056/NEJMoa1805819) enrolled 19,114 community-dwelling adults aged 70 and older in Australia and the United States, all without established cardiovascular disease, dementia, or physical disability. Randomized to aspirin 100 mg daily or placebo, followed for a median of 4.7 years. The primary composite endpoint: death, dementia, or persistent physical disability. Result: no significant difference between groups (HR 1.01, 95% CI 0.92 to 1.11). Major bleeding occurred in 8.6 percent of the aspirin group versus 6.2 percent of the placebo group (HR 1.38, 95% CI 1.18 to 1.62). The finding that generated the most discussion was all-cause mortality: 5.9 percent in the aspirin group versus 5.2 percent in placebo (HR 1.14, 95% CI 1.01 to 1.29), driven substantially by cancer-related deaths. Whether aspirin accelerated cancer progression or whether this was a chance finding in a trial not powered for cancer outcomes remains an open question. Subsequent analyses from ASPREE suggested that aspirin may promote the growth of pre-existing subclinical cancers through effects on tumor microenvironment, but this is not established causation. The signal is worth noting; it is not grounds for panic about past aspirin use.

ARRIVE (Gaziano et al., Lancet 2018, doi.org/10.1016/S0140-6736(18)31924-X) enrolled 12,546 adults at moderate cardiovascular risk, defined as men over 55 with two to four cardiovascular risk factors (hypertension, dyslipidemia, smoking, diabetes, obesity, family history). Aspirin 100 mg daily versus placebo, followed for approximately five years. Primary endpoint: composite of MI, unstable angina, stroke, TIA, or cardiovascular death. Result: no significant difference (HR 0.96, 95% CI 0.81 to 1.13). GI bleeding events were more frequent in the aspirin group (0.97% vs 0.46%, HR 2.11). Importantly, the trial enrolled a population with lower-than-anticipated event rates, which reduced its statistical power to detect a difference. Some investigators argued ARRIVE was underpowered. The practical implication is unchanged: in a moderate-risk population treated with modern preventive therapies, aspirin did not show net benefit.

ASCEND (Bowman et al., NEJM 2018, doi.org/10.1056/NEJMoa1804988) addressed a distinct question: aspirin in patients with diabetes but no established cardiovascular disease. Enrolled 15,480 adults with diabetes, randomized to aspirin 100 mg or placebo, followed for a mean of 7.4 years. There was a modest but statistically significant reduction in serious vascular events: 8.5 percent versus 9.6 percent (rate ratio 0.88, 95% CI 0.79 to 0.97). There was also a significant increase in major bleeding: 4.1 percent versus 3.2 percent (rate ratio 1.29, 95% CI 1.09 to 1.52). The cardiovascular benefit and the bleeding harm were nearly identical in absolute magnitude, leaving net benefit essentially neutral. For patients with diabetes but no prior cardiovascular events, aspirin is no longer routinely recommended.

Why did these trials find results so different from the 1989 Physicians’ Health Study? The answer is in the background. Modern cardiovascular care has significantly reduced baseline event rates: statins cut LDL and stabilize plaque, antihypertensives control blood pressure more effectively, smoking rates have declined, and diabetes management has improved. When the absolute risk of a cardiovascular event is lower, the absolute benefit from aspirin is proportionally smaller. Aspirin’s relative risk reduction for MI was never massive, somewhere in the 20-30 percent range in earlier trials. If the untreated event rate drops from 2 percent per year to 0.6 percent per year because of modern therapy, a 20 percent relative risk reduction goes from an absolute benefit of 0.4 percent per year to 0.12 percent per year. The GI bleeding risk, however, is a property of the drug independent of cardiovascular background risk. That risk has not declined. As preventive care improved, aspirin’s benefit-to-bleeding ratio quietly inverted.

Quantifying the bleeding risk: pooled data from meta-analyses consistently show that low-dose aspirin increases GI bleeding events by approximately 2 to 3 times compared to placebo in primary prevention populations. The absolute risk increase translates to roughly two to three additional major GI bleeding events per thousand patients per year. Intracranial bleeding risk also increases, with estimates around 0.2 additional events per thousand patients per year from meta-analysis data.

The 2022 USPSTF guideline update (doi.org/10.1001/jama.2022.4983) synthesized this evidence into explicit age-based recommendations. For adults aged 40 to 59 with a 10-year cardiovascular risk of 10 percent or greater: the decision to initiate aspirin should be individualized, accounting for bleeding risk, with a small net benefit in this group. For adults 60 and older: do not initiate aspirin for primary prevention. The ACC/AHA 2019 guideline on primary prevention of cardiovascular disease (Arnett et al., JACC 2019) reached a similar conclusion, stating that low-dose aspirin should not be administered on a routine basis for primary prevention and may be considered only for selected adults aged 40 to 70 at higher ASCVD risk who are not at increased bleeding risk.

