evipedia.ai
← Back

Low-Dose Aspirin for Health & Longevity

Evidence Review created on 08/27/2026 using AI4L / Opus 5

Also known as: Aspirin, Acetylsalicylic Acid, ASA, Baby Aspirin, Low-Dose ASA, Enteric-Coated Aspirin, Cardio Aspirin

Motivation

Aspirin is among the oldest and most widely taken medicines in the world. At the small daily amounts covered here — roughly 75 to 100 milligrams, about a quarter of a standard headache tablet — it is not taken for pain. It works by permanently switching off an enzyme inside platelets, the tiny blood cells that make blood clot, so that clots form less readily for the remainder of each platelet’s short life.

Tens of millions of adults have taken a small daily dose for decades to lower the chance of a first or a repeat heart attack or stroke, and the same habit has been studied for a long-delayed reduction in bowel cancer. The clot-blocking action that produces those effects also makes bleeding more likely, and large trials in healthy older adults have sharpened the argument on both sides rather than settling it.

This review examines what the evidence shows about small daily doses of aspirin for heart and blood-vessel events, for cancer, and for bleeding; how that balance shifts with age, body weight, and underlying risk; and where the evidence remains genuinely unsettled.

Benefits - Risks - Protocol - Conclusion

This section collects high-level commentary and overview pieces that frame the aspirin debate for a health- and longevity-oriented audience.

No dedicated aspirin content could be found on hubermanlab.com or lifespan.io. The Huberman Lab on-site search returned only timestamped segments inside broader episodes on pain relievers and drug repurposing, and Lifespan.io’s search surfaced only roundups and unrelated articles that mention aspirin in passing, so neither platform is represented above.

Grokipedia

  • Aspirin

    Covers acetylsalicylic acid’s chemistry, cyclooxygenase pharmacology, antithrombotic use, dosing, and adverse effects in one place, with the primary-prevention controversy treated at length.

Examine

No dedicated Examine article on aspirin exists. Aspirin is a regulated drug rather than a dietary supplement, and Examine’s monographs cover supplements; aspirin appears only as research-feed study summaries, which are excluded here.

ConsumerLab

No dedicated ConsumerLab article on aspirin exists. ConsumerLab tests dietary supplements rather than over-the-counter or prescription drugs, so aspirin appears only inside articles about supplement interactions and never as a product review of its own.

Systematic Reviews

The reviews below cover both sides of the aspirin trade-off: the vascular and cancer benefits, and the bleeding harm.

Mechanism of Action

Aspirin irreversibly acetylates a serine residue in the active site of cyclooxygenase, the enzyme family that converts arachidonic acid into prostaglandins and thromboxanes. COX-1 (cyclooxygenase-1, the form present in platelets and the stomach lining) produces thromboxane A₂, the signal that recruits and clumps platelets, and the prostaglandins that protect the gastric lining. COX-2 (cyclooxygenase-2, induced at sites of inflammation) drives inflammatory prostaglandins and is the main target of other NSAIDs (nonsteroidal anti-inflammatory drugs, the pain-and-inflammation class that includes ibuprofen).

At 75–100 mg daily, aspirin acetylates platelet COX-1 in the portal blood before the liver clears it. Platelets have no nucleus and cannot rebuild the enzyme, so one small daily dose suppresses over 95% of thromboxane A₂ for the platelet’s 8–10 day lifespan while largely sparing COX-2 elsewhere.

Two competing mechanisms are proposed for the cancer effect. One is platelet-mediated: suppressed thromboxane A₂ blocks platelet–tumour cell interactions that help circulating tumour cells seed. The other is direct epithelial COX-2 inhibition plus suppression of NF-κB (nuclear factor kappa B, a master switch for inflammatory gene expression). Aspirin also triggers 15-epi-lipoxin A₄, an inflammation-resolving signal.

Pharmacologically, aspirin’s own half-life is 15–20 minutes; esterases hydrolyse it to salicylate, whose half-life is 2–3 hours at these doses. Salicylate distributes widely, including to synovial fluid and the central nervous system, and is cleared mainly by glucuronidation via UGT1A6 and UGT2B7 (enzymes that attach sugar groups to make drugs water-soluble), with minor hydroxylation by CYP2C9 (a liver enzyme that metabolises many drugs).

Historical Context & Evolution

Willow bark served as a fever and pain remedy for millennia; salicin was isolated in the 1820s, and Bayer marketed acetylsalicylic acid as Aspirin from 1899. Its original use was symptomatic relief of pain, fever and rheumatic inflammation — not clotting, which was not then understood.

The clot-blocking property emerged from two directions. Physicians noticed bleeding in aspirin takers; then in 1971 John Vane showed aspirin blocks prostaglandin synthesis, and later work identified platelet thromboxane A₂ as the product that mattered. That insight, described by Patrono, made aspirin a cardiovascular drug and explained why small doses suffice.

Secondary-prevention trials through the 1980s were decisive, and by the 1990s healthy adults took it too. The 2009 individual-patient meta-analysis then showed the primary-prevention benefit was far smaller than the secondary-prevention one and roughly matched by extra major bleeds — a quantification of the earlier work in a lower-risk population, not a refutation of it.

Three 2018 trials in lower-risk groups — ASPREE, ARRIVE and ASCEND — reported little or no net gain, and the 2022 US Preventive Services Task Force statement recommended against starting aspirin at 60 and above. That panel is government-appointed and unpaid, unlike the cardiology societies issuing competing guidance, whose members earn revenue from cardiovascular care. What changed was not the drug but the background: statins, blood-pressure control and falling smoking lowered the event rate aspirin had to beat. Bleeding risk did not fall with it, so the arithmetic reversed for average-risk people while leaving the high-risk case intact.

