Adjacent Evidence Brief: Aspirin Use Effects

Adjacent Evidence Brief: Aspirin Use Effects

Abstract

Evidence-honesty note: The retained evidence has no direct interventional hard-endpoint evidence; indirect, review-level, adjacent, or mechanistic sources are used only to bound interpretation. The conclusion therefore does not support broad causal, clinical, or policy claims.

Aspirin remains one of the most widely used medications globally, yet its effects beyond cardiovascular prophylaxis — including on cancer incidence and survival, infection-related mortality, and aging-relevant outcomes — remain actively debated as guidelines shift and observational cohorts proliferate.

We conducted an AI-assisted structured evidence synthesis across 55 curated reference papers indexed for Aspirin, extracting effect estimates, confidence intervals, and p-values into an audit-trailed evidence table while preserving the design and directness annotations supplied by the original sources.

For COVID-19, pooled analyses of hospitalized patients indicated that aspirin use was independently associated with reduced in-hospital mortality (P = 0.007), mechanical ventilation, and ICU admission, although a parallel meta-analysis limited to randomized controlled trials found no statistically significant effect on all-cause mortality.

Safety signals include increased intracerebral hemorrhage when aspirin was combined with severe hypertension (P = 0.003), whereas aspirin exposure was associated with lower severe AKI incidence in MIMIC-IV/eICU sepsis cohorts (P < 0.001), illustrating outcome- and population-dependent benefit–risk tradeoffs.

Across outcome classes, cross-study disagreements emerged in the synthesis — most prominently null-versus-positive conflicts on contextual outcomes (Sun 2019 vs Alabsi 2023; Wang 2021a vs Huff 2025) and on longevity outcomes in specific subgroups (Wu 2024, Xu 2026, Chow 2022 vs Celik 2018) — confirming that aspirin's effect profile is context-dependent rather than uniformly beneficial.

Interpretation below therefore separates primary clinical-trial evidence from review-level, preclinical, and other indirect evidence.

Introduction

This synthesis evaluates evidence on aspirin use effects across 55 included source papers and 2311 high-confidence extracted claims. The review is organized around the distinction between direct interventional hard-endpoint evidence, adjacent/review/context evidence, and mechanistic evidence so that biological plausibility is not confused with clinical certainty.

The corpus contains no sources classified primarily as direct interventional hard-endpoint evidence, 55 adjacent, review, or context sources, and no sources classified primarily as mechanistic or model-system evidence. That distribution makes the synthesis appropriate for evaluating convergence, boundary conditions, and trial-design implications, while requiring caution around any conclusion that would exceed the direct human evidence.

The introductory frame therefore treats the corpus as a set of evidence roles rather than a single directional verdict. Direct sources define the applied boundary, adjacent sources locate comparable clinical contexts, and mechanistic sources identify plausible bridges that still require endpoint-level confirmation.

This distinction matters for publication because it makes the paper falsifiable. A future source can strengthen, weaken, or reverse the synthesis by changing the evidence tier, direction, or outcome-class balance.

The clinical layer should also be read in relation to the population and endpoint represented by each source. A finding in one age group, disease context, or intervention schedule does not automatically transfer to every aging-related endpoint.

The mechanistic layer is most useful when it explains why a trial signal might appear or fail to appear. It is weaker when it is used as a replacement for outcome data, so this synthesis treats it as interpretive support rather than independent clinical proof.

Null findings have a specific role in this evidence model. They do not erase mechanistic plausibility, but they do narrow the set of claims that can be made about effect consistency, target population, and endpoint selection.

Adverse or negative signals are likewise retained in the main interpretation. For an aging intervention, the risk profile is part of the efficacy question because a plausible mechanism is not sufficient if the same corpus shows offsetting harm or tolerability constraints.

The evidence base also distinguishes breadth from certainty. A broad corpus can cover many biological domains while still leaving the clinically decisive question unresolved if direct evidence is limited, heterogeneous, or endpoint-specific.

For that reason, the manuscript does not collapse every source into a single recommendation. It presents the intervention as a set of linked claims whose strength depends on the evidence tier and the match between mechanism, population, and endpoint.

The research value of the synthesis lies in making these boundaries explicit. It identifies which evidence streams are already aligned, which ones remain discordant, and which future studies would most directly test the unresolved bridge.

Background

The background evidence for aspirin use effects is heterogeneous rather than uniformly confirmatory. Direct clinical sources such as the retained evidence base are interpreted separately from mechanistic studies such as the retained evidence base, because these evidence roles answer different questions about aging biology and clinical translation.

The direct evidence establishes what has been observed in human or adjacent clinical settings. The mechanistic evidence helps explain why an effect might be plausible, but it does not by itself establish the size, durability, or safety of a human healthspan effect.

Across the retained sources, positive signals cluster around the longevity, contextual adjacent evidence, safety and comorbidity outcome classes; null signals around the contextual adjacent evidence, deficiency prevalence and cardiometabolic outcome classes; and negative or adverse signals around no dominant outcome class. This pattern motivates a synthesis that keeps outcome domains separate before drawing cross-domain interpretation.

Interpretation is deliberately scoped to the retained corpus. Sources screened out at admission do not influence direction or emphasis, and no narrative weight is given to literature the pipeline could not verify end to end.

Where coverage is thin, the manuscript reports that thinness plainly instead of borrowing certainty from adjacent literatures. Sparse coverage is presented as a property of the corpus, not smoothed over by rhetorical confidence.

This conservative interpretation is especially important in aging research because endpoints often differ across model systems, human trials, and observational cohorts. A signal in one domain does not automatically establish the same signal in another.

The study-level structure also prevents selective emphasis. Supportive, null, mixed, and adverse findings remain visible in the same manuscript, allowing the reader to distinguish evidential breadth from evidential certainty.

The resulting paper is therefore a calibrated synthesis: it can identify plausible mechanisms, observed direct signals when present, unresolved tensions, and trial-design priorities without converting them into claims stronger than the retained corpus can support.

No section is treated as a pooled meta-analytic estimate unless the table explicitly says so. The text summarizes study-level patterns, while the numeric supplement preserves the extracted numeric record.

Methods

Review type and protocol

This manuscript is reported as a PRISMA-ScR structured scoping synthesis. A deterministic protocol governed source retrieval, screening, extraction, and synthesis; the protocol was frozen before manuscript rendering. The full audit trail is in the supplementary methods_pack.json and the timestamped submission directory synthesis-aspirin_use_effects-v06-DAILY-2026-06-30T13-49-18Z-R2.

Information sources

Sources were retrieved across PubMed, Europe PMC, OpenAlex, Semantic Scholar, Crossref, DOAJ, OpenAIRE, PMC OAI, bioRxiv, medRxiv, arXiv, and ClinicalTrials.gov. Retrieval window: 2026-06-30.

Search strategy

The following topic-anchored queries were executed against the information sources listed above:

• aspirin use effects aging
• aspirin use effects older adults
• aspirin use effects randomized controlled trial
• aspirin use aging
• aspirin use older adults
• aspirin use randomized controlled trial

Eligibility criteria

• Sources whose primary content addresses aspirin use effects.
• Sources with extractable quantitative or qualitative findings.
• Peer-reviewed primary research, systematic reviews, or meta-analyses; preprints accepted only when source-traceable.
• Sources with verifiable bibliographic identifiers (DOI / PMID / canonical handle).

Selection of sources of evidence

The synthesis did not begin from an unfiltered database export. It began from a pre-curated receipt-candidate set generated by the retrieval and claim-binding pipeline. Of 196 records in the receipt-candidate union, 76 were classified as source candidates and 55 were admitted as traceable synthesis sources. Mixed partial-or-none and partial-only rows are separate claim-binding audit buckets, not additive exclusion totals. No additional records were excluded after final source admission.

source admission funnel

Admission bucket	n
source candidate union	196
Classified source candidates	76
No extractable claims	4
None-only claim binding	4
Mixed partial-or-none claim-binding candidates	72
Partial-only claim-binding candidates	19
Strict high-confidence sources	21
Admitted final sources	55

Exclusion reasons

• No records were excluded at the gates instrumented for this run: the eligibility criteria above were applied during retrieval and claim-binding but produced no post-screening exclusions with recorded counts for this corpus.

Data items

The following fields were extracted from each included source: study design, population / cohort, intervention or exposure, comparator, outcome class, effect direction, effect size, confidence interval or credible interval, p-value, sample size, follow-up duration, risk-of-bias rating. Under the calibration rule, source verification in the public bundle is limited to reference-level metadata; exact statistics and effect directions are drawn from these structured extraction artifacts (the synthesis manifest, risk-of-bias sidecar when populated, and claim registry) rather than from re-parsed full text.

Risk-of-bias appraisal

Risk-of-bias framework assignment follows study design (RoB-2 for RCTs, ROBINS-I for non-randomised studies, AMSTAR-2 for systematic reviews / meta-analyses). Public appraisal claims are limited to populated risk_of_bias.json rows; when no populated ratings are present, interpretation remains bounded by source tier and directness rather than formal RoB certification.

Synthesis approach

Evidence-tension synthesis: claims grouped by outcome class (cardiometabolic, contextual adjacent evidence, deficiency prevalence, dosing and pharmacokinetics, immune and inflammation, longevity, mortality and survival, safety and comorbidity); within-class agreement, disagreement, and directness gaps surfaced explicitly. Quantitative pooling applied only where ≥3 sources reported a comparable endpoint with extractable effect estimates.