To know where you fall, you need your 10-year ASCVD risk calculated using the Pooled Cohort Equations developed jointly by the ACC and AHA. The equation incorporates age, sex, race, total cholesterol, HDL cholesterol, systolic blood pressure, blood pressure treatment status, diabetes, and smoking status. It outputs a 10-year probability of a first atherosclerotic cardiovascular event (fatal or nonfatal MI, or fatal or nonfatal stroke). The calculator is available free at tools.acc.org/ASCVD-risk-estimator-plus. A score below 10 percent in adults aged 40 to 59 means aspirin is not recommended. A score at or above 10 percent opens the conversation, with bleeding risk factored in.

Contraindications that effectively close that conversation: history of gastrointestinal bleeding, active peptic ulcer disease, regular NSAID use (ibuprofen, naproxen), current anticoagulant therapy (warfarin, apixaban, rivaroxaban), or uncontrolled blood pressure. Combining aspirin with NSAIDs multiplies GI bleeding risk. Combining aspirin with anticoagulants can result in major bleeding events at rates that would be unacceptable in a primary prevention context.

Aspirin Resistance: When the Drug Does Not Work as Expected

Laboratory aspirin resistance, defined as the continued production of thromboxane A2 despite regular aspirin use, affects a meaningful fraction of patients on antiplatelet therapy. Estimates in the literature range from 10 to 25 percent depending on the measurement method, the population studied, and the definition used. Understanding why resistance occurs is relevant primarily in secondary prevention patients, where the antiplatelet effect is expected to be providing clinical benefit and where failure of that effect carries direct consequence.

The mechanism of aspirin’s antiplatelet action depends on acetylation of the COX-1 enzyme in platelets. Resistance can occur when this acetylation is incomplete, or when platelets find alternative routes to produce thromboxane A2 through COX-2, which aspirin inhibits less effectively at standard doses, or through alternative arachidonic acid pathways that bypass both cyclooxygenase isoforms entirely.

Several identifiable sources of reduced aspirin response have been documented. Genetic polymorphisms in the COX-1 gene and in the glycoprotein IIb/IIIa receptor have been associated with reduced responsiveness to aspirin in platelet aggregation assays. Elevated platelet turnover, which occurs in obesity and in inflammatory states, replenishes the uninhibited platelet pool more rapidly than standard once-daily aspirin can block. This is one mechanistic explanation for observations that aspirin’s clinical efficacy appears lower in patients with higher body mass index. Cigarette smoking has been shown in multiple studies to impair aspirin’s antiplatelet effect, with platelet activation markers remaining elevated in active smokers despite regular aspirin use.

Enteric coating introduces a separate pharmacokinetic problem. Grosser and colleagues published data in 2013 demonstrating that enteric-coated aspirin produces substantially lower and more variable platelet inhibition than plain aspirin in fasting conditions, due to reduced gastric absorption. When the same patients received plain aspirin, platelet inhibition was consistent and complete. In non-fasting conditions the gap narrows, but the study raised legitimate questions about whether enteric-coated aspirin is providing reliable pharmacological effect in all patients, particularly those with delayed gastric emptying or other absorption variables.

Clinical measurement of aspirin resistance using urinary thromboxane B2 or point-of-care platelet function testing exists, but is not standardized for routine practice. Increasing aspirin dose does not reliably overcome resistance when the mechanism is genetic or receptor-level, and higher doses increase bleeding risk without proportionally greater cardiovascular benefit. In secondary prevention patients with recurrent events despite adequate aspirin use, the appropriate clinical response is typically intensification of antiplatelet therapy, often by adding a P2Y12 inhibitor, rather than adjusting aspirin dose alone. 3 / Early

What to Do This Week

1. Identify which category applies to you. If you have had a heart attack, a stent, a stroke, confirmed coronary artery disease, or peripheral artery disease, you are in secondary prevention. Stay on your aspirin and confirm the dose with your cardiologist. If you have not had any of those events, you are in primary prevention, and the following steps apply to you.

2. Calculate your 10-year ASCVD risk using the ACC/AHA Pooled Cohort Equations at tools.acc.org/ASCVD-risk-estimator-plus. You will need your most recent total cholesterol, HDL, and systolic blood pressure from a lab or clinic visit. The output is a percentage: below 10 percent at ages 40 to 59 means aspirin is not recommended; above 10 percent opens the conversation.

3. If you are currently taking aspirin for primary prevention and are over 60, bring this to your physician’s attention at your next visit. Ask specifically: given my current risk profile and the 2022 USPSTF update, should I be on this? Do not stop abruptly without that conversation, but do not assume continued aspirin is automatically appropriate.

4. Audit your bleeding risk before the appointment. Have you had GI bleeding, ulcers, or frequent NSAID use (ibuprofen for back pain, naproxen for joint pain)? Are you on any blood thinner? These factors shift the benefit-risk calculation against aspirin even if your cardiovascular risk is elevated.

5. If you are in the 40 to 59 age range, your 10-year ASCVD risk exceeds 10 percent, and you have no significant bleeding risk factors, aspirin at 81 mg daily is a reasonable discussion to have with your physician. Not a default. A discussion with numbers on the table.

The clinical picture here is specific: aspirin remains one of the most effective medications in medicine for the right indication, and that indication is secondary prevention. For primary prevention, the data say it is the wrong tool for most men in 2026, not because aspirin changed, but because the cardiovascular landscape around it did.