Expected Benefits

Note on who funded the evidence: most of the large aspirin trials cited here were publicly or charitably funded — ASPREE by the US National Institute on Aging and Australia’s National Health and Medical Research Council, Add-Aspirin by Cancer Research UK and the UK Medical Research Council, CAPP2 by Cancer Research UK and the European Union. The two main exceptions are ARRIVE, sponsored by Bayer, and ASCEND, which received aspirin and funding support from Bayer. Because aspirin is an unpatented generic, no manufacturer has a large financial stake in its continued adoption, and the commercial incentive in this field runs mainly the other way, toward newer patented antiplatelet and lipid-lowering drugs that would displace it. Institutional payers have the mirror-image incentive: a two-cent tablet is far cheaper than the patented alternatives, so insurers and national health systems have a systematic financial reason to favour aspirin, and that asymmetry is a plausible source of structural bias in both guideline formation and research funding.

High 🟩 🟩 🟩

Reduction in Major Atherosclerotic Cardiovascular Events

Suppressing platelet thromboxane A₂ makes arterial clot formation less likely, so fewer heart attacks and ischaemic strokes occur. The benefit is large and unambiguous in people who already have established arterial disease, and much smaller in people who do not, as the individual-patient meta-analysis and the later 13-trial analysis of 164,225 participants both show. Benefit scales with underlying plaque burden, so it is largest in individuals whose imaging or risk markers already show meaningful disease.

Magnitude: Secondary prevention: serious vascular events 6.7% versus 8.2% per year, relative risk (the ratio of event rates in two groups) 0.81. Primary prevention: 0.51% versus 0.57% per year, relative risk 0.88; in the 2019 analysis hazard ratio (the ratio of the rates at which events accumulate over time) 0.89, 95% confidence interval (the range containing the true value with 95% probability) 0.84–0.95, number needed to treat (how many people must take the drug to prevent one event) 265 over five years.

Long-Term Reduction in Colorectal Cancer Incidence and Mortality ⚠️ Conflicted

Pooled 20-year follow-up of randomised vascular trials found fewer colorectal cancer deaths after at least five years of daily aspirin, the benefit appearing only after long latency (Rothwell et al.). A trial in carriers of Lynch syndrome (an inherited bowel-cancer predisposition) found a durable reduction over 10 years (Burn et al.), and a meta-analysis of adenoma-prevention trials found fewer recurrent polyps. Against this, a trial starting aspirin after 70 found more cancer deaths. Net reading: the effect is real but needs years of exposure begun before old age.

Magnitude: Colon cancer death at 20 years, hazard ratio 0.65 (95% confidence interval 0.48–0.88), an absolute 1.76% reduction after ≥5 years of 75–300 mg daily; in Lynch syndrome carriers taking 600 mg daily, colorectal cancer hazard ratio 0.65 (0.43–0.97); recurrent adenoma relative risk 0.83 (0.72–0.96) across 81–325 mg daily.

Prevention of Preeclampsia and Its Complications in At-Risk Pregnancy

In individuals who are pregnant or planning pregnancy and screen at elevated risk, low-dose aspirin started in the first trimester reduces preterm preeclampsia, a placental disorder producing high blood pressure and organ injury. The ASPRE trial randomised 1,776 screened-positive women to 150 mg nightly or placebo, and a meta-analysis shows the effect depends on starting before 16 weeks and on adequate dose. This is the one setting where the evidence for aspirin has strengthened rather than weakened.

Magnitude: Preterm preeclampsia 1.6% versus 4.3%, odds ratio (the ratio of the odds of an event in two groups) 0.38 (95% confidence interval 0.20–0.74); the meta-analysis reports a relative risk of 0.33 (0.19–0.57) in the subgroup started at or before 16 weeks at ≥100 mg.

Reduction in Recurrent Venous Thromboembolism After Anticoagulation

After a first unprovoked clot in a deep vein or the lungs, the risk of a further clot stays high once anticoagulant treatment stops, and low-dose aspirin lowers it by blunting platelet participation in venous clot formation. Two randomised trials and their pooled individual-patient analysis agree on the size of the effect (Simes et al.), with no measurable excess of major bleeding at the low event rates seen in that population.

Magnitude: Recurrent venous thromboembolism 5.1% versus 7.5% per year, hazard ratio 0.68 (95% confidence interval 0.51–0.90); major vascular events 5.7% versus 8.7% per year, hazard ratio 0.66 (0.50–0.86); major bleeding 0.5% versus 0.4% per year.

Medium 🟩 🟩

Lower Incidence of Type 2 Diabetes

A post-hoc analysis of a 19,114-participant randomised trial in healthy adults over 65 found fewer new diagnoses of type 2 diabetes on 100 mg daily, alongside a slower rise in fasting glucose (Zoungas et al.). The proposed mechanism is reduced low-grade inflammatory signalling in insulin-responsive tissue. This comes from a single trial and was not a primary endpoint, so it is suggestive rather than established, and it was accompanied by the same bleeding excess seen for all other endpoints.

Magnitude: Incident type 2 diabetes hazard ratio 0.85 (95% confidence interval 0.75–0.97); 459 versus 536 new cases among 16,209 participants over a median 4.7 years, with fasting plasma glucose 0.048 mmol/L lower at year five on aspirin.

Improved Recurrence-Free Survival in PI3K-Altered Colorectal Cancer

Roughly a third of colorectal tumours carry activating changes in the PI3K pathway (phosphoinositide 3-kinase, a growth-signalling cascade that aspirin appears to interrupt). A Scandinavian randomised trial gave 160 mg daily for three years after surgery to patients selected by tumour genotype (Martling et al.) and roughly halved recurrence. It is one trial in a selected population, but it is the first prospective confirmation that the anticancer effect is genotype-dependent rather than universal.

