An unmarked white tablet rests beside a glass sculpture of red blood cells, with an older adult softly silhouetted in the background.
Aging Science 8 min read

Aspirin May Increase Bleeding Without Extending Disability-Free Life

In older adults without cardiovascular disease, aspirin trials suggest bleeding risks may outweigh preventive benefits, with no demonstrated gain in independent life.

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.

What the independence trial actually tested

Research suggests that taking aspirin as an insurance policy against later-life disability may introduce harm without delivering the hoped-for benefit. In ASPREE, a randomized trial of 19,114 community-dwelling older adults, aspirin did not demonstrate longer survival free of dementia or persistent physical disability during a median 4.7 years. Major bleeding was more frequent in the aspirin group. These findings concern preventive use in people without established cardiovascular disease; they may not apply to someone prescribed aspirin after a cardiovascular event. ASPREE primary report.

Participants lived in Australia or the United States and were generally at least 70. Black and Hispanic participants in the United States could enter from 65. They had no known cardiovascular disease, dementia, or independence-limiting physical disability at enrollment. Researchers randomly assigned them to aspirin or a matching placebo, allowing a more credible comparison than simply observing people who chose whether to take the drug.

The primary outcome combined death, dementia, and persistent physical disability. Its event rates were 21.5 per 1,000 person-years with aspirin and 21.2 with placebo. The hazard ratio was 1.01, with a 95% confidence interval of 0.92 to 1.11. That estimate suggests no detectable advantage for aspirin on the trial’s main measure, while leaving some uncertainty around modest benefit or harm.

The choice of outcome matters. A preventive treatment might reduce one diagnosis yet introduce complications elsewhere. An independence-focused trial asks whether the overall balance may translate into additional time alive without specified major impairments. For readers evaluating an anti-aging claim, that question may be more relevant than whether a drug changes a laboratory measurement.

A combined outcome nevertheless needs careful interpretation. Death, dementia, and persistent physical disability represent different experiences. A neutral combined result cannot establish that every component responded identically, and a favorable-looking component cannot automatically overturn the main result. The strongest interpretation may come from keeping the prespecified question separate from explanations proposed after the results are known.

Why the proposed mechanism also raises concern

Aspirin’s preventive rationale may seem straightforward: reducing platelet aggregation may make certain obstructive blood clots less likely. But the same interference with clot formation may make bleeding more likely. A plausible protective mechanism therefore does not establish a favorable overall result. Both sides of the mechanism need clinical measurement, particularly when treatment is proposed for people who have not had a cardiovascular event.

The American College of Cardiology and American Heart Association’s primary-prevention guideline identifies circumstances that may increase bleeding risk, including previous gastrointestinal bleeding or peptic ulcer disease, kidney disease, and concurrent medicines that affect bleeding. These considerations suggest why age or a cardiovascular risk estimate alone may be insufficient for deciding whether aspirin offers a worthwhile tradeoff. ACC/AHA primary-prevention guideline.

The guideline also discusses why more recent prevention trials may differ from older ones. When other cardiovascular risk factors receive contemporary treatment, aspirin may have less additional benefit to contribute. This interpretation does not require aspirin’s pharmacology to have changed. It suggests that a medicine’s practical value may depend on the care already available around it.

Consider a hypothetical preventive treatment that averts a few clot-related events but produces additional serious bleeds. Counting only the events prevented could make that treatment appear more attractive than a comparison that includes both outcomes. The appropriate assessment may also depend on severity: a brief hospital visit, a disabling stroke, and a fatal hemorrhage cannot reasonably be treated as interchangeable experiences.

Mechanistic explanations may help researchers select outcomes and identify vulnerable groups. They cannot supply an unmeasured gain in independence or lifespan. For aspirin, the relevant question may be whether its effects produce a net clinical benefit in a defined population, over a defined period, against a defined alternative. Calling it an anti-inflammatory drug does not answer that question.

Bleeding numbers need an absolute scale

In the original ASPREE report, major hemorrhage occurred in 3.8% of participants assigned aspirin and 2.8% assigned placebo. The comparison suggests an absolute difference of about one percentage point over the trial, alongside a relative increase that may sound much larger when expressed without its baseline. Both descriptions are useful, provided the time period and population remain attached. ASPREE bleeding results.