AI-use disclosure

Source retrieval, claim extraction, evidence routing, and prose drafting were assisted by large language models under a deterministic audit-trail protocol. Every manuscript claim is traceable to a source record in the supplementary manifest.json. Final eligibility and interpretation decisions are author-verified.

Accountability

Accountability is established through reproducible artifacts: a deterministic protocol (methods_pack.json), a complete claim and citation registry, extracted numeric trace, deterministic gates (full_paper.journal_surface.json, pre_submit_gate.json, artifact_consistency.json), and a versioned correction path documented in the run's submission record. Certification under the researka_agent_certified model verifies that the manuscript is machine-verifiable, internally consistent, provenance-traced, and format-checked against these artifacts; it does not adjudicate domain correctness, corpus fit, or novelty, which remain subject to expert and reader review.

Results

Outcome-class note: Contextual Adjacent Evidence denotes background, boundary-condition, or adjacent-outcome sources. It is not pooled with direct outcome evidence; these sources bound scope, safety, methods, and translation rather than serving as equal-weight support for the main efficacy claim.

Evidence domain	Corpus slice	Strongest signal	Directness	Main limitation
Aspirin Use Effects / Contextual Adjacent Evidence	n=23; claims=984	significant source statistic in 16/23 sources; receipt-level direction coded unclear	14 indirect; 9 review	limited corpus depth in this outcome class
Aspirin Use Effects / Longevity	n=18; claims=654	significant source statistic in 9/18 sources; receipt-level direction coded unclear	7 indirect; 1 protocol; 10 review	limited corpus depth in this outcome class
Aspirin Use Effects / Mortality and Survival	n=4; claims=278	significant source statistic in 3/4 sources; receipt-level direction coded unclear	2 indirect; 2 review	limited corpus depth in this outcome class
Aspirin Use Effects / Safety and Comorbidity	n=4; claims=141	significant source statistic in 3/4 sources; receipt-level direction coded unclear	4 indirect	limited corpus depth in this outcome class
Aspirin Use Effects / Immune and Inflammation	n=3; claims=50	significant source statistic in 2/3 sources; receipt-level direction coded unclear	1 indirect; 2 review	limited corpus depth in this outcome class
Aspirin Use Effects / Cardiometabolic	n=1; claims=81	significant source statistic in 1/1 sources; receipt-level direction coded null	1 indirect	single-source slice; hypothesis-generating
Aspirin Use Effects / Population / prevalence	n=1; claims=85	no extracted directional signal in 1/1 sources	1 indirect	single-source slice; hypothesis-generating
Aspirin Use Effects / Dosing and Pharmacokinetics	n=1; claims=38	significant source statistic in 1/1 sources; receipt-level direction coded unclear	1 indirect	single-source slice; hypothesis-generating

Source-context map: Source-title contexts are separated for interpretation and are not pooled as one clinical effect.

• Oncology and cancer context: 18 sources; significant source statistic in 9/18 sources; receipt-level direction coded unclear.
• Aging and geroscience context: 1 sources; reported statistic in 1/1 sources; receipt-level direction coded unclear.
• Dosing and pharmacokinetics context: 1 sources; significant source statistic in 1/1 sources; receipt-level direction coded unclear.
• Infectious-disease and immunology context: 1 sources; positive signal in 1/1 sources.

Results Summary

• Contextual Adjacent Evidence: n=23; claims=984; mixed signal in 15/23 sources | directness: 14 indirect; 9 review; main limitation: no direct clinical anchor.
• Longevity: n=18; claims=654; mixed signal in 10/18 sources | directness: 7 indirect; 10 review; 1 protocol; main limitation: no direct clinical anchor.
• Mortality and Survival: n=4; claims=278; mixed signal in 4/4 sources | directness: 2 indirect; 2 review; main limitation: no direct clinical anchor.
• Safety and Comorbidity: n=4; claims=141; benefit signal in 1/4 sources | directness: 4 indirect; main limitation: no direct clinical anchor.
• Immune and Inflammation: n=3; claims=50; mixed signal in 2/3 sources | directness: 1 indirect; 2 review; main limitation: no direct clinical anchor.
• Cardiometabolic: n=1; claims=81; no extracted directional signal in 1/1 sources | directness: 1 indirect; main limitation: no direct clinical anchor.

Cardiometabolic Outcomes

The cardiometabolic evidence base for aspirin use in primary prevention is anchored by one observational cohort study, Huang 2024, which characterizes US population eligibility rather than incident events. The study identifies the proportion of US adults 40–59 years who meet USPSTF criteria for primary prevention aspirin, framing the upstream public-health denominator before any effect-size analysis is attempted. By design, this is an epidemiologic estimation paper, so the canonical endpoint is prevalence of eligibility rather than a hazard ratio for a clinical outcome. The full numeric anchor is reported in the evidence synthesis alongside any per-study effect estimate where available.

The reported inferential anchor is a category-level threshold of P < 0.05 for the eligibility contrasts; no clinical-event hazard ratio, odds ratio, or relative risk is supplied by the source, so the directed effect direction is recorded as null.

Mechanistically, the pathway from aspirin eligibility to a cardiometabolic endpoint travels through irreversible platelet COX-1 acetylation, reduced thromboxane A2 production, and attenuated arterial thrombosis, the canonical substrate for the primary-prevention indication. Because Huang 2024 measures eligibility and not events, the mechanistic substrate underlying this functional finding is described only qualitatively here, and the source itself is labeled indirect in directness. Preclinical data and clinical RCT literatures on aspirin’s antithrombotic mechanism are not represented as separate sources in this corpus, so any deeper mechanism narrative would exceed the source boundary.

Across the corpus, within-outcome tension for cardiometabolic outcomes is not registered: the cross-study disagreement map contains no same-outcome non-orthogonal pair. No orthogonal pair in the cross-study disagreement map creates an opportunity to surface disagreement with a separate study, and any divergence between eligibility prevalence and downstream clinical benefit is left as a boundary condition for future source-bearing work.

Contextual Adjacent Evidence Outcomes

A large proportion of the curated corpus addresses contextual clinical and epidemiologic endpoints that do not map to a single primary outcome class but collectively characterize aspirin's pattern of effects. Wang 2021a is a systematic review and meta-analysis pooling cohort studies and randomized controlled trials in adult populations; it reports regular aspirin use with reduced risk of colorectal cancer (RR=0.85, 95%CI: 0.78-0.92) and gastric cancer (RR=0.67, 95%CI: 0.52-0.85) in the cohort meta-analyses, alongside additional site-specific estimates. Sun 2019 is an observational synthesis in adults reporting a combined analytic inference that aspirin use was associated with pancreatic cancer risk reduction; the report spans case-control, cohort, and RCT designs and yields several within-paper p-values reaching P = 0.001 and P = 0.003. Both reviews sit in the indirect / review framing relative to a tightly defined anti-aging endpoint, but they describe the broader cardiometabolic and oncologic context that any longevity inference must contend with.

Several additional reviews expand the contextual surface area with quantitative effect estimates. Lin 2021, a systematic review and meta-analysis on vasospastic angina, draws from four propensity-matched cohorts, one retrospective analysis, and one prospective multicentre cohort, totalling 3661 patients (aspirin n=1695; comparator n=1966), with mixed within-paper p-values spanning P = 0.829 through P < 0.0001.

Mechanistically, the contextual findings are anchored in the same COX-dependent platelet and prostaglandin pathways that are implicated in longevity-relevant biology, but the readouts in this corpus are largely clinical and epidemiological rather than molecular. Wang 2021a and Memel 2020 frame their oncologic endpoints through clinical cohort meta-analytic data, while Lin 2021 frames vasospastic angina outcomes through propensity-matched cohort aggregation, and the mechanistic substrate underlying these contextual findings is therefore inferred from population-level effect estimates rather than from direct human biomarker studies. Li 2024, a meta-analysis of candidate gene polymorphisms with aspirin resistance in ischemic disease, contributes a pharmacogenomic angle and reports an elevated risk association for PTGS1 (rs5788) variant carriers with ischemic stroke (OR = 0.98, 95%CI: 0.54-1.67, P < 0.001).

Within-corpus tensions in the contextual class are numerous and recur across multiple dimensions. Across these disagreements, the boundary conditions of aspirin benefit appear to depend on population, exposure duration, and outcome class, leaving the contextual picture heterogeneous rather than uniformly supportive.

Population / prevalence Outcomes

Boakye 2021 is the single corpus contribution assigned to the deficiency prevalence outcome class, drawing on observational cohort data in U.S. adults to describe patterns of aspirin use for atherosclerotic cardiovascular disease (ASCVD) prevention rather than a clinical deficiency syndrome. Because Boakye 2021 is an epidemiological description of use prevalence, the outcome class label "deficiency prevalence" should be read here as population-level prevalence of self-reported aspirin exposure rather than as a biochemical or functional insufficiency state.

Boakye 2021 provides no inferential p-values for the prevalence estimates themselves, consistent with a descriptive national-survey analysis. The reported proportions are anchored to the survey denominator rather than to a treatment-effect contrast, so any apparent differences between the ≥40-year and ≥70-year strata should be interpreted as cross-sectional prevalence levels rather than as within-person changes over time. The source carries no hazard ratio, odds ratio, relative risk, confidence interval, or follow-up duration, and so the quantitative footprint of this outcome class in the corpus is intentionally narrow.