Magnitude: Three-year recurrence 7.7% versus 14.1% in the PIK3CA (the gene encoding the enzyme unit that drives the PI3K growth signal) exon 9/20 group, hazard ratio 0.49 (95% confidence interval 0.24–0.98).

Low 🟩

Reduction in Liver Fat in Metabolic Dysfunction-Associated Steatotic Liver Disease

In a six-month randomised trial of 80 adults with metabolic dysfunction-associated steatotic liver disease (liver fat driven by insulin resistance, not alcohol), 81 mg daily lowered liver fat on imaging (Simon et al.). The proposed mechanism is suppression of pro-fibrotic lipid mediators; the outcome is an unvalidated imaging surrogate.

Magnitude: Absolute liver fat fell 6.6% on aspirin versus a 3.6% rise on placebo, an adjusted difference of −10.2% (95% confidence interval −27.7% to −2.6%); no difference in liver stiffness at six months.

Possible Reduction in Risk of Cancers Outside the Colon

Pooled analyses report lower incidence and mortality for liver, gastric and oesophageal cancer among aspirin users (Petrelli et al.). The signal outside the colorectum rests mainly on observational cohorts vulnerable to healthy-user bias, and the randomised data are sparse and inconsistent.

Magnitude: Reported hazard ratios cluster around 0.75–0.90 for liver and gastrointestinal cancer incidence in cohort data, with confidence intervals that cross 1.00 in the randomised subsets.

Speculative 🟨

Extension of Lifespan in Animal Models

In the National Institute on Aging’s Interventions Testing Program, aspirin extended median lifespan in male mice but not females (Strong et al.). No human data exist; the basis is animal and mechanistic only.

Benefit-Modifying Factors

  • Body weight and dose matching: In pooled individual-patient data, 75–100 mg reduced vascular events only below about 70 kg, while heavier people needed ≥325 mg; benefit was absent or reversed when dose and weight were mismatched.

  • PI3K-pathway tumour genotype: Colorectal tumours carrying activating PIK3CA changes respond to post-surgical aspirin; the one randomised trial enrolled only genotype-positive patients, so benefit outside that group remains untested rather than excluded.

  • Aspirin-metabolism and platelet-reactivity variants: Variants in UGT1A6 (salicylate clearance) and PEAR1 (a platelet adhesion receptor) shift both salicylate exposure and residual platelet reactivity, and rs2965667 near MGST1 (a detoxifying-enzyme gene) reportedly modifies the colorectal benefit.

  • Lipoprotein(a) genotype: Carriers of the LPA variant rs3798220, who have markedly raised lipoprotein(a) — a cholesterol particle with a clot-promoting tail — showed a much larger vascular benefit than non-carriers in the Women’s Health Study.

  • Baseline biomarker levels: Benefit tracks absolute vascular risk, so higher coronary artery calcium, lipoprotein(a), apolipoprotein B and high-sensitivity C-reactive protein all enlarge the expected gain; low levels shrink it toward the bleeding harm.

  • Pre-existing health conditions: Established coronary, cerebrovascular or peripheral arterial disease multiplies the absolute benefit roughly fivefold over primary prevention; diabetes, chronic kidney disease and Lynch syndrome also raise it.

  • Sex-based differences: Early trials suggested aspirin prevented mainly stroke in women and mainly myocardial infarction in men; pooled data have largely dissolved this, though women in the Women’s Health Study gained mainly stroke reduction.

  • Age at initiation: Benefit for cancer requires years of latency, so starting in the fifties or sixties allows it to accrue; starting after 70 gives bleeding and possible cancer harm immediate expression with no time for the delayed gain.

Sources for these factors: Rothwell et al. (bodyweight and dose), Martling et al. (tumour genotype), Nan et al. (genetic modifiers of colorectal benefit), Chasman et al. (lipoprotein(a) genotype), Verghese et al. (coronary calcium-guided allocation) and Calderone et al. (age stratification).

Potential Risks & Side Effects

High 🟥 🟥 🟥

Major Gastrointestinal Bleeding

Loss of COX-1-derived protective prostaglandins in the stomach lining, combined with impaired platelet plugging, makes ulceration and bleeding both more likely and harder to stop. This is the dominant harm and it is consistent across randomised trials and real-world cohorts (Zheng & Roddick; García Rodríguez et al.; ASPREE). Risk rises steeply with age, with Helicobacter pylori infection, and with concurrent anticoagulants, other anti-inflammatories or antidepressants; it is not reduced by enteric coating.

Magnitude: Major bleeding hazard ratio 1.43 (95% confidence interval 1.30–1.56), number needed to harm (how many people must take the drug to cause one extra event) 210 over five years. Upper gastrointestinal bleeding relative risk 2.3 (2.0–2.6) in observational data; in ASPREE, major haemorrhage 8.6 versus 6.2 events per 1,000 person-years.

Intracranial Haemorrhage

Bleeding into or around the brain is the least common but most consequential haemorrhagic harm, and unlike gastrointestinal bleeding it is frequently disabling or fatal. Randomised data in healthy older adults show a clear excess of intracerebral, subdural and extradural bleeds (Cloud et al.), matched by pooled observational estimates (García Rodríguez et al.). Head trauma, uncontrolled hypertension and cerebral amyloid angiopathy (protein deposits that weaken small brain vessels) amplify it, and the excess is not offset by any reduction in ischaemic stroke in primary prevention.

