An older adult and a clinician review a medication list beside a plain white tablet in a softly lit consultation room.
Aging Science 8 min read

Aspirin May Raise Bleeding Risk After 70; Heart Benefits Remain Unproven

ASPREE suggests preventive aspirin may increase serious bleeding in older adults without established heart disease, while benefits remain unproven in this group.

DISCLAIMER

This article is for informational purposes only and does not constitute medical advice. The statements in this article have not been evaluated by the FDA. The information presented is based on published research and should not be used as a substitute for professional medical guidance. Consult your physician before starting any supplement or health protocol.

The question is whether aspirin prevents a first event

For older adults without established cardiovascular disease, the ASPREE trial suggests that starting preventive aspirin may bring a measurable bleeding risk without a demonstrated cardiovascular benefit. That conclusion concerns people taking a drug to prevent their first cardiovascular event. It should not be transferred automatically to someone prescribed aspirin after a heart attack, an ischemic stroke, or a coronary procedure.

The distinction may explain why familiar advice about aspirin can sound contradictory. Preventing another event in someone with diagnosed disease and preventing a first event in someone without it involve different starting risks. A medicine may have a favorable balance in one setting and an unfavorable balance in another. Its availability without a prescription does not answer which setting applies.

ASPREE offers an unusually direct test of the preventive claim in later life. Rather than asking whether aspirin users appeared healthier in medical records, investigators randomly assigned older volunteers to aspirin or a matching placebo. The results may therefore be more informative about starting preventive treatment than an association between medication use and subsequent health. The original cardiovascular report found no statistically significant reduction in its cardiovascular endpoint, alongside more major bleeding.

For readers considering aspirin as a way to preserve independence, the relevant question may be broader than whether it changes clotting. A preventive medicine needs to improve outcomes that matter enough to justify its harms. A plausible mechanism, an inexpensive tablet, and decades of clinical familiarity may justify testing that proposition; they cannot establish the answer before the outcomes are counted.

ASPREE tested independence, not a blood marker

The trial enrolled 19,114 community-dwelling adults in Australia and the United States. Most were at least 70; eligibility began at 65 for Black and Hispanic participants in the United States. Participants entered without known cardiovascular disease, dementia, or independence-limiting physical disability. That selection makes the findings particularly relevant to prevention in relatively well older adults, while limiting assumptions about people with substantially different health histories.

The main outcome combined death, dementia, and persistent physical disability. Investigators were therefore testing whether aspirin might extend the time people remained alive without those outcomes, rather than merely improve a laboratory result. After a median 4.7 years, the study did not demonstrate a benefit on this measure. The primary trial report also documented more major hemorrhage among participants assigned aspirin.

A combined endpoint may be useful when an intervention could influence several competing outcomes. Preventing one diagnosis would be less persuasive if another serious outcome became more common. However, combining outcomes also requires care: the overall result cannot reveal every possible effect on each component, nor establish that the components have equal importance to every participant.

For interpretation, it helps to separate a trial’s clinical question from the broader phrase healthy aging. ASPREE evaluated a particular medication strategy in a defined population over a defined interval. It did not measure an all-purpose rejuvenation effect. Its failure to demonstrate longer disability-free survival may undermine that specific preventive strategy without resolving every question about aspirin in every disease setting.

The time horizon also matters. Nearly five years may capture clinically important outcomes, but it cannot establish what would happen over every remaining year of life. Conversely, an appeal to benefits that might emerge much later should not erase harms observed during the actual experiment. A longer-term claim needs longer-term observations, rather than an assumption that continued treatment must eventually become worthwhile.

Clot prevention and bleeding may share a mechanism

Aspirin’s antiplatelet action may help explain both its clinical usefulness and its liabilities. Pharmacologic research describes how aspirin inhibits platelet cyclooxygenase-1, reducing production of thromboxane A2, a signal involved in platelet activation. Less platelet activity may reduce some arterial clot formation, while also making bleeding harder to arrest. These are connected consequences of changing the clotting response, rather than unrelated properties of the same tablet. The mechanism is discussed in a clinical pharmacology review.

That mechanism does not specify the net outcome for an individual. A reduction in clot formation might matter greatly to someone at substantial risk of another clot-related event. The same effect may offer less absolute benefit when the chance of such an event is lower. Whether an intervention helps overall therefore depends on both the event it might prevent and the harm it might produce.