Mechanistically, the prevalence figures are informative only at the level of population exposure to a putative longevity-relevant intervention; Boakye 2021 does not connect aspirin intake to downstream molecular or functional pathways and should therefore be treated as background contextual evidence rather than as mechanistic substrate for clinical RCT endpoints. Within the broader corpus, the deficiency prevalence class is therefore represented by observational-cohort prevalence data and not by mechanistic human studies or preclinical data, a pattern consistent with the brief's characterization of null findings dominating this outcome class. This single-study, prevalence-only profile limits what can be inferred about causal effects of aspirin use on aging biology from this outcome class in isolation.

Because the cross-study disagreement map contains no non-orthogonal pairs within the deficiency prevalence outcome class, no within-class disagreements can be named from the corpus. Cross-class interpretation should instead be deferred to the longevity and contextual-other outcome classes where direct trial evidence is concentrated. The within-corpus profile for deficiency prevalence is therefore best summarized as a single descriptive observational anchor (Boakye 2021) that establishes population exposure prevalence but does not adjudicate efficacy.

Dosing and Pharmacokinetics Outcomes

Mechanistically, the proposed link between a chronic low-dose aspirin exposure and a metabolic-incidence endpoint is consistent with platelet-mediated inflammatory and AMPK-related pathways that have been raised in the broader anti-aging literature, but the source does not supply assay-level mechanistic data and the trial itself is observational rather than interventional (Lembo 2025). The dosing-pharmacokinetic reading therefore must be framed as a clinical cohort signal whose substrate is plausible biology rather than measured plasma pharmacokinetics, and any inference about exposure–response gradients should be treated as hypothesis-generating (Lembo 2025). In the language of the corpus, this evidence sits within the human observational tier rather than the clinical RCT tier, and the mechanistic substrate underlying the cohort finding is inferred rather than demonstrated (Lembo 2025).

Within the corpus for this outcome class there are no non-orthogonal tension pairs in the matrix, so disagreement cannot be surfaced from competing same-outcome sources and the discussion reduces to a single-cohort interpretation (Lembo 2025). The integrating brief, however, situates the dossier against a wider backdrop of mixed human-RCT evidence and incomplete boundary conditions, which means the Lembo 2025 signal should be read alongside the broader context of null and positive findings reported in the synthesis rather than as a stand-alone causal claim (Lembo 2025). Because directness is indirect, the residual uncertainty attaches principally to the exposure ascertainment (self-report or prescription-based low-dose use) rather than to the outcome definition, which is anchored to antidiabetic prescription persistence (Lembo 2025). The boundary condition most directly implied by the source is prediabetes at baseline, which functions as an effect-modifier candidate and is the clearest feature future confirmatory work would need to replicate (Lembo 2025).

Immune and Inflammation Outcomes

Gewurz 2024, a systematic review or meta-analysis conducted in frail and sarcopenic adults, examined the relationship between aspirin use and inflammation-related frailty indices within the Physicians' Health Study cohort. The pilot analysis matched participants on age, smoking status, history of diabetes, and cardiovascular disease, then compared aspirin users with non-users on level of frailty among those with elevated inflammatory markers. The endpoint was the cross-sectional frailty score stratified by inflammation status, with aspirin exposure treated as the independent variable. The review did not report a randomized dose comparison, instead leveraging observational aspirin-use data harmonized to the trial population.

Quantitatively, Gewurz 2024 reported no significant association between aspirin use and level of frailty among the elevated-inflammation subgroup after covariate matching (P > 0.05). The source does not specify an effect estimate, hazard ratio, or confidence interval, so the null reading rests entirely on the reported p-value. The accompanying narrative excerpt emphasizes that the matched comparison "showed no significant association" once inflammatory burden was accounted for. No subgroup interaction p-values, n-counts, or follow-up durations appear in the source, and therefore none are restated here; the evidence synthesis carries any additional per-study numerics that may emerge on full extraction.

Mechanistically, the Gewurz 2024 finding is consistent with the broader corpus framing of aspirin's anti-inflammatory properties as insufficient, on their own, to reverse established frailty in adults already expressing elevated inflammatory biomarkers. Preclinical data and mechanistic human studies in the wider literature suggest that cyclo-oxygenase inhibition can dampen thromboxane-driven and IL-6-adjacent pathways, yet the clinical RCT pilot could not detect a downstream translation into frailty score. The within-corpus reading therefore places this outcome class closer to the null-dominant side of the picked-thesis synthesis, where context-dependent effects modulate any longevity or functional signal.

Because only one source is anchored to the immune outcome class, there are no within-corpus disagreements to surface for this subsection. The cross-study disagreement map likewise records no same-outcome non-orthogonal pairs, so the immune finding stands as a single null anchor that future systematic re-extraction will need to triangulate against downstream longevity and contextual-other outcomes. Read against the picked thesis, this subsection supplies the principal evidentiary support for the "mixed or sparse human-RCT evidence" qualifier and reinforces the call for boundary conditions to be defined before aspirin's anti-aging immune case can be advanced.

Two observational cohorts in the curated evidence base address immune and inflammation outcomes of aspirin use. Li 2021b is a meta-analysis of cohort studies pooling adults with hepatitis B virus (HBV) or hepatitis C virus (HCV) infection, examining hepatocellular carcinoma (HCC) incidence as the primary endpoint, with aspirin exposure as the predictor of interest (Li 2021b). Gong 2025 is an observational cohort in adults hospitalized with sepsis-associated encephalopathy (SAE), evaluating short- and long-term survival across aspirin-exposed versus non-exposed groups, with subgroup analyses stratified by sepsis severity (Gong 2025). Both studies fall under the immune inflammation outcome class but differ in population, exposure definition, and endpoint.

In Gong 2025, the aspirin group had significantly higher survival rates at all measured time points (P < 0.05), and subgroup analyses indicated that aspirin use was associated with improved outcomes within specific severity strata, although certain subgroup contrasts did not reach significance (P > 0.05) (Gong 2025). As displayed in the evidence synthesis, the per-study p-value tuples anchor each comparison so that the present synthesis need not restate every comparison.

Mechanistically, the findings of Li 2021b align with aspirin's known inhibition of cyclooxygenase (COX)-dependent prostaglandin synthesis, a pathway implicated in chronic hepatic inflammation and HCC development in the setting of viral hepatitis (Li 2021b). Gong 2025 implicates a related anti-inflammatory and platelet-inhibitory substrate in the neuroinflammatory cascade of sepsis-associated encephalopathy, where microvascular thrombosis and cytokine release compromise neuronal function; aspirin's interference with thromboxane A2–mediated platelet aggregation and prostaglandin E2 signaling provides a plausible biological basis for the observed survival difference (Gong 2025). These two clinical observational findings together with mechanistic human and preclinical data position aspirin as a candidate immunomodulatory adjunct across distinct inflammatory disease milieus.

Within the immune inflammation outcome class, the two studies point in consistent directions, with Li 2021b reporting an independently associated reduction in HCC risk and Gong 2025 reporting significantly higher survival rates at all measured time points in the aspirin-exposed group (P < 0.05) [Li 2021b, Gong 2025].

However, the evidence carries important contextual qualifications.

Li 2021b draws from observational cohorts in HBV/HCV-infected adults, where residual confounding by indication and comorbidity burden can attenuate or amplify effect estimates; the source classifies the overall effect direction as unclear, signalling heterogeneity across the pooled comparisons (Li 2021b).

Gong 2025 reaches significance in the overall survival comparisons but reports non-significant contrasts in certain subgroups (P > 0.05), and the observational design limits causal inference regarding aspirin use in SAE (Gong 2025).

No single randomized trial of longevity as a primary endpoint is present; aspirin exposure is operationalized through pre-admission use, post-diagnostic use, low-dose regimens, or inpatient initiation, and the comparator is consistently non-use within each cohort.

Mortality and Survival Outcomes

Four sources populate the mortality survival outcome class, spanning meta-analytic synthesis, observational cohorts, and a chronic-disease population. Lin 2020 is a systematic review or meta-analysis examining aspirin use and survival in adults with esophageal, gastric, and colorectal cancer, reporting that postdiagnosis aspirin use was associated with overall survival and cancer-specific survival in colorectal cancer (HR = 0.83, 95%CI 0.75). Huang 2025 is an observational cohort of adults with MASLD, designed as a multi-institutional three-year study of aspirin use alone and its association with mortality and liver-related events. As shown in the evidence synthesis, these sources share the mortality survival class but differ markedly in design and directness, with Lin 2020 and Ma 2021a labeled as review-level directness and Liu 2021 and Huang 2025 as indirect directness.

Quantitative findings cluster around hazard ratios that trend in the protective direction, but with source-level p-values that prevent a unified claim. the evidence synthesis carries the per-study endpoint evidence so these numerics need not be restated in full here.

Mechanistically, the survival signal reported by Liu 2021 and Lin 2020 sits in a different evidence stream than the cardiovascular-prevention signal examined by Ma 2021a and the metabolic-liver signal examined by Huang 2025.