Magnitude: Intracranial bleeding hazard ratio 1.38 (95% confidence interval 1.03–1.84) in ASPREE — 108 versus 79 events over about 4.7 years — and relative risk 1.4 (1.2–1.7) in pooled observational data.

Aspirin-Exacerbated Respiratory Disease and Hypersensitivity

In a subset of adults with asthma and nasal polyps, COX-1 blockade diverts arachidonic acid into leukotriene production, provoking bronchospasm and severe nasal symptoms within hours of a dose; urticaria (hives) and angioedema (rapid deep-tissue swelling, sometimes of the airway) occur separately. Prevalence has been quantified across 27 studies pooled in a meta-analysis. Reactions are reproducible, can be severe, and are a class effect shared with other anti-inflammatories rather than a true allergy.

Magnitude: Present in 7.2% (95% confidence interval 5.3–9.0%) of adults with asthma, rising to 14.9% in severe asthma, about 9.7% among those with nasal polyps and about 8.7% among those with chronic rhinosinusitis.

Medium 🟥 🟥

Iron-Deficiency Anaemia

Chronic occult blood loss too small to be noticed still depletes iron stores over years. In the ASPREE trial, daily 100 mg raised incident anaemia and lowered ferritin independently of any clinically recognised bleeding event (McQuilten et al.). This is a single large randomised trial, but the effect was internally consistent across haemoglobin and ferritin measurements. It is reversible on stopping and detectable long before it becomes symptomatic, which makes it one of the more monitorable harms.

Magnitude: Incident anaemia hazard ratio 1.20 (95% confidence interval 1.12–1.29); haemoglobin declined about 0.6 g/L per five years faster on aspirin, and ferritin at three years was 11.5% lower than on placebo.

Higher All-Cause and Cancer Mortality When Started After Age 70 ⚠️ Conflicted

In healthy adults over 70, aspirin was associated with more deaths overall, driven mainly by cancer deaths and by more advanced-stage cancer at diagnosis (McNeil et al., mortality; McNeil et al., cancer). No mechanism is established, and longer post-trial follow-up (Orchard et al.) has attenuated the signal, while decades-long data point the other way. Net reading: an unexplained, possibly chance finding specific to late initiation that has not been reproduced elsewhere.

Magnitude: All-cause mortality hazard ratio 1.14 (95% confidence interval 1.01–1.29), 12.7 versus 11.1 deaths per 1,000 person-years; cancer-related death hazard ratio 1.31 (1.10–1.56), with metastatic disease at diagnosis hazard ratio 1.19.

Incident Heart Failure

A patient-level pooling of six observational cohorts totalling 30,827 people at cardiovascular risk found aspirin use associated with more first heart-failure events, and the association survived propensity-score matching (pairing users with statistically similar non-users) and adjustment (Mujaj et al.). A plausible mechanism is prostaglandin-mediated sodium and water retention with a rise in afterload. Because the data are observational, confounding by indication (sicker people being likelier to be prescribed it) cannot be excluded, and no randomised trial has tested the question.

Magnitude: Incident heart failure hazard ratio 1.26 (95% confidence interval 1.12–1.41) overall and 1.27 (1.10–1.46) in the 73.6% of participants without prior cardiovascular disease.

Serious Falls Requiring Hospital Presentation

Within the ASPREE trial, participants on aspirin presented to hospital with falls more often than those on placebo, although fracture rates were unchanged (Barker et al.). The likely explanation is not more falling but more bleeding-related injury and lower thresholds for imaging after minor head trauma. The finding comes from one prespecified substudy in adults over 70 and may not extend to younger, more robust users.

Magnitude: Falls resulting in hospital presentation incidence rate ratio (the ratio of the number of events per unit of person-time in two groups) 1.17 (95% confidence interval 1.03–1.33); fractures hazard ratio 0.97, that is, no difference.

Low 🟥

Small-Intestinal Mucosal Injury

Capsule-endoscopy studies in healthy volunteers show that a fortnight of low-dose aspirin produces mucosal lesions in the small bowel in a majority of takers (Endo et al.). These lesions are usually silent, are not prevented by acid suppression, and their link to clinically important outcomes is unproven.

Magnitude: Small-bowel lesions appeared in 80% of volunteers after 14 days of low-dose aspirin versus 20% off drug in a crossover capsule-endoscopy study, with frank mucosal breaks in 30% versus none.

Raised Serum Uric Acid and Reduced Kidney Filtration

At antiplatelet doses aspirin retains urate in the renal tubule, raising serum uric acid while filtration falls alongside it (Caspi et al.). The data come from one small, uncontrolled dose-escalation study in hospitalised elderly patients, chiefly those with low albumin, so relevance to healthy long-term users is uncertain.

Magnitude: At 75 mg daily the urate excretion rate fell about 15%, with a small but statistically significant rise in serum uric acid; creatinine clearance fell in parallel and remained below baseline a week after aspirin was stopped.

Speculative 🟨

Blunting of Training-Induced Skeletal Muscle Adaptation

Cyclooxygenase inhibition suppresses prostaglandin signalling that supports post-exercise muscle protein synthesis, but the human work used ibuprofen rather than aspirin (Trappe et al.). No outcome data exist at antiplatelet doses; the basis is mechanistic.

Risk-Modifying Factors

  • Age: Bleeding risk rises steeply and non-linearly after 70, while the vascular benefit does not, which is why net harm in trials appears almost entirely in the oldest participants rather than in those aged 40–59.

  • Prior ulcer or gastrointestinal bleed: A previous peptic ulcer or upper gastrointestinal bleed is the single strongest predictor of recurrence on aspirin, raising absolute risk several-fold above the age-matched background.