This is why an anti-inflammatory explanation alone may be an inadequate argument for preventive aspirin. Describing a pathway does not establish that modifying it improves survival or independence. Different actions of a medicine can occur together, and a promising action may coexist with an unfavorable clinical balance. The experiment needs to connect the proposed mechanism to outcomes in the population being advised to use it.

The comparison also illustrates a limitation of the word protective. Protection against one process may introduce vulnerability elsewhere. A reader assessing a prevention claim can ask what was measured: platelet activity, a diagnosed cardiovascular event, hospitalization, or survival. Each may contribute useful information, but they answer different questions. A mechanistic observation should not quietly become a claim that a person will live longer.

For the same reason, a product description that emphasizes familiarity or a low-dose formulation may leave the central question unresolved. Neither description supplies the missing comparison with otherwise similar people who did not receive the drug. The clinical balance may only become visible when benefits and harms are collected together, using the same follow-up period and comparable definitions.

The bleeding signal was clearer than the heart benefit

In ASPREE, cardiovascular disease occurred at reported rates of 10.7 events per 1,000 person-years with aspirin and 11.3 with placebo. Major hemorrhage rates were 8.6 and 6.2 per 1,000 person-years, respectively. The cardiovascular hazard ratio was 0.95, with a 95% confidence interval of 0.83 to 1.08; the bleeding hazard ratio was 1.38, with an interval of 1.18 to 1.62. These trial estimates suggest greater certainty about increased bleeding than about cardiovascular benefit.

The absolute rate difference for major bleeding was 2.4 events per 1,000 person-years. That arithmetic may help put the relative increase in perspective: a 38% higher hazard does not mean that 38% of participants experienced a major bleed. Relative comparisons describe the contrast between groups; absolute rates help readers understand how frequently events occurred in the trial.

Person-years also need careful reading. One thousand person-years could represent approximately 1,000 people observed for one year, or a different combination of people and observation time. An event rate expressed this way is not automatically an individual’s probability over five years. Converting it into a personal forecast would require assumptions about changing risk, follow-up, and competing events.

The cardiovascular confidence interval crossed 1, the value representing no difference in the hazard comparison. That result does not prove aspirin had precisely zero effect. It suggests that the study could not distinguish its cardiovascular estimate reliably from no difference. The appropriate correction to an overconfident prevention claim is therefore that benefit was unproven in this population, rather than that aspirin could never prevent any event.

Nor should the numerical differences in cardiovascular events and bleeding simply be subtracted to create a universal score. A heart attack and a major hemorrhage may have different consequences, and individual preferences may differ. The useful interpretation is that a clear harm signal accompanied an uncertain benefit signal. A personalized decision may require more than counting the two categories as interchangeable events.

Later observations did not establish a delayed benefit

A 2025 ASPREE extension examined cardiovascular events and major bleeding across the trial and subsequent observational follow-up through 2021–2022. Across the combined period, the hazard ratio for major adverse cardiovascular events was 1.04, with a 95% confidence interval of 0.94 to 1.15. Major hemorrhage remained more frequent in the group originally assigned aspirin, with a hazard ratio of 1.24. The extended follow-up report did not establish a long-term cardiovascular advantage from the original aspirin assignment.

The extension also reported a higher cardiovascular event rate during the post-trial period among those originally assigned aspirin. That finding may deserve further investigation, but it should not be presented as a fresh randomized comparison of continuing versus stopping treatment. Original allocation was randomized; medication use after the intervention ended was not maintained as the same assigned experiment.

This distinction may matter more than the extra years on the calendar. Longer follow-up can reveal outcomes that the original trial was too short to observe. Yet once treatment becomes a matter of clinical care, the interpretation changes. An analysis based on original assignment may estimate a lasting effect of the earlier strategy, while answering a different question from continuous treatment throughout follow-up.

The extension therefore narrows one optimistic interpretation of the original results: the available longer observation did not reveal an overall cardiovascular dividend from the earlier assignment. It does not establish that all medication changes after the trial were harmless or harmful. In particular, a post-trial comparison should not become advice for readers to discontinue a prescription on their own.

Another analysis suggests that the harm assessment may extend beyond dramatic bleeding episodes. In a 2023 secondary analysis, aspirin assignment was associated with roughly a 20% higher hazard of developing anemia. Investigators also examined hemoglobin and ferritin, a marker used in assessing iron stores. The National Institutes of Health summary describes the finding as a reason clinicians may consider periodic blood monitoring in older aspirin users.