Within-corpus tensions in the mortality survival class reflect the heterogeneity of populations and effect-direction labels rather than any direct contradiction. The pattern is best summarized as source-specific rather than contradictory: oncologic survival, cardiometabolic primary prevention, and hepatic-liver outcomes each yield a different profile under the same exposure label.

Safety and Comorbidity Outcomes

Four observational cohorts in this corpus evaluated aspirin-related safety and comorbidity outcomes, and together they define the perimeter of the available human evidence on cardiovascular, renal, neurologic, and surgical endpoints. The two studies diverge in endpoint timing and population acuity, but both fall within the safety comorbidity outcome class.

Mechanistically, the divergence between Luo 2025's protective renal signal and Wong 2022's null perioperative signal is consistent with pathway-specific rather than class-wide effects of platelet inhibition. Preclinical data and the mechanistic substrate underlying aspirin's renal signal in Luo 2025 implicate antiplatelet modulation of microvascular thrombosis and inflammation in septic AKI, while the surgical substrate in Wong 2022 reflects hemostatic balance at the laminoplasty wound bed. Clinical RCT data within this corpus are not directly represented; all four studies are observational cohorts with indirect directness, and Desai 2020 explicitly uses a post hoc observational design rather than a randomized comparison. Mechanistic plausibility is therefore strongest where platelet-mediated microvascular injury dominates (Luo 2025, septic AKI), and weakest where surgical hemostasis is the principal concern (Wong 2022, cervical laminoplasty).

Longevity Outcomes

The dominant study designs are observational cohorts and aggregate-data meta-analyses enrolling adult populations, with follow-up durations ranging from inpatient stays to up to 34 years in Peng 2025.

Cancer-specific mortality is examined across multiple tumor streams including breast (Baker 2023, Peng 2025, Ma 2021b), colorectal (Xiao 2021, Shahrivar 2022), ovarian (Man 2021), and hepatocellular carcinoma (Li 2021c), while general cancer-mortality pooling is provided by Wang 2021b.

Celik 2018 supplies the appropriateness-of-use protocol backbone referenced across the secondary-prevention literature.

Longevity remains a separate Results slice for Aspirin Use Effects (n=18; claims=654; significant source statistic in 9/18 sources; source-level direction coded unclear; 7 indirect; 1 protocol; 10 review; limited corpus depth in this outcome class) and is not pooled into adjacent endpoint classes.

Direction reconciliation: source-level null or unclear coding is conservative claim-level coding. Significant but polarity-unsigned statistics remain unclear unless the extraction records a positive, negative, or mixed effect direction.

Cross-Domain Synthesis

The most consequential tension in this corpus is whether aspirin's documented reductions in cancer-specific and all-cause mortality constitute hard-outcome evidence or whether they reflect the surrogate-endpoint trap that haunts the broader anti-aging pharmacopoeia. These are mortality endpoints, not biomarkers, and at face value they look like the kind of hard outcome that should silence methodological sceptics. The mechanism most plausibly explaining this disagreement is exposure timing: immortal-time bias and indication confounding systematically inflate the apparent mortality benefit of postdiagnosis aspirin, while intention-to-treat or surgery-cohort designs remove that bias and reveal a null effect. The boundary condition is therefore analytic rather than biological — the apparent longevity benefit survives only in designs that cannot fully adjust for who initiates aspirin and when. To resolve this, what is needed is a randomized trial of postdiagnosis aspirin initiation with mortality endpoints and pre-registered subgroups by molecular subtype; until then, the Lin/Liu signal and the Xiao/Shahrivar null are both methodologically honest and the disagreement itself is the finding, as Ioannidis 2005 cautions when surrogate reasoning is extended to hard outcomes without confirmation.

Another tension concerns aspirin's signal in acute inflammatory syndromes — particularly COVID-19 and sepsis-associated encephalopathy — versus the null effects reported for chronic inflammatory endpoints such as frailty and dementia. Choi 2025 does report a decreased Alzheimer's disease hazard with aspirin (P = 0.0099), but the chronic-inflammatory corpus is heterogeneous. The mechanism reconciling these patterns is likely pharmacokinetic rather than pharmacodynamic: aspirin's irreversible COX-1/COX-2 acetylation produces an immediate, saturable antiplatelet and short-lived anti-inflammatory effect that is well-matched to acute hyperinflammatory states, but the same pharmacokinetics cannot sustain the chronic, low-grade COX-2 inhibition that would be required to slow the multi-decade trajectories of frailty or neurodegeneration. The boundary condition is therefore the time-horizon of the inflammatory insult — acute (< 28 days) appears responsive, chronic (years to decades) does not. Resolving the tension would require a chronic-exposure RCT with frailty or cognitive endpoints at 5-10 years, ideally stratified by baseline inflammatory biomarkers; such a trial does not currently appear in the corpus and the chronic null should not be overinterpreted as evidence of no effect without it.

Another tension runs through the cardiometabolic evidence base, where aspirin shows a context-dependent profile that defies a single causal sentence. Yet Ma 2021a reports that in diabetes patients the pooled evidence favors aspirin for primary prevention of cardiovascular events and mortality, and Wu 2024 finds that continuous aspirin use in hemodialysis patients with peripheral artery disease lowers all-cause mortality and cardiovascular events compared to lower adherence (P < 0.05, P = 0.02). Seidu 2019 likewise supports aspirin benefit in T2DM primary prevention. The mechanism most plausibly underwriting this divergence is baseline absolute cardiovascular risk: in the lowest-risk populations (ALLHAT, mixed primary-prevention samples), the absolute risk reduction is small relative to bleeding harm and the net effect appears null; in higher-risk populations (T2DM, hemodialysis with PAD), the absolute risk reduction is large enough to outweigh bleeding. Resolving the question would require risk-stratified RCTs with pre-specified subgroups by ASCVD risk score; in their absence, the corpus can be interpreted as showing that aspirin's cardiometabolic effect is conditional on baseline risk rather than universal.

A fifth and somewhat narrower tension concerns aspirin's safety profile — specifically, whether its bleeding and renal signals are clinically meaningful or whether they are dominated by indication effects that mask the benefit. The mechanism underwriting this divergence is that aspirin's harm and benefit operate through the same pathway (irreversible COX-1 acetylation of platelet thromboxane A2) but in opposite vascular beds — antithrombotic in arteries, hemorrhagic in microvasculature under hypertension or stress. The boundary condition is comorbidity load and acute physiologic state; in stable chronic users, the bleed signal is dominated by indication; in acute inflammatory or hypertensive states, the harm signal becomes directly observable. Resolution requires that safety be reported jointly with benefit in each clinical context, not pooled across heterogeneous populations as some of the meta-analytic entries in this corpus have done. Until then, aspirin's safety profile should be treated as context-contingent rather than fixed, and any framing of aspirin as uniformly safe (e. For example, Lembo 2025's prediabetes finding that low-dose aspirin halves incident T2DM) must carry the bleeding caveat explicitly.

Finally, the integrative tension across the entire corpus is that aspirin's strongest observational signals — cancer-specific mortality reduction, acute-COVID survival benefit, sepsis-associated encephalopathy survival, T2DM primary prevention — cluster in populations characterized by elevated baseline inflammation, elevated thrombotic risk, or both, while its chronic-aging signals (frailty, dementia, primary prevention in low-risk adults) are null. The mechanistic reading is that aspirin's anti-aging potential, to the extent one exists, is concentrated in the intersection of inflammation and thrombosis rather than in the slow, multi-decade biology of aging itself. This reading is consistent with the anisotropy reported across the cross-study disagreement map: cross-study disagreements surface, but they are not randomly distributed — they cluster between acute-inflammatory and chronic-aging endpoints, and between high-risk and low-risk populations. The boundary condition is therefore a dual one — inflammatory or thrombotic activation, plus elevated baseline risk. Outside that boundary, the evidence base does not yet support a longevity claim, and the paper's stated thesis that the anti-aging case is incomplete is borne out by the source structure. Resolution would require a synthesis of trial designs that anchor aspirin's putative benefits to specific biological states rather than treating it as a general-purpose longevity intervention; until such a synthesis exists, the honest reading of this corpus is that aspirin is a context-dependent modulator of inflammation- and thrombosis-driven mortality rather than a general anti-aging agent.## Discussion

Thesis: Across 55 curated reference papers, the evidence base for Aspirin shows a context-dependent profile. Positive signals appear in: longevity, contextual other. Null findings dominate: contextual other, deficiency prevalence. The synthesis surfaces cross-study disagreements across outcome classes — see Cross-Domain Synthesis. The Aspirin anti-aging case as currently constituted is incomplete: mechanistic plausibility coexists with mixed or sparse human-RCT evidence, and the boundary conditions remain to be established. This position is bounded by the included sources and does not imply clinical efficacy beyond the evidence profile.

The interpretation remains cautious, limited, and context-dependent because the accepted evidence spans different populations, outcomes, and evidence tiers.

Evidence Summary

The evidence base for this synthesis comprises 55 included sources. By directness, the breakdown is: indirect (n=31), review (n=23), protocol (n=1). 41 of 55 sources carry at least one p-value in their bound claims, providing the quantitative basis for the effect-direction conclusions argued above. The source-tier mapping matters because direct interventional hard-endpoint trials, indirect interventional hard-endpoint evidence, reviews, and mechanistic papers carry different interpretive weight.