An illustrative translation may make the scale easier to assess: applying those observed percentages to two groups of 1,000 would correspond to approximately 38 people experiencing major hemorrhage in one group and 28 in the other. This arithmetic is a description of the trial proportions, not a prediction for a particular reader. Individual risk may differ substantially from the study average.

Relative and absolute measures answer different questions. A relative comparison may describe the strength of an association between treatment assignment and an event. An absolute comparison may help convey how frequently the event occurred. Reporting one without the other can obscure the practical meaning, especially when a familiar medicine is discussed as a routine addition to daily life.

Person-years require similar care. A rate per 1,000 person-years incorporates both participants and their observation time; it is not simply the percentage of people affected. A trial following participants for unequal lengths of time may use this denominator to make groups more comparable. Readers should be cautious when a headline silently converts such a rate into a personal annual probability.

For an elective preventive medicine, an absence of demonstrated benefit may be particularly consequential when a credible harm signal is present. That does not mean every person taking aspirin will bleed. It suggests that familiarity, availability, and a plausible mechanism may be inadequate reasons to assume that adding aspirin improves the prospect of remaining independent.

Anemia may reveal a less visible cost

A separate ASPREE analysis published in 2023 suggests that aspirin may increase anemia risk even when major bleeding does not explain the comparison. Researchers used annual hemoglobin measurements and examined ferritin, a marker related to iron stores, in a subset. Anemia rates were 51.2 per 1,000 person-years in the aspirin group and 42.9 with placebo; the hazard ratio was 1.20. Anemia analysis.

The aspirin group also had a greater decline in ferritin. Analyses accounting for major bleeding produced similar findings, which may indicate that counting only major hemorrhages misses part of the treatment burden. However, this was a post hoc analysis, hemoglobin was measured annually, and researchers lacked data establishing the causes of anemia. The findings may strengthen concern without proving the pathway in every affected participant.

That distinction limits the mechanism story. A pattern involving anemia and declining iron stores may be compatible with blood loss, but compatibility does not establish where blood was lost or exclude other explanations. Describing all additional anemia as proven hidden gastrointestinal bleeding would go beyond what this analysis could determine.

The National Institutes of Health’s account suggests that older adults taking aspirin and their clinicians may wish to consider periodic hemoglobin or red-cell monitoring. That is a prompt for individualized review, rather than a universal testing schedule or an instruction to add iron. The study does not establish a supplement regimen that may cancel aspirin-related risks. NIH explanation.

A monitoring result and a treatment decision answer different questions. A blood count may identify a change that deserves investigation; it cannot by itself establish that a preventive medicine is worthwhile. Likewise, a normal result on one occasion may not settle the future balance of benefit and harm. Monitoring should not be interpreted as proof that an otherwise uncertain intervention has become beneficial.

The anemia analysis also illustrates why secondary studies can be valuable without carrying the same interpretive weight as a trial’s primary endpoint. Researchers may identify a clinically relevant consequence that was not the original central question. Confidence may depend on the measurement quality, consistency of related findings, and transparency about which analyses were planned before the data were examined.

Longer follow-up may not rescue the benefit claim

A 2025 ASPREE follow-up examined cardiovascular events and major hemorrhage across the original trial and subsequent observation. Across the combined period, original aspirin assignment did not demonstrate a reduction in major adverse cardiovascular events: the hazard ratio was 1.04, with a 95% confidence interval of 0.94 to 1.15. Major hemorrhage was more frequent overall, with a hazard ratio of 1.24. These results suggest no delayed cardiovascular advantage large enough to emerge in that analysis. Extended follow-up.

The investigators also observed more cardiovascular events during the post-trial period among those originally assigned aspirin. That finding warrants caution because the extension was observational follow-up after the randomized treatment period. It should not be treated as proof that stopping aspirin causes a rebound event, or that every individual would have benefited from continuing it.

The distinction between assignment and actual use becomes especially important after a trial ends. Original assignment may remain useful for studying the consequences of the earlier treatment strategy. It does not necessarily describe the medicine each participant took throughout later observation. Personal prescribing decisions, newly diagnosed disease, and changes in treatment may complicate attempts to identify why a later difference appeared.

Longer observation may improve the chance of detecting delayed benefits, but duration alone does not solve every design problem. Attrition, changes in exposure, and the selection of people available for further follow-up may affect interpretation. A longer study can answer additional questions while also requiring more careful language about what remained randomized and what became observational.

