Skip to main content
Evidence Based Longevity (EBL)
Latest changes

Longevity is long on hype and short on human evidence. We separate the marketing and the mouse studies from what’s been tested in people.

Supplements, longevity drugs, devices, diets and lifestyle protocols, graded A to D on published human clinical trials. The letter grades the evidence, not the product. Look any one up, free.

  • 7 questions
  • Free
  • About 2 minutes
  • No account
  • We never see your answers

For you and your family: plain answers first. Where to begin → For clinicians: the full evidence trail. For clinicians →

Sound familiar?

  • Eight bottles on the counter. No idea what’s behind any of them.
  • The study everyone cites used a different dose.
  • It worked in mice. Was it ever tested in people?
  • It changed a lab number. Did anyone live longer or better?

No more guessing.

Quick tour

The whole site in under two minutes. Sound is off until you turn it on.

Or start from your goal: Heart and blood vessels Brain and mood Blood sugar Bones Strength and falls Sleep Hearing and vision Living longer

Full tour, under two minutes, with music. Sound is off; use the player to turn it on.
Read the tour as text
  1. Supplements. Devices. Medicines. Longevity protocols. What’s actually been tested in people? More than 600 pages, more than 500 graded uses, more than 3,000 studies cited.
  2. Sound familiar? Eight bottles on the counter. No idea what’s behind any of them. The study everyone cites used a different dose. It worked in mice. Was it ever tested in people? It changed a lab number. Did anyone live longer or better? No more guessing.
  3. What the letters mean. Every letter is about human trials. A, Replicated The same finding reproduced by a separate research group. A result seen once can be chance or bias; a replicated one is harder to explain away.: replicated trials on a qualifying endpoint The outcome a trial measures to decide whether the treatment worked, such as heart attacks, blood pressure or a symptom score.. B, Supported: a good trial, backed by more. C, Limited: small, mixed or unreplicated trials. D, Not shown: no human trial shows benefit. A D is never one bucket: not yet tested, tested with no benefit, or tested with harm found.
  4. Seven questions. About two minutes. Free. We never see your answers. One report: metformin can deplete B12 and folate, and possibly CoQ10, not proven, and gets no letter; vitamin D3, B, fewer fractures, with calcium, in care homes; vitamin E, D, tested for cancer prevention, harm found.
  5. Same supplement. Two uses. Two letters. Creatine for strength, with weight training, over 50: B. A trial large enough to count found bigger strength gains; the largest, longest trials did not. Creatine for bone density, women after menopause: D. Bone density did not improve. The letter grades the evidence for what was tested.
  6. What moves the needle. A better lab number is not the same as better health. From the most direct evidence of better health to the least: deaths and disease events, such as heart attacks, strokes, fractures, cancers and deaths; what people can do or feel, such as strength, fitness, alertness and sleep; and a test number only, where a blood test or other measurement moved.
  7. More than a supplement checker. Supplements, medications, devices, protocols, conditions.
  8. The letter can move down. That is the point. For example: CoQ10 for chronic heart failure, A to B, September 2026. The A rested on one trial no independent group has repeated. Every change, dated.
  9. Source Check: a claim that circulates, and the study behind it. The claim, as circulated: high-protein diets shorten lifespan. Set beside what the study actually found.
  10. Every answer here starts with a question, not a shopping list. The Medication and Supplement Check: free, about two minutes, no account. evidencebasedlongevity.co

580

graded uses, one letter each

3,421

published studies cited

$0

commission on any supplement EBL grades

108

grade changes logged, each one dated

What you get

What you take, checked against the trials.

One person's list, checked. Medications are checked, not graded, so they get no letter. Open a row to see why.

Medication and Supplement CheckSample report

What one person takes

  • MetforminMedication can lower your vitamin B12 and folate; CoQ10, possible, not provenNo letter

    In plain wordsCan lower your vitamin B12 and folate. Possibly CoQ10, not proven.

    EvidenceMedications are checked, not graded, so there is no letter. Evidence tier on each depletion A fall in a nutrient's level caused by a medication, for example through reduced absorption or increased loss. record: vitamin B12, well documented; folate, well documented; CoQ10, possible, not proven.

    The full metformin record →

  • Vitamin D3Fracture prevention with calcium, older adults in residential careB

    In plain wordsIn care homes, vitamin D taken with calcium meant fewer broken bones. Alone, it did not.