  • Helicobacter pylori colonisation: This stomach bacterium multiplies ulcer-bleeding risk on aspirin; a randomised eradication trial in aspirin takers aged 60 and over reduced ulcer-bleeding hospitalisations over the first two and a half years.

  • Baseline haemoglobin, ferritin and platelet count: Low iron stores or a low platelet count before starting leave no reserve for chronic occult loss, so anaemia appears sooner and at lower blood-loss volumes.

  • Kidney function: Reduced filtration raises bleeding risk through uraemic platelet dysfunction (waste products that build up when kidneys fail and stop platelets working), and aspirin’s own transient effect on renal prostaglandins compounds it alongside diuretics.

  • Uncontrolled hypertension: Systolic pressure above roughly 160 mmHg substantially raises intracranial bleeding risk on any antiplatelet drug, making it a modifiable prerequisite rather than a fixed characteristic.

  • Sex-based differences: Absolute major-bleeding rates are somewhat lower in women than men at the same age, largely tracking lower background ulcer disease; relative risk increases from aspirin are similar in both.

  • Genetic variants: CYP2C9 poor-metaboliser alleles and UGT1A6 variants slow salicylate clearance, and G6PD deficiency (an enzyme defect that leaves red cells vulnerable to oxidative stress) raises haemolysis risk at higher salicylate exposure.

Sources: ASPREE bleeding data, García Rodríguez et al., the HEAT eradication trial, McQuilten et al. and Calderone et al..

Key Interactions & Contraindications

  • Oral anticoagulants (warfarin, apixaban, rivaroxaban, dabigatran): Caution to absolute contraindication without a specific indication; combined use roughly doubles to triples major bleeding. Mitigation: combined use is confined to a cardiologist-documented indication, with acid suppression added.

  • Other antiplatelet drugs (clopidogrel, ticagrelor, prasugrel): Caution; dual therapy raises major bleeding roughly twofold and small-bowel injury further. Mitigation: the overlap is time-limited to the interval specified after stenting, then reassessed.

  • Non-aspirin anti-inflammatories (ibuprofen, naproxen, diclofenac): Caution; these both add ulcer risk and competitively block aspirin’s access to platelet COX-1, abolishing its antiplatelet effect. Mitigation: aspirin taken at least two hours before ibuprofen, or paracetamol substituted.

  • Corticosteroids (prednisone, dexamethasone): Caution; concurrent use multiplies ulcer and bleeding risk beyond either drug alone. Mitigation: a proton pump inhibitor (an acid-suppressing drug class such as omeprazole) is co-prescribed for the duration of steroid treatment.

  • Antidepressants that block serotonin reuptake (sertraline, fluoxetine, escitalopram): Caution; these deplete platelet serotonin and raise upper gastrointestinal bleeding roughly twofold when combined with aspirin. Mitigation: acid suppression, with an antidepressant of another class preferred where feasible.

  • Methotrexate: Caution; salicylate displaces methotrexate from albumin and competes for renal tubular secretion, raising toxicity. Mitigation: the combination is avoided above 15 mg methotrexate weekly, with blood-count monitoring.

  • Urate-lowering drugs (probenecid, sulfinpyrazone): Caution; low-dose aspirin blunts their uricosuric action and can precipitate gout flares. Mitigation: allopurinol or febuxostat, which act on urate production, are preferred instead.

  • Alcohol: Caution; three or more drinks daily with aspirin multiplies upper gastrointestinal bleeding risk, an interaction carried on the drug’s own label. Mitigation: intake below that threshold, rather than added acid suppression.

  • Supplements with additive antiplatelet or anticoagulant effect: Caution; fish oil at high dose, ginkgo, garlic extract, vitamin E above 400 IU, nattokinase, curcumin and high-dose ginger each add measurably to bleeding time. Mitigation: a 7–10 day pause before any procedure.

  • Supplements that reduce risk or interact favourably: Monitor only; zinc carnosine and some probiotic strains have small randomised data suggesting reduced aspirin-induced gastric mucosal damage, while vitamin C may modestly raise salicylate levels. Mitigation: none needed beyond dose awareness.

  • Other interventions: Caution around any planned surgery, dental extraction, spinal injection, colonoscopy with polypectomy or cosmetic procedure. Mitigation: for secondary prevention, continuing aspirin usually outweighs bleeding risk; for primary prevention, a 7-day pause is standard.

Populations who should avoid Low-Dose Aspirin:

  • Children and adolescents under 16 with a viral illness, because of Reye syndrome (a rare, rapidly fatal swelling of the brain with liver failure)
  • Anyone with active peptic ulceration or gastrointestinal bleeding within the preceding 12 months
  • People with documented aspirin-exacerbated respiratory disease, aspirin-induced urticaria, or salicylate hypersensitivity
  • People with severe liver impairment (Child-Pugh Class C, the most advanced grade of liver failure) or severe kidney impairment (estimated glomerular filtration rate under 30 mL/min/1.73 m²)
  • People with an inherited bleeding disorder such as haemophilia or von Willebrand disease, or platelet count under 50 × 10⁹/L
  • Pregnant women at analgesic doses (325 mg or above) from 20 weeks’ gestation onward; the 75–150 mg preventive doses prescribed for preeclampsia risk are the exception
  • People with a prior spontaneous intracerebral haemorrhage or known cerebral amyloid angiopathy
  • People with uncontrolled hypertension above 160/100 mmHg until pressure is controlled
  • People with severe glucose-6-phosphate dehydrogenase deficiency

Risk Mitigation Strategies

  • Helicobacter pylori eradication before long-term use: A randomised trial in aspirin takers aged 60 and over showed a one-week eradication course cut ulcer-bleeding hospitalisations over the first 2.5 years; stool antigen or urea breath testing identifies carriers first.