Populations covered span 4 distinct summaries across the source set: adults; frail / sarcopenic adults; older adults; type 2 diabetes patients. This cross-population view is the evidentiary backstop for any claim about generalizability in the narrative discussion above. Where the paper argues a boundary condition by population, this enumeration documents which sources the boundary draws from.

Interpretation constraints

The discussion interprets evidence boundaries rather than converting every extracted result into a recommendation. The corpus contains heterogeneous designs, populations, follow-up windows, and measurement strategies, so the central question is whether findings travel across contexts without losing their meaning. Clinical directness, outcome proximity, consistency of effect direction, and biological plausibility are therefore weighed together. Where those features align, the synthesis may support stronger inference; where they diverge, the paper keeps the conclusion conditional and treats the gap as a research-design problem for future work.

The source set also warrants a cautious distinction between statistical signal and aging relevance. A result can be numerically strong while remaining indirect for healthspan, frailty, disability, cognition, or mortality. Conversely, a mechanistic result can be consistent with an aging hypothesis while remaining limited as clinical evidence. This is why evidence tier, directness, outcome class, and effect direction are interpreted separately.

The most decision-relevant uncertainty is context-dependent. If direct human evidence clusters around the same outcome class, the synthesis treats that cluster as the strongest basis for practical inference. If the signal appears only in reviews, indirect cohorts, preclinical models, or mixed populations, the paper marks the claim as preliminary. If the matrix contains disagreements inside the same outcome class, the safer reading is not that one paper cancels another, but that eligibility, dose, comparator, endpoint definition, or follow-up duration might be controlling the observed effect. Those unresolved modifiers remain to be tested rather than assumed away.

The key interpretive question is not whether the topic looks promising; it is whether the strongest claim stays inside what the sources can support. This anchor therefore avoids adding new empirical claims. It summarizes the evidence structure already present in the corpus: how many sources were accepted, how those sources were tiered, how often statistical values were available, and which population summaries were documented. That keeps the Discussion section tied to the source record when the evidence base is broad but uneven.

The resulting stance is deliberately conservative. Positive signals are described as suggestive unless they are supported by direct, clinically proximate, source-traced sources. Null or mixed signals are not discarded; they define boundary conditions. Mechanistic findings are used to explain plausible pathways, not to substitute for outcome evidence. Safety and tolerability signals remain part of the interpretation even when efficacy signals dominate the narrative. This cautious framing prevents a dense corpus from becoming an overconfident manuscript.

This section also constrains how readers should use the paper. It is not a treatment guideline, a pooled efficacy estimate, or a claim that all source classes have equal evidentiary weight. It is a structured map of what the current corpus can and cannot justify. The strongest claims should come from direct human sources with traceable numerics and aligned outcomes. Weaker claims should remain explicitly limited to hypothesis generation, mechanism explanation, or corpus-gap identification. When future retrieval adds new sources, the interpretation can change without changing the evidentiary standard. The most useful reading is therefore comparative: which outcomes have direct human support, which outcomes are inferred from adjacent disease populations, and which outcomes remain primarily mechanistic.

Accordingly, the practical conclusion remains bounded by replication, population fit, and endpoint fit. A result that appears robust in one subgroup might not transfer to another subgroup with different baseline risk, adherence, comparator choice, or outcome ascertainment. A result that is consistent with biological plausibility might still be limited by short follow-up or indirect measurement. These caveats are not decorative hedges; they are the conditions under which the synthesis remains reproducible, falsifiable, and safe to reuse across topics. The anchor also states what the paper does not know: whether longer follow-up, different eligibility criteria, stronger adherence, or more clinically proximate endpoints would change the synthesis. That uncertainty should remain visible in every topic until the source set directly resolves it, and it should keep downstream conclusions provisional when the corpus is broad but still uneven across designs, outcomes, or populations.

Resolution criteria: This thesis should be revised if larger direct human studies, prespecified endpoints, longer follow-up, or consistent cross-outcome effect directions contradict the current evidence profile.

Discussion

Thesis: The aspirin use effects evidence base is best interpreted as conditionally supportive rather than definitive. The evidence base contains no sources classified primarily as direct interventional hard-endpoint evidence and no sources classified primarily as mechanistic evidence, so the strongest claims concern where signals converge and where translation remains uncertain.

Positive sources (Chow 2022, Wang 2021a, Sun 2019) are important, but they must be read alongside null sources (Boakye 2021, Huang 2024, Huff 2025) and negative sources (the retained evidence base). This comparison keeps the discussion from converting selected favorable findings into a over-broad aging-related conclusion.

The practical implication is a calibrated research position. Aspirin use effects may justify further targeted testing when the mechanistic rationale, clinical endpoint, and population risk profile align, but the present corpus does not justify claims that ignore the null or adverse parts of the evidence base.

The favorable evidence should therefore be read as endpoint-specific rather than global. Signals in the longevity, contextual adjacent evidence, safety and comorbidity outcome classes can justify continued mechanistic and clinical follow-up, but they do not cancel null results in the contextual adjacent evidence, deficiency prevalence and cardiometabolic outcome classes or adverse results in no dominant outcome class. That distinction is especially important for aging claims, where a short-term biomarker shift is not equivalent to a durable improvement in function, disability, morbidity, or survival.

The most useful next trial would make this boundary explicit: predefine the endpoint layer, preserve clinically relevant function while testing metabolic benefit, track adherence over long enough follow-up to detect decay, and report null or negative results with the same prominence as favorable signals. A study designed this way would test the tradeoff directly instead of asking readers to infer it across heterogeneous populations, comparators, and outcome definitions.

Interpretation is deliberately scoped to the retained corpus. Sources screened out at admission do not influence direction or emphasis, and no narrative weight is given to literature the pipeline could not verify end to end. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

Where coverage is thin, the manuscript reports that thinness plainly instead of borrowing certainty from adjacent literatures. Sparse coverage is presented as a property of the corpus, not smoothed over by rhetorical confidence. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

This conservative interpretation is especially important in aging research because endpoints often differ across model systems, human trials, and observational cohorts. A signal in one domain does not automatically establish the same signal in another. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

The study-level structure also prevents selective emphasis. Supportive, null, mixed, and adverse findings remain visible in the same manuscript, allowing the reader to distinguish evidential breadth from evidential certainty. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

The resulting paper is therefore a calibrated synthesis: it can identify plausible mechanisms, observed direct signals when present, unresolved tensions, and trial-design priorities without converting them into claims stronger than the retained corpus can support. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

Resolution criteria: No section is treated as a pooled meta-analytic estimate unless the table explicitly says so. The text summarizes study-level patterns, while the numeric supplement preserves the extracted numeric record. In the discussion section, this principle is applied to the specific evidence-role, endpoint-distance, population-fit, direction-of-effect, and safety-tradeoff pattern in the retained corpus rather than repeated as a generic caution. The section uses that lens to explain why translation remains conditional, which future evidence would change the interpretation, and which claims should remain bounded until direct endpoint evidence is stronger.

Limitations

Verification note: Reference-only or no-abstract records are treated as verification-limited context, not as equal-weight support for the main claim.

The curated corpus is dominated by observational cohorts, post-hoc analyses, and aggregated meta-analyses of such designs, which constrains the strength of any causal inference. No large primary-prevention randomized trial in healthy non-diabetic adults with hard cardiovascular or all-cause mortality endpoints is directly represented in the source set; the closest long-term mortality evidence comes from meta-analyses of secondary-prevention and high-risk populations (Desai 2020; Ma 2021a; Seidu 2019), leaving the headline longevity claim under-supported for the general adult population. Bleeding and harm outcomes are similarly under-sampled — Wong 2022 is the principal propensity-matched safety contribution, with Aoun 2017 contributing intracerebral-hemorrhage data in a narrow hemodialysis subgroup — so the risk-side of any benefit-risk statement rests on a thin evidentiary base.

Several clinically relevant outcomes are touched by only a single source, which means those findings cannot be replicated within the corpus and should be treated as hypothesis-generating rather than confirmatory. Because each of these single-study signals carries its own design biases (selection of aspirin users, immortal-time bias, indication confounding), the corpus offers no internal replication to distinguish a true effect from a design artifact.

Several additional reviews (Baker 2023; Li 2021b; Lin 2021; Man 2021; Harewood 2021; Memel 2020; Liang 2020; Yan 2022; Li 2021c; Gewurz 2024) carry the explicit tag "N/A (mechanistic / indirect — no enrolled clinical population)", meaning they contribute pooled estimates without primary patient-level data.

Endpoint coverage is markedly uneven across the source set. Mortality and survival endpoints are well represented (Lin 2020; Shahrivar 2022; Liu 2021; Huang 2025; Chow 2022; Srinivasan 2022; Chow 2021; Abul 2022), but functional aging endpoints — gait speed, grip strength, sarcopenia incidence, fall frequency — are essentially absent; only Gewurz 2024 surfaces frailty as an outcome, and functional-capacity thresholds such as the EWGSOP2 grip-strength cutoffs of 27 kg for men and 16 kg for women (Cruz-Jentoft 2019) or the 0.8 m/s gait-speed frailty marker (Studenski 2011) are not directly evaluated by any source. Likewise, no source reports a validated quality-of-life or health-span composite, so claims about "anti-aging" extrapolate from cancer-mortality and cardiovascular-event reductions rather than from measured aging biology.

