Research
What does a longevity study really prove?
Research discovery should show what was registered, what was done, what was reported, and what can reasonably be concluded—without treating those as the same thing.
A connected record
From the original paper to the claim you see
A paper can describe a study; a registry can describe a trial; neither automatically establishes that a claim is true.
Source or paper
The citable publication, preprint, registry entry, correction, or regulatory document.
Study record
The design, population or model, comparison, endpoints, and limitations of the work itself.
Trial record
The registered protocol, phase, operational status, sponsor role, arms, and milestones.
Reported result
A dated finding tied to an endpoint, population, analysis, and publication status.
Scientific claim
A specific proposition assessed against supporting, contradictory, indirect, or insufficient evidence.
Article or update
The reader-facing explanation of what changed, how certain it is, and what it does not show.
Browse the model
Start with the study behind the headline
Start with published coverage or use the demonstrations to understand the records behind it.
Separate screening invitations from completed procedures.
In NordICC, inviting adults to one colonoscopy reduced colorectal-cancer incidence at 13 years. Colorectal-cancer mortality was not significantly lower, and participation limits what the invitation result says about completing screening.
Read the screening analysisKeep selected eligibility beside the mortality benefit.
NLST found fewer lung-cancer deaths with three annual low-dose CT screens than with chest radiography in selected current and former heavy smokers. The benefit came with many positive results that were false positives under the trial definition.
Read the screening analysisRead mortality benefit beside false positives and extra diagnoses.
After 21 years in Rotterdam’s ERSPC section, repeated PSA-based screening reduced prostate-cancer mortality and metastatic disease in the primary age group, while producing substantially more prostate-cancer diagnoses.
Read the screening analysisKeep hormone regimens, symptom treatment, disease outcomes and mortality separate.
Read WHI through formulation, uterus status, absolute outcomes and the limits of general longevity prevention.
Read the WHI hormone analysisRead fracture outcomes before bone-density surrogates.
Follow HORIZON’s vertebral, hip, duration and safety evidence in its defined osteoporosis population.
Read the HORIZON fracture analysisKeep LIFTMOR attached to screening and supervision.
Separate BMD and function findings from fracture prevention, unsupervised safety and longevity.
Read the LIFTMOR explainerSeparate FINGER’s small package-level test result from dementia prevention.
Keep the combined intervention, population, cognitive endpoint, adverse events and untested clinical outcomes visible.
Read the FINGER analysisRead the MIND randomized comparison against its active control.
Keep shared calorie restriction and weight-loss support visible beside the null cognitive and MRI results.
Read the MIND trial analysisKeep ACTIVE’s trained skills, mortality and disputed dementia subgroup separate.
Three evidence lanes prevent a protocol-specific result from becoming a generic brain-training promise.
Read the ACTIVE explainerKeep vitamin D prevention separate from deficiency treatment.
Follow VITAL’s cancer, cardiovascular, and fracture outcomes without transferring them to medically indicated care.
Read the vitamin D claim checkRead cognitive tests and aging clocks on their own scales.
Compare related COSMOS substudies without turning small test or biomarker changes into dementia prevention or longer life.
Read the multivitamin analysisTranslate resveratrol from mechanism to mixed human trials.
Keep memory, metabolic, exercise-adaptation, safety, healthspan, and lifespan claims separate.
Read the resveratrol explainerRead SPRINT through outcomes and tradeoffs.
Separate cardiovascular events, mortality, cognition, adverse events, and population limits without prescribing a personal blood-pressure target.
Read the SPRINT analysisSeparate statin prevention questions in older adults.
Keep secondary prevention, late-life primary prevention, PROSPER’s mixed population, cognition, and ongoing trials distinct.
Read the statin explainerDo not treat every omega-3 product as one intervention.
Compare VITAL, OMEMI, and REDUCE-IT without transferring results across formulations, populations, comparators, or outcomes.
Read the omega-3 claim checkLead with ACHIEVE’s null overall result.
Separate randomized total-cohort cognition from recruitment-source and modeled-risk subgroup signals.
Read the hearing analysisCompare eating windows without erasing the comparator.
Follow three randomized trials and both TREAT corrections without turning short outcomes into slower aging.
Read the meal-timing explainerKeep creatine-plus-training distinct from longevity.