  • Acid suppression where bleeding risk is elevated: A proton pump inhibitor alongside aspirin reduces upper gastrointestinal bleeding by roughly 70–80% and is cost-effective above about a 1% annual bleed risk; it does not protect the small bowel.

  • Uncoated rather than enteric-coated tablets: Enteric coating does not lower gastrointestinal bleeding but does cause erratic absorption and apparent non-response in 10–20% of users, so uncoated tablets remove that risk of silent treatment failure.

  • Dose matched to body weight: 75–100 mg is the dose studied below about 70 kg, and 150–325 mg above it; a mismatch produces bleeding without the vascular benefit that justifies it.

  • Blood pressure control before initiation: Systolic pressure under 140 mmHg, and certainly under 160 mmHg, is the highest-leverage way to reduce the intracranial haemorrhage risk aspirin adds.

  • Iron-status screening and re-screening: A full blood count and ferritin at baseline, at 12 months, then every 1–2 years catches the chronic occult blood loss that drives aspirin-associated anaemia before symptoms appear.

  • Investigation of anaemia rather than supplementation past it: Falling haemoglobin on aspirin warrants faecal occult blood testing and, if positive or unexplained, endoscopy, because aspirin can mask an underlying colorectal lesion.

  • Two-hour separation from ibuprofen dosing: Immediate-release aspirin taken two hours before any ibuprofen dose prevents the competitive block that silently abolishes platelet inhibition.

  • Pre-procedural pause in primary prevention: Stopping 7 days ahead allows enough platelet turnover for haemostasis; for secondary prevention, continuation usually outweighs procedural bleeding.

  • Annual reassessment of the decision after age 70: Bleeding risk climbs with age while benefit does not, so a decision that was net-positive at 62 may not be at 74.

Therapeutic Protocol

A note on the parties behind the three approaches below. The US Preventive Services Task Force is a government-appointed volunteer panel whose members draw no revenue from the recommendations they publish. The preventive-cardiology groups favouring imaging-guided allocation do derive revenue from the cardiac imaging and specialist consultations their approach generates. The functional-medicine practitioners advocating substitution derive revenue from the supplements and consultations they sell instead. None of the three is disinterested, and none is treated here as the default.

  • Standard dose: 75–100 mg once daily is the conventional antiplatelet dose worldwide (81 mg in the United States, 75 mg in the United Kingdom, 100 mg in continental Europe and Australia); higher doses add bleeding without adding platelet inhibition.

  • Formulation: Plain uncoated or chewable tablets are preferred over enteric-coated ones, which absorb erratically and produce apparent non-response; a secondary analysis of the ADAPTABLE trial found coating conferred no safety advantage.

  • Time of day: Bedtime dosing modestly blunts the morning peak in platelet reactivity that coincides with peak cardiac event rates, without changing 24-hour inhibition; a randomised crossover trial supports this but no outcome trial has tested it.

  • Half-life: Aspirin itself is cleared within about 20 minutes and salicylate within 2–3 hours, yet the antiplatelet effect lasts the platelet’s 8–10 day lifespan; duration of effect is therefore unrelated to plasma half-life.

  • Single versus split dosing: Once daily suffices for most people, but where platelet turnover is accelerated — diabetes, obesity, essential thrombocythaemia (overproduction of platelets) — thromboxane recovers before 24 hours and twice-daily dosing restores full suppression.

  • Competing approach — risk-calculator allocation: The 2022 US Preventive Services Task Force statement allocates by 10-year risk score alone: individual decision at 40–59 with risk above 10%, and against starting at 60 or above.

  • Competing approach — imaging-guided allocation: Preventive cardiologists at Harbor-UCLA and Johns Hopkins instead allocate by coronary artery calcium score, a computed-tomography measure of plaque, with scores above 100 marking a favourable balance regardless of age per a preventive-cardiology editorial.

  • Competing approach — integrative substitution: Functional-medicine practitioners including Chris Kresser argue for addressing clotting risk through diet, omega-3 intake and inflammation control first, reserving aspirin for documented arterial disease.

  • Genetic polymorphisms influencing dose choice: UGT1A6 and CYP2C9 variants alter salicylate clearance, PEAR1 variants alter residual platelet reactivity, and PIK3CA tumour status determines whether an anticancer indication exists at all.

  • Sex-based differences in dosing: No sex-specific dose is established; the apparent stroke-versus-infarction split between women and men in early trials is now largely attributed to differences in age and baseline risk rather than pharmacology.

  • Age-related considerations: Below 60 the calculus favours starting where risk is documented; between 60 and 70 it is genuinely close; above 70 initiation adds bleeding immediately while the delayed cancer benefit has no time to accrue.

  • Baseline biomarkers guiding the decision: Coronary artery calcium, lipoprotein(a), apolipoprotein B, high-sensitivity C-reactive protein and haemoglobin together determine whether expected benefit exceeds expected bleeding for a given person.

  • Pre-existing conditions guiding the decision: Established arterial disease, diabetes of long duration, chronic kidney disease and Lynch syndrome all shift the balance toward use; prior ulcer, prior brain bleed and thrombocytopenia (a low platelet count) shift it away.

Discontinuation & Cycling

  • Intended duration: Low-dose aspirin is taken indefinitely rather than in courses; both the vascular and the cancer benefits depend on continuous multi-year exposure, and the cancer effect requires at least five years before it appears.

  • Withdrawal and rebound: Stopping is followed by a transient excess of cardiovascular events in established users — a Swedish cohort of 601,527 people found a 37% higher event rate after discontinuation, concentrated in the first months.