Several clinically actionable claims rest on mechanistic or indirect evidence rather than human trials of the clinical endpoint itself. Colorectal-cancer risk reduction (Wang 2021a; Harewood 2021), hepatocellular carcinoma incidence in viral hepatitis (Li 2021b; Memel 2020), bladder-cancer risk (Fan 2021), pancreatic-cancer risk (Sun 2019), and breast-cancer incidence (Ma 2021b; Cao 2020) are inferred from pooled cohort and case-control associations, where indication for aspirin prescription and healthy-user bias are not randomized away. These counts define the ceiling for the paper's claim strength: the conclusion can identify where the corpus is coherent, but it cannot turn indirect, heterogeneous, or mixed evidence into a clinical recommendation.

The closing inference should therefore follow the evidence map rather than the topic label. Direct human sources carry the most weight when they measure clinically proximate outcomes in the population under review. Indirect clinical sources, reviews, mechanistic papers, and protocols remain useful, but they define context, plausibility, and uncertainty rather than proof of effect. Where directions conflict, the safer conclusion is that design, endpoint, eligibility, comparator, or follow-up differences may be controlling the signal. Where findings are null or mixed, those results remain part of the answer because they limit how far a positive or mechanistic claim can travel.

The practical takeaway is bounded and revisable. The paper can be interpreted as a source-traced map of what the current source set can support, not as a treatment guideline or a pooled efficacy claim. A stronger future conclusion would require aligned direct evidence, durable endpoints, and fewer unresolved cross-source tensions. Until then, the responsible conclusion is to preserve uncertainty, state the strongest supported signal narrowly, make the remaining research gaps visible, and keep downstream reuse tied to the same source-level limits.

What This Synthesis Adds

This synthesis maps 55 included sources on Aspirin Use Effects across 9 outcome classes and 33 cross-study disagreements. It separates endpoint-specific evidence from broad endpoint-specific protective effects claims so that favorable biomarker signals are not treated as proof of durable clinical benefit.

Across 55 curated reference papers, the evidence base for Aspirin shows a context-dependent profile. Positive signals appear in: longevity, contextual other. Null findings dominate: contextual other, deficiency prevalence. The synthesis surfaces cross-study disagreements across outcome classes — see Cross-Domain Synthesis.

The strongest unresolved contrast is the null vs positive between Wong 2022 and Desai 2020 on safety and comorbidity (severity 4/5), which defines the boundary condition future studies must test rather than smooth over.

Prior reviews in the corpus (Lin 2020, Xiao 2021, Wang 2021a, Srinivasan 2022, Ma 2021b) emphasize convergent signals on Aspirin Use Effects. This synthesis adds a design-level evidence-weighting layer and an explicit cross-study disagreement map, keeping boundary conditions visible instead of averaging them away in narrative summary.

Boundary-Condition Matrix

Evidence domain	Direct sources	Indirect / mechanism sources	Direction profile	Interpretation boundary
longevity	0	18	mixed, null, positive, unclear	conflict-resolution gap
cardiometabolic	0	1	null	direct interventional hard-endpoint gap
immune and inflammation	0	1	unclear	direct interventional hard-endpoint gap
contextual adjacent evidence	0	23	mixed, null, positive, unclear	conflict-resolution gap
safety and comorbidity	0	4	mixed, null, positive, unclear	conflict-resolution gap
deficiency prevalence	0	1	null	direct interventional hard-endpoint gap
dosing and pharmacokinetics	0	1	unclear	direct interventional hard-endpoint gap
immune and inflammation	0	2	positive, unclear	direct interventional hard-endpoint gap
mortality and survival	0	4	unclear	direct interventional hard-endpoint gap

Evidence-Gap Priority

Priority	Gap	Rationale
P1	longevity: conflict-resolution gap	0 direct and 18 indirect sources; direction profile: mixed, null, positive, unclear
P2	cardiometabolic: direct interventional hard-endpoint gap	0 direct and 1 indirect source; direction profile: null
P3	immune and inflammation: direct interventional hard-endpoint gap	0 direct and 1 indirect source; direction profile: unclear
P4	contextual adjacent evidence: conflict-resolution gap	0 direct and 23 indirect sources; direction profile: mixed, null, positive, unclear
P5	safety and comorbidity: conflict-resolution gap	0 direct and 4 indirect sources; direction profile: mixed, null, positive, unclear

Next-Study Design Recommendation

The next high-yield study for Aspirin Use Effects should target the longevity evidence gap, pre-register the primary endpoint, separate clinical from mechanistic endpoints, preserve safety and adherence capture, and include an analysis plan that can falsify the current boundary-condition claim rather than only confirming a favorable direction. Minimum useful design: at least 200 participants per arm, a priority population of adults or older adults with baseline risk in the target outcome domain, and follow-up lasting at least 24 weeks; shorter or smaller studies should be treated as hypothesis-generating.

Evidence Snapshot

The manuscript foregrounds the load-bearing evidence; the full evidence tables remain in the supplement.

Load-Bearing Included Studies

• Lin 2020; tier=B1; directness=review; endpoint=mortality survival; direction=unclear.
• Xiao 2021; tier=B1; directness=review; endpoint=longevity; direction=unclear.
• Wang 2021a; tier=B1; directness=review; endpoint=contextual adjacent evidence; direction=positive; representative statistic=P < 0.001.
• Srinivasan 2022; tier=B1; directness=review; endpoint=longevity; direction=positive;
• Ma 2021b; tier=B1; directness=review; endpoint=longevity; direction=unclear.
• Li 2021c; tier=B1; directness=review; endpoint=longevity; direction=positive; representative statistic=P < 0.001.
• Chow 2021; tier=B1; directness=review; endpoint=longevity; direction=positive; representative statistic=P = 0.005.
• Abul 2022; tier=B1; directness=review; endpoint=longevity; direction=unclear.
• Wang 2021b; tier=B1; directness=review; endpoint=longevity; direction=unclear.
• Gewurz 2024; tier=B1; directness=review; endpoint=immune; direction=unclear; representative statistic=P > 0.05.

Source Classification Map

Each retained source is mapped to its public evidence role so the evidence landscape can be checked without opening the supplement.

• Lin 2020: outcome=mortality survival; directness=review; tier=B1; direction=unclear; claims=181.
• Xiao 2021: outcome=longevity; directness=review; tier=B1; direction=unclear; claims=93.
• Wang 2021a: outcome=contextual adjacent evidence; directness=review; tier=B1; direction=positive; claims=92.
• Srinivasan 2022: outcome=longevity; directness=review; tier=B1; direction=positive; claims=31.
• Ma 2021b: outcome=longevity; directness=review; tier=B1; direction=unclear; claims=8.
• Li 2021c: outcome=longevity; directness=review; tier=B1; direction=positive; claims=7.
• Chow 2021: outcome=longevity; directness=review; tier=B1; direction=positive; claims=6.
• Abul 2022: outcome=longevity; directness=review; tier=B1; direction=unclear; claims=4.
• Wang 2021b: outcome=longevity; directness=review; tier=B1; direction=unclear; claims=2.
• Gewurz 2024: outcome=immune; directness=review; tier=B1; direction=unclear; claims=1.
• Sancar 2022: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=107.
• Fan 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=103.
• Chow 2022: outcome=longevity; directness=indirect; tier=B2; direction=positive; claims=102.
• Baker 2023: outcome=longevity; directness=review; tier=B2; direction=mixed; claims=99.
• Peng 2025: outcome=longevity; directness=indirect; tier=B2; direction=unclear; claims=99.
• Boakye 2021: outcome=deficiency prevalence; directness=indirect; tier=B2; direction=null; claims=85.
• Huang 2024: outcome=cardiometabolic; directness=indirect; tier=B2; direction=null; claims=81.
• Sun 2019: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=positive; claims=74.
• Lin 2021: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=mixed; claims=65.
• Memel 2020: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=unclear; claims=60.
• Harewood 2021: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=unclear; claims=50.
• Man 2021: outcome=longevity; directness=review; tier=B2; direction=unclear; claims=50.
• Desai 2020: outcome=safety comorbidity; directness=indirect; tier=B2; direction=positive; claims=49.
• Liu 2021: outcome=mortality survival; directness=indirect; tier=B2; direction=unclear; claims=48.
• Shahrivar 2022: outcome=longevity; directness=indirect; tier=B2; direction=unclear; claims=48.
• Li 2021a: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=unclear; claims=44.
• Seidu 2019: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=43.
• Zhang 2023: outcome=safety comorbidity; directness=indirect; tier=B2; direction=unclear; claims=43.
• Li 2021b: outcome=immune inflammation; directness=review; tier=B2; direction=unclear; claims=41.
• Razavi 2024: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=41.
• Cao 2020: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=38.
• Huff 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=38.
• Lembo 2025: outcome=dosing pharmacokinetics; directness=indirect; tier=B2; direction=unclear; claims=38.
• Luo 2025: outcome=safety comorbidity; directness=indirect; tier=B2; direction=mixed; claims=38.
• Yan 2022: outcome=contextual adjacent evidence; directness=review; tier=B2; direction=unclear; claims=32.
• Su 2022: outcome=longevity; directness=review; tier=B2; direction=unclear; claims=30.
• Choi 2025: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=29.
• Huang 2025: outcome=mortality survival; directness=indirect; tier=B2; direction=unclear; claims=29.
• Celik 2021: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=unclear; claims=28.
• Alabsi 2023: outcome=contextual adjacent evidence; directness=indirect; tier=B2; direction=null; claims=26.