    EvidenceGrade B. Measured on a hard clinical endpoint An outcome that matters directly to a person and is counted without judgment, such as death, heart attack, stroke or fracture.. 4 PubMed-indexed papers on this use. Replicated benefit on hip fracture in older adults living in institutions. Not A: the large community trials did not reproduce it.

    Caution on the recordHypercalcemia Too much calcium in the blood. Mild cases cause few symptoms; higher levels can bring thirst, confusion, kidney stones and heart rhythm problems. from any cause. Sarcoidosis, tuberculosis and other granulomatous disease A disease in which the immune system walls off irritation in small clusters of inflamed cells called granulomas, as in sarcoidosis and tuberculosis. These clusters can activate vitamin D on their own., where extrarenal Outside the kidneys. The kidneys normally control how much vitamin D is turned into its active form; activation elsewhere in the body escapes that control. conversion can produce hypercalcemia at ordinary doses. Primary hyperparathyroidism. Sustained intake above 10,000 IU/day without monitoring.

    The full Vitamin D3 record and its trials →

  • Vitamin ECancer prevention, tested at scaleD

    In plain wordsLarge trials found it did not prevent cancer. In one, it raised prostate cancer in healthy men.

    EvidenceGrade D, Route 2 Human randomized trials have been run and did not show benefit on the stated outcome. One of the two routes to the lowest letter on EBL's scale.: TESTED, HARM FOUND. Measured on a hard clinical endpoint. 8 PubMed-indexed papers on the record. Every large randomized trial of vitamin E for cancer prevention found no benefit. In SELECT, 400 IU a day of synthetic vitamin E increased prostate cancer in healthy men.

    Caution on the recordAnyone already at raised bleeding risk. Separation from surgery is the usual clinical margin.

    The full Vitamin E record and its trials →

Scale A · B · C · DOpen a row to see why

Each grade is for one use, not the bottle.

Your own list

It stays on this device. Nothing you type is sent anywhere.

The check is step 1 of 6. See the full method →

What the letters mean

Every letter is about human trials.

Each grade is for one use, not the bottle.

Grade A: Replicated

Replicated trials on a qualifying endpoint

Trials in people found a real benefit on an outcome that counts, and separate research groups found it again.

Grade B: Supported

A good trial, backed by more

A good trial in people found a benefit, and other studies back it up.

Grade C: Limited

Small, mixed or unreplicated trials

The trials are small, mixed, or not yet repeated. Too soon to be sure.

Grade D: Not shown

No human trial shows benefit

Either no one has tested it in people, or the trials found no benefit or found harm.

A D is never one bucket. We say which:

  • Not yet tested
  • Tested, no benefit
  • Tested, harm found

580 graded uses. Most land at C.

  • A Replicated: 32
  • B Supported: 99
  • C Limited: 280
  • D Not shown: 169

How grades work →

One supplement, two grades

Same supplement. Two uses. Two letters.

B

Creatine for strength

With resistance training, in adults over 50.

A trial large enough to count found bigger strength gains. The largest, longest trials did not.

It measures strength, not falls, fractures or survival.

D

Creatine for bone density

Women after menopause, in two-year trials.

Bone density did not improve.

The letter grades the evidence for what was tested. See the creatine page →

What moves the needle

A better lab number is not the same as better health.

What did the trial actually change? A letter earned on heart attacks or fractures is a different kind of evidence from a letter earned on a blood test.

Deaths and disease events

Heart attacks, strokes, fractures, cancers, deaths. Things that happen to people and are counted.