  • Tapering: No taper is possible or useful, because the effect resides in irreversibly inhibited platelets rather than in a drug level; function returns as new platelets appear over 7–10 days.

  • Cycling: No rationale for cycling exists, and the evidence points against it; intermittent use recreates the rebound window repeatedly while forfeiting the continuous exposure the cancer benefit requires.

  • How to stop when a stop is chosen: Simple cessation is appropriate for primary prevention. In established arterial disease the decision belongs with the treating cardiologist, and a substitute antiplatelet is usually put in place.

Sourcing and Quality

  • Regulatory baseline: Aspirin is a monograph drug manufactured under pharmaceutical good manufacturing practice, so identity and content are already assured; third-party seals of the kind used for supplements add nothing here.

  • What to look for on the label: A “USP” designation confirms the tablet meets United States Pharmacopeia dissolution and content standards; “delayed release” or “safety coated” identifies the enteric formulation generally best avoided.

  • Preferred formulation: Plain uncoated 81 mg tablets, or chewable 81 mg tablets, give reliable absorption. Buffered products add calcium carbonate or magnesium oxide, which does not reduce ulcer risk but can affect other drugs’ absorption.

  • Storage and shelf life: Aspirin hydrolyses to acetic acid and salicylic acid in humidity. A vinegar smell indicates degradation and loss of potency; dry storage in the original bottle within the expiry date preserves it.

  • Reputable manufacturers: Bayer, Ecotrin and St. Joseph among branded products, and store-brand generics from major pharmacy chains, all meet the same monograph. Price differences reflect branding rather than quality.

  • Compounding: Compounding pharmacies play no role here; aspirin is inexpensive, universally stocked, and available in the exact strengths the evidence supports.

Practical Considerations

  • Time to effect: Platelet inhibition is essentially complete within 24 hours of the first dose (or within an hour if a 162–325 mg chewable loading dose is used); cardiovascular benefit accrues over years and cancer benefit only after five to ten.

  • Common pitfall — choosing enteric-coated tablets: The coating is chosen for a stomach protection it does not deliver, while causing erratic absorption that leaves 10–20% of users with no measurable platelet inhibition (Grosser et al.).

  • Common pitfall — casual ibuprofen use: A single ibuprofen dose taken before aspirin can block platelet COX-1 access for the day, and habitual use silently negates the whole intervention (Schuijt et al.).

  • Common pitfall — stopping abruptly for minor procedures: Established users who stop for a dental visit or dermatology procedure enter the rebound window described above for a bleeding risk that is usually trivial.

  • Common pitfall — assuming the label dose fits: 81 mg is a historical American convention, not a weight-adjusted dose, and it under-treats heavier people while the bleeding risk is unchanged.

  • Regulatory status: Aspirin is available over the counter worldwide without prescription and is not FDA-approved for primary prevention of a first cardiovascular event, making that particular use off-label self-medication.

  • Cost and accessibility: Aspirin costs roughly two to five US cents per day, is stocked universally, and is one of the least expensive interventions in medicine — a fact that also means no manufacturer funds promotion of it.

Interaction with Foundational Habits

  • Sleep: Indirect and mostly favourable. Aspirin does not disturb sleep architecture at antiplatelet doses, and bedtime dosing exploits the circadian morning peak in platelet reactivity. Salicylate can worsen tinnitus (ringing in the ears) at higher doses, which occasionally disturbs sleep onset; the low doses discussed here rarely do.

  • Nutrition: Direct and bidirectional. Taking aspirin with food slows absorption without reducing platelet inhibition and reduces gastric irritation. Alcohol above two to three drinks daily potentiates bleeding, high-dose fish oil, garlic and ginkgo add antiplatelet effect, and chronic use depletes iron stores, so iron-rich food matters.

  • Exercise: Direct and potentially blunting. Cyclooxygenase inhibition suppresses prostaglandin signalling that contributes to post-exercise muscle protein synthesis, though the human evidence used full anti-inflammatory doses of other drugs. In endurance athletes, aspirin combined with dehydration and heat stress raises gastrointestinal bleeding risk during long events.

  • Stress management: Indirect and weakly favourable. Aspirin does not affect cortisol or the stress axis, but chronic psychological stress raises platelet reactivity and gastric acid output, so it simultaneously enlarges the potential benefit and the ulcer risk. Stress reduction therefore complements rather than substitutes for the intervention.

Monitoring Protocol & Defining Success

Before starting, a baseline set of measurements serves two purposes: establishing whether expected benefit exceeds expected harm, and creating the reference values against which later drift is judged. That means a full blood count with ferritin to confirm iron reserve, kidney and liver panels, blood pressure documented at target, and a decision-grade risk assessment — coronary artery calcium score, lipoprotein(a), apolipoprotein B and high-sensitivity C-reactive protein. Testing for Helicobacter pylori belongs in that baseline set too, given the years of exposure that follow.