Nor should the extended cardiovascular analysis be described as a new demonstration about every aspect of independent living. Its reported endpoints differ from the original combined independence outcome. The two reports may inform the same clinical discussion, but their conclusions should retain the boundaries of the outcomes each measured.

The prescription’s purpose may change the conclusion

The US Preventive Services Task Force recommends against starting aspirin to prevent a first cardiovascular event in adults aged 60 or older. Its recommendation concerns primary prevention in people without known cardiovascular disease, and reflects a judgment that initiating treatment in this group offers no net benefit. It should not be generalized into a claim that aspirin has no legitimate cardiovascular role. USPSTF recommendation.

The same statement distinguishes starting treatment from continuing treatment begun earlier. It also emphasizes individualized decisions for selected younger adults. These qualifications suggest why a headline about older participants cannot replace a review of the original prescription, current cardiovascular history, and bleeding risk. Someone already taking prescribed aspirin should discuss changes with the treating clinician rather than treating this article as a stopping instruction.

A useful medication review may begin with a concrete question: was aspirin prescribed after an event or procedure, or added to prevent a first event? A second question may be whether the reason remains documented and applicable. A third may concern other medicines or medical conditions that alter bleeding risk. Those details may be more informative than whether a product is marketed as a familiar daily preventive.

The age threshold in a recommendation also needs interpretation. It may summarize how expected benefits and harms balance across groups; it does not imply that a person’s physiology changes abruptly on a birthday. Trial eligibility, guideline thresholds, and an individual’s clinical circumstances serve different purposes and should not be collapsed into a single rule about every older adult.

ASPREE may provide a useful standard for assessing other proposed anti-aging medicines: identify who was studied, what outcome mattered, how treatment was assigned, and which harms were measured. A claim about additional independent years should be supported by outcomes that can meaningfully address that claim. A mechanism or a favorable intermediate measurement may justify further investigation without establishing longer life.

For aspirin used solely as a later-life preventive, the available findings suggest a restrained conclusion. A large randomized trial did not demonstrate more disability-free time, subsequent analyses identified additional concerns, and extended follow-up did not establish a delayed cardiovascular benefit. The practical implication may be a review of the indication and tradeoffs, with no promise that starting, continuing, or stopping aspirin will improve an individual’s lifespan.

Frequently asked questions

May aspirin help older adults live independently for longer?

ASPREE did not demonstrate longer disability-free survival over a median 4.7 years in older adults without known cardiovascular disease. That finding may not apply to other prescribing indications.

Could aspirin contribute to anemia without a major bleed?

A secondary ASPREE analysis suggests it may. Higher anemia incidence persisted in analyses accounting for major bleeding, although the study could not establish the causes of anemia.

Should someone already taking aspirin stop?

These findings do not establish an individual stopping decision. The balance may differ according to the prescription’s purpose, so changes should be discussed with the treating clinician.


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

Frequently Asked Questions

May aspirin help older adults live independently for longer?
ASPREE did not demonstrate longer disability-free survival over a median 4.7 years in older adults without known cardiovascular disease. That finding may not apply to other prescribing indications.
Could aspirin contribute to anemia without a major bleed?
A secondary ASPREE analysis suggests it may. Higher anemia incidence persisted in analyses accounting for major bleeding, although the study could not establish the causes of anemia.
Should someone already taking aspirin stop?
These findings do not establish an individual stopping decision. The balance may differ according to the prescription's purpose, so changes should be discussed with the treating clinician.

Sources

  1. New England Journal of Medicine: Effect of Aspirin on Disability-free Survival in the Healthy Elderly(2026)
  2. ACC/AHA: 2019 Guideline on the Primary Prevention of Cardiovascular Disease(2026)
  3. Annals of Internal Medicine: Aspirin Versus Placebo and Incidence of Anemia(2026)
  4. National Institutes of Health: Low-dose Aspirin May Increase Risk of Anemia in Older Adults(2026)
  5. European Heart Journal: Extended Follow-up of the ASPREE Trial(2026)
  6. USPSTF: Aspirin Use to Prevent Cardiovascular Disease(2026)
aspirin ASPREE primary prevention bleeding risk

Stay Updated on Longevity Science

Weekly research digests. No spam, unsubscribe anytime.

Subscribe

Related Articles