Human diabetes incidence and microvascular outcomes · long randomized follow-up
Population or model2,776 surviving DPP participants with elevated glucose and overweight or obesity who joined long-term follow-up
randomized lifestyle or metformin versus placebo, followed observationally after the original trial
Released bottom line
Over 15 years, cumulative diabetes incidence was 55% with intensive lifestyle, 56% with metformin and 62% with placebo. The randomized groups did not differ significantly in aggregate microvascular prevalence.
What this does not establish
The follow-up did not demonstrate that either randomized assignment reduced the aggregate microvascular outcome or extended life.
Checked
intensive lifestyle intervention for weight loss and physical activity · major cardiovascular events, weight and cardiometabolic risk factors
Human cardiovascular events and function · large randomized trial
Population or model5,145 adults age 45–75 with type 2 diabetes and overweight or obesity
intensive lifestyle intervention versus diabetes support and education
Released bottom line
The primary cardiovascular outcome occurred in 403 intensive-lifestyle participants and 418 controls (hazard ratio 0.95; P=0.51). Weight loss and several risk factors improved more with intensive lifestyle.
What this does not establish
Look AHEAD did not establish that its intensive lifestyle program reduced major cardiovascular events or extended life in this population.
Checked
intensive multi-drug glycemia strategy targeting HbA1c below 6% · major cardiovascular events, all-cause mortality and treatment harms
Human cardiovascular events, mortality and safety · large randomized trial
Population or model10,251 adults with established type 2 diabetes and high cardiovascular risk
intensive glycemia target below 6% versus standard target of 7.0% to 7.9%
Released bottom line
The primary cardiovascular outcome was not significantly lower (352 versus 371 events; hazard ratio 0.90; P=0.16). Deaths were higher with intensive treatment (257 versus 203; hazard ratio 1.22; P=0.04), prompting early discontinuation.
What this does not establish
ACCORD does not show that all glucose lowering is harmful; it tested one intensive, multi-drug target strategy in a high-risk population.
Checked
urolithin A · walking distance, muscle endurance and ATP production
Human function and biomarkers · small randomized trial
Population or modelSelected older adults with lower mitochondrial function
small randomized trial
Released bottom line
ENERGIZE found no significant benefit for six-minute walk distance or maximal ATP production. Secondary endurance and biomarker signals remain preliminary and do not establish disability prevention or longer healthy life.
What this does not establish
Urolithin A missed both primary outcomes in older adults
Checked
low-dose aspirin · disability-free survival · major bleeding
Human function and safety · large randomized trial
Population or model19,114 community-dwelling older adults without cardiovascular disease, dementia or independence-limiting disability at enrollment
large randomized trial
Released bottom line
In 19,114 selected older adults, daily low-dose aspirin did not improve the trial’s composite of death, dementia or persistent physical disability over a median 4.7 years. Major hemorrhage was higher.
What this does not establish
ASPREE found no gain in disability-free survival—and more major bleeding
Checked
testosterone · six-minute walking distance and a 50-meter improvement threshold
Human physical-function outcomes · one-year randomized trials
Population or model790 symptomatic men age 65 or older with repeatedly low testosterone; 390 enrolled in the Physical Function Trial
one-year randomized trials
Released bottom line
The prespecified Physical Function Trial threshold was not significant in its enrolled subgroup; broader secondary walking results do not establish disability prevention, healthspan or longevity.
What this does not establish
Testosterone’s walking result was modest—not an anti-aging effect
Checked
CBT-I · sleep latency, sleep efficiency, insomnia remission and durability
Human BMD and function · small supervised randomized trial
Population or model101 screened postmenopausal women with low bone mass; mean age 65
small supervised randomized trial
Released bottom line
LIFTMOR found BMD and function gains in a small, screened, closely supervised trial—not fracture prevention or proof that unsupervised heavy training is safe.
What this does not establish
LIFTMOR improved bone density in a small supervised trial—not fracture prevention
Checked
creatine · lean tissue, strength and bone or muscle imaging
Human cardiovascular and safety outcomes · large randomized trial
Population or model19,114 community-dwelling older adults without cardiovascular disease at enrollment
large randomized trial
Released bottom line
In selected older adults without cardiovascular disease, low-dose aspirin did not significantly reduce the trial’s cardiovascular composite. Major hemorrhage occurred more often.
What this does not establish
ASPREE did not lower cardiovascular events—and increased major bleeding
Checked
testosterone · major cardiovascular events and cardiovascular safety outcomes
Human cardiovascular outcomes · large randomized noninferiority trial
Population or model5,246 men age 45–80 with symptoms, repeatedly low testosterone and preexisting or high cardiovascular risk
large randomized noninferiority trial
Released bottom line
Among selected men with hypogonadism and elevated cardiovascular risk, testosterone gel was noninferior to placebo for major cardiovascular events; the trial did not test longer life or general anti-aging use.