Ongoing monitoring is lighter. The full blood count and ferritin are repeated at 3 months, then at 12 months, then every 1–2 years; blood pressure every 6 months; and kidney function annually. Success is the absence of drift: stable haemoglobin, ferritin holding above baseline thresholds, no visible or occult blood, and controlled blood pressure.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Haemoglobin 135–160 g/L (men), 120–150 g/L (women) Detects chronic occult blood loss A fall of more than 10 g/L from personal baseline warrants investigation even inside the reference range
Ferritin 50–150 µg/L Iron stores deplete before haemoglobin falls Acute-phase reactant; interpreted alongside hs-CRP (high-sensitivity C-reactive protein, an inflammation marker). Conventional labs flag only below 15–30 µg/L
Platelet count 200–350 × 10⁹/L Identifies thrombocytopenia that makes aspirin unsafe Below 50 × 10⁹/L is a contraindication; above 450 × 10⁹/L raises the question of twice-daily dosing
Faecal immunochemical test Negative Screens for the occult bleeding aspirin causes and the lesion it may mask FIT (faecal immunochemical test); no dietary restriction needed, unlike older guaiac tests. Annual once on long-term aspirin
Serum uric acid 210–360 µmol/L (3.5–6.0 mg/dL) Aspirin blocks tubular urate secretion and can precipitate gout Conventional upper limit is 420 µmol/L; functional targets are lower. Fasting sample preferred
Estimated glomerular filtration rate Above 60 mL/min/1.73 m², stable year to year Falling filtration raises bleeding risk and contraindicates use below 30 eGFR (estimated glomerular filtration rate). Paired with urine albumin-to-creatinine ratio; not measured after intense exercise
Blood pressure Below 120/80 mmHg Uncontrolled hypertension is the main modifiable driver of brain bleeding on aspirin Home readings over clinic readings; conventional threshold of 140/90 is too permissive for anyone on an antiplatelet drug
High-sensitivity C-reactive protein Below 1.0 mg/L Quantifies the inflammatory risk that enlarges aspirin’s expected benefit Measurement deferred 2 weeks after any infection or hard training block; conventional labs report below 3.0 mg/L as normal
Lipoprotein(a) Below 75 nmol/L (30 mg/dL) Identifies the clot-prone phenotype most likely to benefit Largely genetic; a single lifetime measurement suffices. Non-fasting acceptable
Coronary artery calcium score 0 under age 60; above 60 no established target exists, so track the change from the individual’s own prior scan Best single discriminator of who gains more than they lose Above 100 the benefit-to-bleeding balance turns favourable at any age. Repeated no sooner than 5 years; the score is reported in Agatston units
Helicobacter pylori stool antigen Negative Colonisation multiplies ulcer-bleeding risk on aspirin Proton pump inhibitors stopped 2 weeks and antibiotics 4 weeks before testing, or the result is falsely negative

Qualitative markers worth tracking alongside the laboratory values:

  • Unusual bruising, gum bleeding when brushing, or nosebleeds lasting more than a few minutes
  • Black, tarry or unusually foul-smelling stools, or any visible blood
  • New or worsening indigestion, burning just below the breastbone, or feeling full unusually early
  • Exercise tolerance and recovery, since falling haemoglobin shows up as breathlessness before it shows up as a symptom at rest
  • New or worsening ringing in the ears, which signals salicylate accumulation
  • Wheeze, nasal congestion or flushing within hours of a dose, which signals hypersensitivity

Emerging Research

  • Add-Aspirin, adjuvant cancer trial: NCT02804815 randomises 11,000 people after curative treatment of breast, colorectal, gastro-oesophageal or prostate cancer to 100 mg, 300 mg or placebo for five years. Primary completion is estimated for October 2026 and could substantially strengthen the anticancer case.

  • COLCOT-T2D, twice-daily uncoated dosing: NCT05633810 enrols 10,000 adults aged 55–80 with type 2 diabetes and no prior coronary event, testing 40 mg non-enteric-coated aspirin twice daily and colchicine factorially. It is the first outcome trial of split uncoated dosing; completion is estimated for December 2027.

  • ASPREE-XT, post-treatment follow-up: NCT07224347 follows all 19,114 original participants after randomised treatment ended. Its cancer analysis has now reported an attenuated mortality signal and no legacy effect; the dementia, disability and bleeding endpoints continue through 2026.

  • Gastric cancer chemoprevention: NCT04214990 randomises 1,700 patients in Korea after endoscopic resection of early gastric cancer to aspirin or placebo, testing whether the colorectal chemoprevention finding extends to the upper gastrointestinal tract.

  • Genotype-directed use could narrow the indication: Martling et al., 2025 showed benefit confined to PI3K-pathway-altered tumours. If replicated, this weakens the case for undirected chemoprevention in everyone and reframes aspirin as a targeted agent for a genetically defined minority.

  • Unresolved harm signals needing randomised testing: The heart-failure association reported by Mujaj et al., 2022 rests on observational pooling and has never been tested prospectively; a positive randomised result would materially worsen the risk side of the ledger.

  • Cognitive endpoints appear closed: Ryan et al., 2020 found no effect on dementia or cognitive decline over 4.7 years, and a Cochrane review by Jordan et al., 2020 reached the same conclusion, so future research is unlikely to revive this as a reason for use.

Conclusion

Low-dose aspirin is a cheap, irreversible platelet blocker whose effects have been measured more thoroughly than almost any other drug. What the evidence supports is a clear reduction in heart attacks and clot-driven strokes in people who already have narrowed arteries, a much smaller reduction in people who do not, fewer repeat clots in the veins after a first unexplained one, and a delayed reduction in bowel cancer that takes at least five years of continuous use to appear and seems to depend on tumour biology. Set against this is a consistent increase in serious bleeding — in the gut, and less often but far more seriously in the brain — together with slow iron depletion, and an unexplained rise in deaths among people who first started it after seventy.

The evidence base is unusually strong and unusually free of commercial shaping, since no company profits from an unpatented tablet. The bodies that write the guidance are not disinterested in the same way — specialist societies and imaging-led practices earn from the pathways they endorse, as do the practitioners selling alternatives — so their positions are claims to be weighed rather than conclusions to be adopted. What separates a favourable balance from an unfavourable one is not a guideline threshold but the individual’s own artery narrowing, body weight, bleeding history, blood pressure and age at starting. Where those things have been measured, the balance is knowable; where they have not, it is guesswork in both directions.

Top - Benefits - Risks - Protocol