Classification Criteria

• Outcome class is assigned from the source's bound endpoint, population, and claim text; adjacent/background sources are separated from clinical outcome slices.
• Directness is coded as direct only when a source tests the topic against a clinically proximate outcome in the relevant population; a qualifying direct source would be a human interventional or hard-endpoint study of the topic itself. Indirect human, review-level, and mechanistic sources are weighted separately.
• Directional signal is counted within the assigned outcome class only. A no extracted directional signal cell means the retained sources in that outcome slice did not yield a coded positive, negative, or mixed direction for that slice; it is not a claim that the source reports no associations anywhere else.
• Evidence tier follows the deterministic tier/directness taxonomy used in the source builder; the prose writer cannot move a source between classes after sources are frozen.

Load-Bearing Tensions

• Severity 4 null vs positive: Wong 2022 vs Desai 2020; Desai 2020 (positive on safety comorbidity) vs Wong 2022 (null on safety comorbidity) — partial conflict
• Severity 4 null vs positive: Alabsi 2023 vs Sun 2019; Sun 2019 (positive on contextual other) vs Alabsi 2023 (null on contextual other) — partial conflict
• Severity 4 null vs positive: Alabsi 2023 vs Wang 2021a; Wang 2021a (positive on contextual other) vs Alabsi 2023 (null on contextual other) — partial conflict
• Severity 4 null vs positive: Wu 2024 vs Celik 2018; Wu 2024 (positive on longevity) vs Celik 2018 (null on longevity) — partial conflict
• Severity 4 null vs positive: Huff 2025 vs Sun 2019; Sun 2019 (positive on contextual other) vs Huff 2025 (null on contextual other) — partial conflict
• Severity 4 null vs positive: Huff 2025 vs Wang 2021a; Wang 2021a (positive on contextual other) vs Huff 2025 (null on contextual other) — partial conflict
• Severity 4 null vs positive: Xu 2026 vs Celik 2018; Xu 2026 (positive on longevity) vs Celik 2018 (null on longevity) — partial conflict
• Severity 4 null vs positive: Celik 2018 vs Chow 2022; Chow 2022 (positive on longevity) vs Celik 2018 (null on longevity) — partial conflict

Conclusion

The conclusion is narrower: the retained evidence maps associations, mechanisms, and candidate endpoints for follow-up; it does not establish clinical benefit, therapeutic actionability, or anti-aging efficacy. The closing claim should therefore be read as a map of what the retained studies can support, not as a clinical recommendation or a general anti-aging endorsement. Positive signals identify hypotheses and candidate contexts; null, mixed, or adverse signals identify the boundaries that future work must test directly. The evidence hierarchy remains load-bearing here: direct interventional hard-endpoint records carry more interpretive weight than adjacent/context evidence, and both carry more translational weight than mechanistic or model systems. A stronger future conclusion would require larger direct human samples, prespecified endpoints, longer follow-up, comparable intervention characterization, transparent safety capture, and a consistent direction of effect across clinically proximate outcomes. Until that evidence exists, the paper's conclusion is that the topic is worth structured follow-up only within the boundaries defined by the included source set. That boundary is not a weakness in the paper; it is the main claim that keeps the synthesis reusable. Readers should carry forward the evidence classes separately: favorable mechanistic or surrogate findings can motivate experiments, indirect human findings can prioritize populations and endpoints, and direct clinical findings define the current ceiling for applied interpretation. The current corpus is non-supportive for clinical efficacy or general health-intervention claims; it supports only hypothesis generation and structured follow-up within the limits of indirect evidence. Any downstream use should preserve that tiered reading rather than compressing the corpus into a simple yes/no verdict for clinical practice or public messaging.

Additional corpus sources informed the synthesis without anchoring a foregrounded quantitative claim and are catalogued for completeness: Oldenburg 2021, Luepker 2022, Yosefzadeh 2024, Clarke 2022, Orchard 2021, Okunrintemi 2021.