Compare pooled strength and lean-tissue signals with null one-year findings and unavailable disclosure metadata.
Read the creatine briefFollow a functional outcome without turning it into lifespan.
Trace LIFE and a very-old inpatient trial with their populations, adverse events and durability limits visible.
Read the mobility explainerKeep cardiovascular evidence separate from anti-aging claims.
Connect SELECT event counts, sponsor, safety analysis and current label to the exact approved population.
Read the GLP-1 analysisSeparate NAD target engagement from meaningful outcomes.
Compare NR, NMN, registry and FDA records without transferring evidence across compounds or routes.
Read the NAD claim checkSeparate sleep signals from a causal longevity intervention.
Compare duration, regularity and fragmentation across reviews and cohorts while keeping reverse causation visible.
Read the sleep explainerRead PREDIMED with its correction history attached.
See absolute cardiovascular event counts, allocation deviations, revised analyses and the limit of lifespan inference.
Read the PREDIMED analysisKeep diabetes evidence separate from geroscience claims.
Trace MASTERS, a small tissue study, an active registry and the current metformin label.
Read the metformin analysisSeparate calorie-restriction outcomes from lifespan inference.
Follow randomized metabolic results, cardiometabolic outcomes, mixed clock findings, and the linked correction.
Read the CALERIE briefSee how far human senolytic evidence has actually progressed.
Compare three tiny disease-specific pilots without converting feasibility or biomarkers into efficacy.
Read the senolytics analysisAsk what an epigenetic clock was trained to predict.
Separate repeatability, association, intervention response, surrogate validation and clinical utility.
Read the clock explainerTrace an early human safety study back to its protocol and mouse context.
Separate recruitment and registered safety outcomes from efficacy, lifespan, and rejuvenation claims.
Read the reprogramming analysisRead a randomized trial through its primary outcome and limitations.
Keep a null visceral-fat outcome, exploratory signals, safety, and conflicts distinct.
Read the PEARL briefCompare mature functional evidence with indirect longevity claims.
Follow randomized outcomes, systematic review evidence, population guidance, and remaining uncertainty.
Read the exercise explainerLead with SmartAge’s null prespecified memory result.
Keep exploratory signals, product-specific dosing, conflicts and untested dementia or longevity outcomes separate.
Read the spermidine explainerRead ENERGIZE through both missed primary outcomes.
Separate nonsignificant walking and ATP results from secondary endurance and biomarker signals in a selected short trial.
Read the urolithin A analysisKeep GlyNAC’s many exploratory measures attached to its tiny sample.
Distinguish within-arm change from randomized comparison and biomarkers from demonstrated reversal of aging.
Read the GlyNAC explainerBrowse 133 primary studies and governing syntheses
Filter real publication records by design, population or model, and measured outcome while keeping source identity separate from editorial interpretation. The fixture-only library demonstration remains available for record-contract inspection.
Browse covered studiesBrowse 60 registry records connected to released reporting
Filter operational status, results availability and topic while keeping registry facts separate from evidence conclusions. The fixture-only tracker demonstration remains available for record-contract inspection.
Browse covered clinical trialsTrace a claim back to its original source
Keep source identity, role, dates, confidence, editorial status, and validity independent.
Open the fixture provenance recordSee what changed—and why
Compare current state with append-only correction, supersession, and retraction events.
Open the fixture history ledgerWho funded or conducted the work?
Inspect identity, aliases, programs, dated roles, funding links, disclosures, and provenance without implying endorsement.
Open the fixture organization directoryWhich researchers are attached to the evidence?
Keep authorship, dated affiliations, roles, disclosures, coverage, and claim validity independent.
Open the fixture researcher directoryWhy do studies agree, conflict, or leave a claim open?
Keep claim scope, study design, population, outcome, limitations, provenance, and history visible without a composite score.
Open the fixture evidence mapFind the studies behind connected coverage
Distinguish reporting, status changes, timeline events, and editorial curation without an opaque relatedness or popularity score.
Open connected coverageWhat are people talking about—and what is still unproven?
Keep observation, clustering, scientific importance, source verification, and editorial disposition independent.
Open the fixture discovery boundaryLearn to spot strong evidence and weak inference
Use the research guides to interpret study design, clinical trial structure, biomarkers, and uncertainty.
Open Research GuidesConnected exploration