What this does not establish
TRAVERSE found cardiovascular noninferiority—not a longevity benefit
Checked
testosterone · clinical fractures and bone-density measures
Human fracture and bone-density outcomes · randomized trials
Population or model5,204 TRAVERSE participants: middle-aged and older men with symptoms, repeatedly low testosterone and preexisting or high cardiovascular risk
randomized trials
Released bottom line
Clinical fractures were more frequent with testosterone than placebo in TRAVERSE; a smaller substudy’s bone-density gain cannot substitute for fracture outcomes.
What this does not establish
Testosterone did not prevent fractures in TRAVERSE
Checked
CPAP · major cardiovascular events, sleepiness, symptoms and quality of life
Population or model7,765 postmenopausal women with osteoporosis; mean age about 73
large randomized trial
Released bottom line
HORIZON reduced vertebral and hip fractures in postmenopausal osteoporosis, with outcome-specific duration and safety limits—not evidence of slower aging.
What this does not establish
Zoledronic acid reduced fractures in osteoporosis—not aging itself
Checked
vitamin D · invasive cancer, major cardiovascular events and clinical fractures
Human clinical outcomes · large randomized prevention trial
Population or model25,871 generally healthy U.S. adults not selected for vitamin D deficiency, low bone mass or osteoporosis
large randomized prevention trial
Released bottom line
In generally healthy adults, vitamin D3 did not significantly reduce invasive cancer, major cardiovascular events, or fractures; deficiency treatment is a different question.
What this does not establish
Vitamin D is essential. VITAL still found no broad prevention benefit
Population or model9,361 adults age 50 or older at elevated cardiovascular risk, without diabetes or prior stroke
large randomized trial
Released bottom line
Intensive treatment lowered cardiovascular events and mortality in eligible high-risk adults, with more hypotension, electrolyte problems and kidney injury.
Human clinical outcomes · formulation-specific randomized trials
Population or modelPrimary-prevention adults, older adults after myocardial infarction and statin-treated high-risk adults with elevated triglycerides
formulation-specific randomized trials
Released bottom line
Two mixed EPA+DHA trials were neutral on primary outcomes; prescription icosapent ethyl benefited a defined high-risk population with safety and comparator limits.
What this does not establish
“Omega-3” is not one intervention—and the trial results do not transfer
Checked
meal timing · weight, body composition, visceral fat and cardiometabolic markers
Population or modelOlder adults with subjective cognitive decline
small randomized phase 2b trial
Released bottom line
In the 12-month SmartAge trial, a wheat-germ spermidine extract did not improve the prespecified memory outcome or secondary outcomes. Exploratory signals do not establish dementia prevention or longer life.
What this does not establish
Spermidine did not improve the trial’s primary memory outcome
Checked
shingles vaccine · herpes-zoster disease and recorded dementia diagnoses
Human disease outcomes · randomized and non-randomized evidence
Population or modelAdults age 50 or older in ZOE-50 plus older vaccine-eligible cohorts in two dementia analyses
randomized and non-randomized evidence
Released bottom line
Randomized trials established shingles prevention. Dementia findings come from observational and quasi-experimental studies and do not prove prevention.
Human cognitive tests · randomized multidomain trial
Population or model1,260 Finnish adults age 60–77 at elevated dementia risk and with average-to-lower cognitive performance
randomized multidomain trial
Released bottom line
A combined two-year program produced a small cognitive-test advantage; it cannot identify a winning component or establish dementia, disability, mortality, healthspan, or lifespan benefit.
What this does not establish
FINGER slightly improved cognitive-test scores. It did not prove dementia prevention
Checked
MIND diet · global cognition and structural MRI measures
Human cognition and MRI · randomized dietary trial
Population or model604 adults age 65 or older without cognitive impairment, with family dementia history, overweight and a suboptimal diet
randomized dietary trial
Released bottom line
Both groups improved with counseling and calorie restriction; the between-group cognitive difference was not significant and MRI measures were similar.
What this does not establish
The MIND diet trial found no cognitive edge over its active control
Checked
cognitive training · trained abilities, daily function, all-cause mortality and claims-diagnosed dementia
Human trained abilities and long-term outcomes · randomized trial plus disputed secondary analysis
Population or model2,832 independent U.S. adults age 65–94
randomized trial plus disputed secondary analysis
Released bottom line
ACTIVE supports trained-ability gains and some function signals, found no mortality benefit, and leaves a claims-based dementia subgroup methodologically disputed.
What this does not establish
Cognitive training improved practiced skills—not survival or proven dementia prevention
Checked
multivitamins · cognitive-test scores, word recall and DNA-methylation biomarkers
Human cognitive tests and biomarkers · correlated randomized substudies
Population or modelOlder adults in related telephone, web, clinic and blood-based COSMOS substudies
correlated randomized substudies
Released bottom line
COSMOS found small cognitive-test differences and modest movement in two DNA-methylation clocks; neither establishes dementia prevention or longer life.