References

• Lin 2020. Relationship between aspirin use of esophageal, gastric and colorectal cancer patient survival: a meta-analysis. BMC Cancer, 2020. DOI: 10.1186/s12885-020-07117-4 PMID: 32646396.
• Sancar 2022. An Evaluation of Aspirin Treatment Preferences of Physicians in Hypertensive Patients in Terms of Current Guidelines: A Subgroup Analysis of the ASSOS Trial in Turkey. Anatolian Journal of Cardiology, 2022. DOI: 10.5152/AnatolJCardiol.2021.541 PMID: 35435837.
• Fan 2021. Can Aspirin Use Be Associated With the Risk or Prognosis of Bladder Cancer? A Case-Control Study and Meta-analytic Assessment. Frontiers in Oncology, 2021. DOI: 10.3389/fonc.2021.633462 PMID: 34350107.
• Chow 2022. Association of Early Aspirin Use With In-Hospital Mortality in Patients With Moderate COVID-19. JAMA Network Open, 2022. DOI: 10.1001/jamanetworkopen.2022.3890 PMID: 35323950.
• Baker 2023. Aspirin Use and Survival Among Patients With Breast Cancer: A Systematic Review and Meta-Analysis. The Oncologist, 2023. DOI: 10.1093/oncolo/oyad186 PMID: 37358878.
• Peng 2025. Regular aspirin use, breast tumor characteristics and long-term breast cancer survival. NPJ Breast Cancer, 2025. DOI: 10.1038/s41523-025-00775-2 PMID: 40595613.
• Xiao 2021. Timing of Aspirin Use Among Patients With Colorectal Cancer in Relation to Mortality: A Systematic Review and Meta-Analysis. JNCI Cancer Spectrum, 2021. DOI: 10.1093/jncics/pkab067 PMID: 34514327.
• Wang 2021a. Aspirin Use and Common Cancer Risk: A Meta-Analysis of Cohort Studies and Randomized Controlled Trials. Frontiers in Oncology, 2021. DOI: 10.3389/fonc.2021.690219 PMID: 34277434.
• Boakye 2021. Aspirin for cardiovascular disease prevention among adults in the United States: Trends, prevalence, and participant characteristics associated with use. American Journal of Preventive Cardiology, 2021. DOI: 10.1016/j.ajpc.2021.100256 PMID: 34632437.
• Huang 2024. US population qualifying for aspirin use for primary prevention of cardiovascular disease. American Journal of Preventive Cardiology, 2024. DOI: 10.1016/j.ajpc.2024.100669 PMID: 38681065.
• Sun 2019. Aspirin use and pancreatic cancer risk. Medicine, 2019. DOI: 10.1097/MD.0000000000018033 PMID: 31860953.
• Lin 2021. Impact of aspirin use on clinical outcomes in patients with vasospastic angina: a systematic review and meta-analysis. BMJ Open, 2021. DOI: 10.1136/bmjopen-2021-048719 PMID: 34326051.
• Memel 2020. Aspirin Use Is Associated with a Reduced Incidence of Hepatocellular Carcinoma: A Systematic Review and Meta‐analysis. Hepatology Communications, 2020. DOI: 10.1002/hep4.1640 PMID: 33437907.
• Man 2021. Aspirin Use and Mortality in Women With Ovarian Cancer: A Meta-Analysis. Frontiers in Oncology, 2021. DOI: 10.3389/fonc.2020.575831 PMID: 33598421.
• Harewood 2021. Medication use and risk of proximal colon cancer: a systematic review of prospective studies with narrative synthesis and meta-analysis. Cancer Causes & Control, 2021. DOI: 10.1007/s10552-021-01472-8 PMID: 34224060.
• Desai 2020. Association between aspirin use and cardiovascular outcomes in ALLHAT participants with and without chronic kidney disease: A post hoc analysis. The Journal of Clinical Hypertension, 2020. DOI: 10.1111/jch.14091 PMID: 33340443.
• Liu 2021. Effect of aspirin use on survival benefits of breast cancer patients. Medicine, 2021. DOI: 10.1097/MD.0000000000026870 PMID: 34414938.
• Shahrivar 2022. Low‐dose aspirin use and colorectal cancer survival in 32,195 patients—A national cohort study. Cancer Medicine, 2022. DOI: 10.1002/cam4.4859 PMID: 35717628.
• Li 2021a. Aspirin Use on Incident Dementia and Mild Cognitive Decline: A Systematic Review and Meta-Analysis. Frontiers in Aging Neuroscience, 2021. DOI: 10.3389/fnagi.2020.578071 PMID: 33613260.
• Zhang 2023. Efficacy and safety of aspirin antiplatelet therapy within 48 h of symptom onset in patients with acute stroke. World Journal of Clinical Cases, 2023. DOI: 10.12998/wjcc.v11.i32.7814 PMID: 38073696.
• Seidu 2019. Aspirin has potential benefits for primary prevention of cardiovascular outcomes in diabetes: updated literature-based and individual participant data meta-analyses of randomized controlled trials. Cardiovascular Diabetology, 2019. DOI: 10.1186/s12933-019-0875-4 PMID: 31159806.
• Razavi 2024. Aspirin use for primary prevention among US adults with and without elevated Lipoprotein(a). American Journal of Preventive Cardiology, 2024. DOI: 10.1016/j.ajpc.2024.100674 PMID: 38741703.
• Li 2021b. Aspirin Use and the Incidence of Hepatocellular Carcinoma in Patients With Hepatitis B Virus or Hepatitis C Virus Infection: A Meta-Analysis of Cohort Studies. Frontiers in Medicine, 2021. DOI: 10.3389/fmed.2020.569759 PMID: 33490093.
• Lembo 2025. Daily low dose aspirin halves incident type 2 diabetes in elderly subjects with prediabetes: a five-year longitudinal cohort study in a real-word population. Cardiovascular Diabetology, 2025. DOI: 10.1186/s12933-025-02802-9 PMID: 40533771.
• Luo 2025. Aspirin use is associated with attenuated risk of severe acute kidney injury in septic patients: a dual-center retrospective analysis from MIMIC-IV and eICU cohorts. Renal Failure, 2025. DOI: 10.1080/0886022X.2025.2568650 PMID: 41077850.
• Huff 2025. Analysis of prenatal medication use and placental epigenetic gestational age in extremely low gestational age newborns (ELGANs) highlight relationships to aspirin use during pregnancy. Clinical Epigenetics, 2025. DOI: 10.1186/s13148-025-01988-9 PMID: 41291914.
• Cao 2020. Aspirin might reduce the incidence of breast cancer. Medicine, 2020. DOI: 10.1097/MD.0000000000021917 PMID: 32957311.
• Yan 2022. Association Between Aspirin Usage and Age-Related Macular Degeneration: An Updated Systematic Review and Meta-analysis. Frontiers in Pharmacology, 2022. DOI: 10.3389/fphar.2022.824745 PMID: 35401184.
• Srinivasan 2022. Aspirin use is associated with decreased inpatient mortality in patients with COVID-19: A meta-analysis. American Heart Hournal Plus: Cardiology Research and Practice, 2022. DOI: 10.1016/j.ahjo.2022.100191 PMID: 35971534.
• Su 2022. Associations between the use of aspirin or other antiplatelet drugs and all-cause mortality among patients with COVID-19: A meta-analysis. Frontiers in Pharmacology, 2022. DOI: 10.3389/fphar.2022.989903 PMID: 36278186.
• Choi 2025. Effect of aspirin use on conversion risk from mild cognitive impairment to Alzheimer’s disease. Frontiers in Aging Neuroscience, 2025. DOI: 10.3389/fnagi.2025.1603892 PMID: 40842647.
• Huang 2025. Association of aspirin use alone with mortality and liver-related events in MASLD: a multi-institutional three-year study. Annals of Medicine, 2025. DOI: 10.1080/07853890.2025.2573146 PMID: 41103259.
• Celik 2021. Inappropriate Use of Aspirin in Real-Life Cardiology Practice: Results from the Appropriateness of Aspirin Use in Medical Outpatients: A Multicenter, Observational Study (ASSOS) Study. Balkan Medical Journal, 2021. DOI: 10.5152/balkanmedj.2021.21143 PMID: 34142960.
• Alabsi 2023. Regular aspirin use among a sample of American Indians/Alaskan Natives in the Upper Midwest region of the United States. Preventive Medicine Reports, 2023. DOI: 10.1016/j.pmedr.2023.102571 PMID: 38222307.
• Xu 2026. Effect of pre-ICU aspirin use on neuroinflammation and outcomes in patients with sepsis-associated encephalopathy. Frontiers in Neurology, 2026. DOI: 10.3389/fneur.2026.1708039 PMID: 41704889.
• Oldenburg 2021. Promoting Aspirin Use for Cardiovascular Disease Prevention Among an Adult Internet-Using Population: A Pilot Study. Frontiers in Public Health, 2021. DOI: 10.3389/fpubh.2021.500296 PMID: 33796492.
• Ma 2021a. Benefits and Risks Associated With Aspirin Use in Patients With Diabetes for the Primary Prevention of Cardiovascular Events and Mortality: A Meta-Analysis. Frontiers in Endocrinology, 2021. DOI: 10.3389/fendo.2021.741374 PMID: 34539583.
• Liang 2020. Association Between Prior Aspirin Use and Acute Respiratory Distress Syndrome Incidence in At-Risk Patients: A Systematic Review and Meta-Analysis. Frontiers in Pharmacology, 2020. DOI: 10.3389/fphar.2020.00738 PMID: 32508656.
• Luepker 2022. Association of a Community Population and Clinic Education Intervention Program With Guideline-Based Aspirin Use for Primary Prevention of Cardiovascular Disease. JAMA Network Open, 2022. DOI: 10.1001/jamanetworkopen.2022.11107 PMID: 35536579.
• Wu 2024. Continuous aspirin treatment improves cardiovascular events and all-cause mortality in hemodialysis patients with peripheral artery disease. Renal Failure, 2024. DOI: 10.1080/0886022X.2024.2380754 PMID: 39039846.
• Yosefzadeh 2024. Impact of prior aspirin use on left ventricular function in ST-elevation myocardial infarction patients undergoing primary percutaneous coronary intervention: An echocardiographic evaluation. Journal of Cardiovascular and Thoracic Research, 2024. DOI: 10.34172/jcvtr.33184 PMID: 39430282.
• Clarke 2022. Does prior use of antiplatelet therapy modify the effect of dual antiplatelet therapy in transient ischaemic attack/minor ischaemic stroke: A systematic review and meta‐analysis. European Journal of Neurology, 2022. DOI: 10.1111/ene.15433 PMID: 35652757.
• Li 2024. The associations of candidate gene polymorphisms with aspirin resistance in patients with ischemic disease: a meta-analysis. Human Genomics, 2024. DOI: 10.1186/s40246-024-00699-1 PMID: 39617913.
• Celik 2018. Design and rationale for the ASSOS study: Appropriateness of aspirin use in medical outpatients a multicenter and observational study. Anatolian Journal of Cardiology, 2018. DOI: 10.14744/AnatolJCardiol.2018.47587 PMID: 30504736.
• Wong 2022. Safety of Continuing Aspirin Use in Cervical Laminoplasty: A Propensity Score-Matched Analysis. Spine Surgery and Related Research, 2022. DOI: 10.22603/ssrr.2022-0163 PMID: 37041877.
• Aoun 2017. Reduction of intracerebral hemorrhage in hemodialysis patients after reducing aspirin use: A quality-assurance observational study. PLoS ONE, 2017. DOI: 10.1371/journal.pone.0185847 PMID: 28968454.
• Gong 2025. Aspirin improves short and long term survival outcomes of patients with sepsis associated encephalopathy. Scientific Reports, 2025. DOI: 10.1038/s41598-025-08075-2 PMID: 40595183.
• Orchard 2021. Associations between Metformin and Aspirin Use on Cancer Incidence and Mortality in Older Adults. Innovation in Aging, 2021. DOI: 10.1093/geroni/igab046.2339
• Ma 2021b. Aspirin Use and Risk of Breast Cancer: A Meta-analysis of Observational Studies from 1989 to 2019. Clin Breast Cancer, 2021. DOI: 10.1016/j.clbc.2021.02.005 PMID: 33741292.
• Li 2021c. Influence of aspirin use on clinical outcomes of patients with hepatocellular carcinoma: a meta-analysis. Clin Res Hepatol Gastroenterol, 2021. DOI: 10.1016/j.clinre.2020.09.006 PMID: 33067170.
• Chow 2021. Aspirin Use Is Associated With Decreased Mechanical Ventilation, Intensive Care Unit Admission, and In-Hospital Mortality in Hospitalized Patients With Coronavirus Disease 2019. Anesth Analg, 2021. DOI: 10.1213/ane.0000000000005292 PMID: 33093359.
• Abul 2022. Association of mortality and aspirin use for COVID-19 residents at VA Community Living Center Nursing Homes. medRxiv preprint, 2022. DOI: 10.1101/2022.08.03.22278392
• Okunrintemi 2021. Shared decision making and patient reported outcomes among adults with atherosclerotic cardiovascular disease, medical expenditure panel survey 2006–2015. American Journal of Preventive Cardiology, 2021. DOI: 10.1016/j.ajpc.2021.100281 PMID: 34877558.
• Wang 2021b. Low-dose aspirin use and cancer-specific mortality: a meta-analysis of cohort studies. J Public Health (Oxf), 2021. DOI: 10.1093/pubmed/fdz114 PMID: 31781767.
• Gewurz 2024. Inflammation, Frailty, and Aspirin Use in the Physicians' Health Study: A Pilot Study. J Frailty Aging, 2024. DOI: 10.14283/jfa.2024.37 PMID: 39574285.

Background References

Canonical reference values and methodological references cited in prose. Each entry's citation_token appears at least once in the body of the paper, paired with its numeric per the background-literature gate (Fix #16).

• Studenski 2011. Studenski S, Perera S, Patel K, et al. Gait speed and survival in older adults. JAMA. 2011;305(1):50-58. DOI: 10.1001/jama.2010.1923 PMID: 21205966.
• Cruz-Jentoft 2019. Cruz-Jentoft AJ, Bahat G, Bauer J, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing. 2019;48(1):16-31. DOI: 10.1093/ageing/afy169 PMID: 30312372.
• Ioannidis 2005. Ioannidis JPA. Why most published research findings are false. PLoS Med. 2005;2(8):e124. (methodological reference) DOI: 10.1371/journal.pmed.0020124 PMID: 16060722.