That analysis may identify a clinically relevant signal without diagnosing why every affected participant became anemic. It should not be translated into a claim that all fatigue in an aspirin user reflects blood loss, or that taking iron would resolve the problem. A secondary outcome can prompt a more focused clinical question while leaving the cause, significance, and appropriate response uncertain for a particular person.

The prescribing reason may determine the interpretation

The US Preventive Services Task Force’s 2022 recommendation advises against initiating aspirin for primary cardiovascular prevention in adults aged 60 or older. The statement concerns starting aspirin to prevent a first event; it does not cover every reason a clinician might prescribe it. The official recommendation also advises people already taking aspirin to discuss continuation or discontinuation with their clinician.

For a medication review, the most useful starting point may be the original indication. A record of a prior heart attack or a coronary intervention may change how relevant ASPREE is. By contrast, a note saying aspirin was started solely because someone reached a certain birthday may raise a different question. The label aspirin user combines people whose expected benefits may be very different.

A clinician discussion may therefore be more informative when it connects three pieces of information: why aspirin was started, what cardiovascular risk it is intended to address, and what bleeding concerns now apply. A medication list can help make that discussion concrete. The purpose is to clarify the treatment’s rationale, rather than infer that everyone of the same age should receive the same answer.

For someone considering aspirin independently, the trial may provide a reason to pause the preventive assumption before adding a daily medicine. The relevant uncertainty is not whether aspirin has biological activity. It is whether that activity produces enough benefit in a person without established disease to justify the accompanying risk. ASPREE suggests that age alone does not establish such a benefit.

There are also questions this study cannot settle. Its findings should not be treated as a comparison of every possible cardiovascular prevention strategy, an evaluation of occasional aspirin for pain, or a prediction for every genetic subgroup. A claim that a selected subgroup benefits would need a credible analysis and, ideally, confirmation. It should not be presumed merely because the overall result was disappointing.

For readers following healthy-aging science, the methodological lesson may be especially useful: prioritize the population, the assigned comparison, and the clinical endpoint before the mechanistic story. A medicine may alter an important pathway without extending independent life. Here, the available trial and follow-up findings suggest caution about starting aspirin for broad prevention in later life, while preserving the need to assess established medical indications individually.

Frequently asked questions

May daily aspirin help people over 70 live independently for longer?

ASPREE did not demonstrate longer disability-free survival in its study population. That finding leaves a broad independence benefit unproven.

Does a low-dose formulation remove the bleeding concern?

The trial suggests it does not: major bleeding was more frequent despite use of a low-dose regimen.

Should an older adult already taking aspirin stop it?

That decision may depend on the prescribing reason. Someone taking aspirin should discuss changes with their clinician, particularly after a cardiovascular event or procedure.


This article summarises published research and is not medical advice. Consult your healthcare provider before changing your routine.

Frequently Asked Questions

May daily aspirin help people over 70 live independently for longer?
ASPREE did not demonstrate longer disability-free survival in its study population. That finding leaves a broad independence benefit unproven.
Does a low-dose formulation remove the bleeding concern?
The trial suggests it does not: major bleeding was more frequent despite use of a low-dose regimen.
Should an older adult already taking aspirin stop it?
That decision may depend on the prescribing reason. Someone taking aspirin should discuss changes with their clinician, particularly after a cardiovascular event or procedure.

Sources

  1. ASPREE: Effect of Aspirin on Cardiovascular Events and Bleeding in the Healthy Elderly, NEJM, 2018(2026)
  2. ASPREE: Effect of Aspirin on Disability-free Survival in the Healthy Elderly, NEJM, 2018(2026)
  3. The Multifaceted Clinical Readouts of Platelet Inhibition by Low-Dose Aspirin, JACC, 2015(2026)
  4. Aspirin, cardiovascular events, and major bleeding in older adults: extended follow-up of the ASPREE trial, European Heart Journal, 2025(2026)
  5. NIH: Low-dose aspirin may increase risk of anemia in older adults, 2023(2026)
  6. USPSTF: Aspirin Use to Prevent Cardiovascular Disease, 2022 recommendation(2026)
aspirin ASPREE bleeding risk primary prevention

Stay Updated on Longevity Science

Weekly research digests. No spam, unsubscribe anytime.

Subscribe

Related Articles