What this does not establish
Multivitamins moved some tests and aging clocks. What does that mean?
Checked
resveratrol · memory tests, metabolic markers, body composition, inflammation and exercise adaptation
Human biomarkers and function · very small randomized trial
Population or modelVery small sample of older adults
very small randomized trial
Released bottom line
A 16-week study randomized 24 older adults—12 per arm—and reported many biomarker and function measures. Its size, multiplicity and exploratory analyses do not prove that GlyNAC reverses aging or extends healthspan.
What this does not establish
GlyNAC’s small trial cannot establish reversal of aging
Checked
habitual winter swimming and cold challenge · brown-fat activity and thermogenesis
Human physiology biomarkers · small cross-sectional study
Population or modelVery small selected sample of young healthy men
cross-sectional swimmer versus matched comparison group
Released bottom line
A small study of young male winter swimmers found physiology differences. It did not test disease, function, healthspan or lifespan, and it cannot show that cold exposure caused the findings.
What this does not establish
Cold-water studies measure thermogenesis—not longer life
Checked
supervised hyperbaric-oxygen research program · selected blood-cell telomere and senescence markers
Human blood-cell biomarkers · uncontrolled pre/post study
Population or modelSmall selected group of healthy adults age 64 or older
single-arm pre/post study without a control
Released bottom line
A small uncontrolled study reported telomere and senescent-cell changes in selected blood-cell subtypes. It did not test clinical aging, healthspan or lifespan, and commercial conflicts matter.
Population or model1.48 million adults age 20–79 in four national cohorts, with a separate 202,248-person US linked-cohort analysis
observational current and former cigarette smokers compared with never smokers; cessation timing compared with continued smoking
Released bottom line
Current cigarette smoking was associated with roughly three times the mortality of never smoking; stopping was associated with lower excess mortality at every age studied.
What this does not establish
The analyses did not establish an exact causal effect for every subgroup, forecast an individual lifespan, or compare cessation treatments.
Checked
invitation to one colonoscopy · colorectal-cancer incidence and mortality
Human cancer incidence and mortality · large randomized invitation trial
Population or modelAdults age 55–64 in Norway, Poland and Sweden
randomized invitation versus usual care
Released bottom line
At 13 years, colorectal cancer was diagnosed in 1.46% of the invited group and 1.80% of the usual-care group. Colorectal-cancer mortality was 0.41% versus 0.47%; the confidence interval included no difference.
What this does not establish
A colorectal-cancer mortality or all-cause survival benefit was not demonstrated.
Checked
three annual low-dose CT screens · lung-cancer and all-cause mortality
Human lung-cancer mortality · large randomized screening trial
Population or modelSelected current and former heavy smokers age 55–74
randomized low-dose CT versus chest radiography
Released bottom line
The original NLST analysis reported 247 versus 309 lung-cancer deaths per 100,000 person-years, a 20% relative reduction with low-dose CT. It did not test never-smokers or the general population.
What this does not establish
Benefit for never-smokers, average-risk adults or overall lifespan.
Checked
repeated PSA-based screening · prostate-cancer mortality, metastasis and diagnoses
Human prostate-cancer mortality and diagnosis · long randomized screening follow-up
Population or modelMen age 55–74 in Rotterdam; primary group age 55–69
randomized screening invitation versus usual care
Released bottom line
In the primary age group, prostate-cancer mortality was lower with screening (RR 0.73), as was metastatic disease (RR 0.67). Screening also produced 57 excess prostate-cancer diagnoses per 1,000 men randomized.
What this does not establish
An all-cause survival benefit or favorable balance for every individual.
Checked
habitual sauna use · cardiovascular and all-cause mortality association
One Finnish male cohort linked frequent sauna use with lower mortality. The design cannot show that sauna caused longer life, and randomized evidence has not tested that endpoint.
What this does not establish
Sauna use was associated with lower mortality—not proven to extend life
Checked
low-dose aspirin · cancer mortality · long-term follow-up
Human mortality and cancer outcomes · randomized trial plus observational extension
Population or model19,114 original ASPREE participants; 14,907 cancer-free survivors in the post-trial analysis
randomized trial plus observational extension
Released bottom line
The randomized trial found an unexpected cancer-mortality imbalance. Longer follow-up found no difference in overall cancer incidence and no post-trial mortality legacy effect.
What this does not establish
ASPREE’s cancer signal changed across longer follow-up
Checked
Method
One released story, one primary group.
All 55 canonical stories appear exactly once. A story may expose several measured-result filters, but it is never duplicated. Open the full article for sources, endpoints, limitations, funding, conflicts, safety context and corrections.