  • AHard clinical endpoint
    Mediterranean diet + extra-virgin olive oilMajor cardiovascular events (PREDIMED)
    6 fewer per 1,000 people, over a median 4.8 yearsOutcome major cardiovascular events (counted once: the first heart attack, stroke or cardiovascular death)In 7,447 adults aged 55 to 80 at high cardiovascular risk, without cardiovascular disease, Spain (PREDIMED)Compared Mediterranean diet with extra-virgin olive oil against advice to cut dietary fatPMID PubMed ID: the permanent number the US National Library of Medicine gives a published paper. Each one links to that paper's record on PubMed. 29897866
  • AHard clinical endpoint
    SemaglutideCardiovascular death, heart attack and stroke in people with cardiovascular disease or type 2 diabetes
    15 fewer per 1,000 people, over a mean 39.8 monthsOutcome major adverse cardiovascular events (counted once: the first cardiovascular death, nonfatal heart attack or nonfatal stroke)In 17,604 adults aged 45 and over with cardiovascular disease and overweight or obesity, without diabetes (SELECT)Compared semaglutide 2.4 mg weekly against placebo An inactive substitute made to look like the real treatment, so participants cannot tell which one they received.PMID 37952131
  • BHard clinical endpoint
    Coenzyme Q10 (CoQ10)Chronic heart failure events, including hospitalization, as an adjunct A treatment tested on top of standard prescribed therapy rather than in place of it. to standard therapy
    110 fewer per 1,000 people, over 2 yearsOutcome major adverse cardiovascular events, the trial's long-term combined endpoint (counted once: the first such event)In 420 adults with moderate to severe chronic heart failure on standard therapy (Q-SYMBIO)Compared CoQ10 100 mg three times a day against placebo, both added to standard therapyPMID 25282031

73 graded uses measured here.

What people can do or feel

Strength, fitness, alertness, sleep. Measured with a test, or reported by the person.

177 graded uses measured here.

A test number only

A blood test or other measurement moved. Whether people lived longer or better was not measured.

  • BSurrogate biomarker
    Alcohol reductionBlood pressure in people drinking more than two drinks a day
  • BSurrogate biomarker
    Amla (Indian Gooseberry, Emblica officinalis)Lipid profile (LDL Low-density lipoprotein cholesterol, the cholesterol fraction most closely tied to plaque in the arteries., triglycerides)
  • BSurrogate biomarker
    RetatrutideApnea-hypopnea index in adults with obesity and moderate to severe obstructive sleep apnea

161 graded uses rest on a marker alone.

Each figure answers a different question, about a different outcome, group of people, comparison and length of time. The numbers cannot be compared with each other.

See all 580, ranked →

How a grade is made

Behind every letter, in short.

  1. We find the trials

    Only studies in people count toward a letter. Animal and test-tube studies do not.

  2. We check what was measured

    The dose, who was studied, and whether people's health changed or only a lab number did.

  3. We show our work

    Every letter opens its trials, so you can see exactly why it was given.

The full method, step by step →

Why you can trust it

The letter can move down. That is the point.

No grade for sale
No company can pay for a letter. EBL takes no commission on any supplement it grades. Where EBL earns on a device or a test, the page says so beside the letter.

Last grade change: November 2026 Creatine for strength in older adults moves from A to B; the headline moves to B. 108 grade changes logged, each with its reason. See every change →

One letter that moved down

CoQ10 for chronic heart failure: A to B

The A rested on one trial that no independent group has repeated.

September 2026 · Every change, dated

One editor: Arnold Sandlow, D.C., California-licensed, Diplomate of the American Board of Anti-Aging Health Practitioners (A4M), former Founding Full Professor of Anti-Aging Studies, and author of the chapter “Enhancing Adrenal Function” in the textbook Pain Management, 6th edition. One published standard. About the editor →

Source Check · A claim that circulates, and the study behind it

The claim, as circulated

“High-protein diets shorten lifespan”

What we found: source reversed.

See what the study actually found →

Every answer here starts with a question, not a shopping list.

EBL shows you what the evidence says. It does not tell you what to take, and it does not replace your doctor.

Question of the week

Metformin is widely taken for longevity. In two randomized trials, older adults took metformin or a placebo while doing supervised exercise training. What did metformin do to the benefits of the exercise?

  • A)Roughly doubled them
  • B)Made no difference either way
  • C)Reduced them
  • D)It has never been tested alongside exercise
Show the answer

C) Reduced them

In the MASTERS trial, 94 adults aged 65 and older did 14 weeks of supervised strength training; those on placebo gained more lean body mass and thigh muscle than those taking 1,700 mg of metformin a day. In a 12-week aerobic training trial in adults in their early 60s, metformin blunted the gains in VO2max The most oxygen the body can use during all-out exercise, a standard measure of aerobic fitness. and insulin sensitivity and canceled the rise in muscle mitochondrial function that exercise produced on placebo.

On EBL, Metformin (for longevity) is Grade D.

Read the full record and the trials behind the grade

Earlier questions, and one question by email every Monday

Check what I take

Free · About 2 minutes · No account

Build 9b19d78, November 2026