MAGELLAN LONGEVITY
Home › Evidence record

The evidence record: where we disagree with the market

Every product on this site carries a grade for the strength of the human evidence behind it, and those grades are assigned independently of affiliate commissions. This section is what that policy costs us. It is the running, public list of longevity claims our own reading of the evidence does not support — including claims attached to products in our own catalogue.

The format is the same on every page: what is genuinely true, what has been oversold, the trial record including the studies that failed or were inconclusive, and the specific result that would change our mind. Where a product is still legitimately worth considering for a narrower claim, we say so and link to it. Where it is not, there is no buy button — only a link to the graded product page and to something with better evidence behind it.

Of the 12 compounds and devices documented here, we have withdrawn the buy recommendation on 3 and narrowed the claim on the rest.

The record

Compound or deviceOur gradeThe claim we do not supportDo we still recommend buying it?
NMN (nicotinamide mononucleotide)
Partly true — for a much narrower claim than the one being sold
Moderate evidenceNAD+ falls with age, NMN raises NAD+, therefore NMN reverses aging.Yes — for the narrower claim only
Resveratrol
The mechanism story outran the outcome data by about fifteen years
Moderate evidenceResveratrol activates sirtuins, mimics caloric restriction, and is the reason red wine is good for you.Yes — for the narrower claim only
Vitamin D
One of the best-tested supplements in medicine, and most of the big tests were negative
Moderate evidenceEveryone is deficient, and vitamin D prevents cancer, heart disease, fractures, falls and early death.Yes — for the narrower claim only
Omega-3 fish oil
The lipid effect is solid. The heart-attack claim is contested, and the AF signal is real
Moderate evidenceFish oil protects your heart, so everyone should take it — and more is better.Yes — for the narrower claim only
Blue-light filtering glasses
Tested properly. They did not work. We grade this at the bottom
Tested — did not workBlue-light filtering glasses reduce digital eye strain, protect your retina, and improve your sleep.No — see the alternative on the page
Fisetin
A strong preclinical case that has not survived its most rigorous animal test
Emerging evidenceFisetin clears senescent zombie cells and extends lifespan — take it in monthly high-dose pulses.Yes — for the narrower claim only
Spermidine
A superb animal story, an ambiguous human one, and a bioavailability problem
Moderate evidenceSpermidine induces autophagy, extends lifespan across species, and dietary intake is linked to five fewer years of mortality risk.Yes — for the narrower claim only
Magnesium L-threonate
A rodent mechanism, two small human trials, and no independent replication
Emerging evidenceThe only magnesium that crosses the blood-brain barrier — raises brain magnesium, builds synapses and improves memory.Yes — for the narrower claim only
Apigenin
Two real findings welded into one claim that neither of them supports
Emerging evidenceApigenin inhibits CD38, so it preserves NAD+ and slows aging — and it is the calming molecule in chamomile, so it helps you sleep.Yes — for the narrower claim only
Glucosamine for longevity
The most-cited longevity association in the supplement aisle, and the most contested
Moderate evidenceGlucosamine users have 27% lower all-cause mortality — it is a longevity supplement disguised as a joint supplement.Yes — for the narrower claim only
Low-dose lithium orotate
A landmark mouse paper, an interesting epidemiology, and no human trial at the dose being sold
Emerging evidenceNature showed lithium deficiency causes Alzheimer's and lithium orotate reverses it — so take 5 mg a day for your brain.No — see the alternative on the page
Saliva cortisol stress testing
A legitimate assay attached to a claim it cannot support — our lowest grade in diagnostics
PreliminaryA four-point daily saliva cortisol curve measures your stress, reveals adrenal fatigue, and tells you what to fix.No — see the alternative on the page

A short note on how we grade

Ingestibles and interventions are graded on the strength of human evidence for the active compound. Diagnostics and wearables are graded on how well the thing they measure predicts health — not on a treatment effect. For well-characterised compounds a board-certified physician assigns the grade from the human and mechanistic literature, and a dose is published only where a verified standard range exists. Effect statements on product pages are quoted verbatim from a cited study; we never paraphrase findings into stronger claims than the authors made.

Strong evidence

Consistent randomized human trials — or large, convergent human cohort data — show a real benefit, usually for a specific measurable outcome rather than for lifespan.

Moderate evidence

Several human studies point the same way, but with limits — size, duration, funding, or mixed results.

Emerging evidence

Mechanism is plausible and early human data exists, but it is small, short, or not yet replicated.

Preliminary

Mostly mechanistic or animal data, or very few human studies. Interesting, not proven.

Measurement tool

A test or sensor, graded on what the marker predicts and how reliably it measures it.

Tested — did not work

Adequate evidence pointing at nothing. Reserved for things that were properly tested and failed, which is different from things that are merely unproven.

From our published method: where human trials are inconsistent or have failed to show benefit, we say so on the product page. A marketplace that only ever surfaces flattering studies is marketing in a lab coat. Open the interactive grading rubric in the store →

Inclusion is never paid. No brand can buy a listing, a grade, or placement, and Magellan does not sell its own supplement line. Magellan earns an affiliate commission through the Amazon Associates program at no extra cost to you; that commission does not influence the grade.

Educational information, not medical advice. Nothing here is intended to diagnose, treat, cure, or prevent any disease. Evidence describes a compound or a category of device — it is not a claim about any specific commercial product. Talk to your physician before starting or stopping any supplement, test, or therapy, especially if you are pregnant, nursing, or taking medication.

Where to go next

The evidence-graded supplement guide · Longevity research map · Pathway universe · Senolytics · NMN buyer's guide · Creatine — the strongest-graded supplement here · Omega-3 · Resveratrol · Spermidine · Urolithin A · Sulforaphane · Berberine

About this section

Why would a storefront publish pages arguing against products it sells?

Because the grade is the product. Magellan grades every listing on the strength of human evidence, and those grades are assigned independently of affiliate commissions. If a grade can never be bad news for revenue, it is not a grade. These pages are what the firewall looks like when it is working: 3 of the 12 compounds documented here carry no buy button from us at all.

Does a page here mean the compound is useless?

No. Most of these compounds do something real — the problem is the distance between what a trial measured and what the marketing implies. Each page separates the two explicitly: a 'what is true' section built from the same primary sources we cite on the product page, and a 'what is oversold' section built from the trials that failed, were inconclusive, or measured something narrower than the claim.

How is the evidence graded?

Ingestibles are graded on the strength of human evidence for the active compound; diagnostics and wearables are graded on how well the thing they measure predicts health, not on a treatment effect. A board-certified physician assigns grades for well-characterised compounds, and effect statements are quoted verbatim from a cited study rather than paraphrased into stronger claims than the authors made.

How often is this updated?

These pages were reviewed on 2026-08-11. A grade moves when the evidence moves — each page lists, in plain language, the specific result that would change our mind. If one of those results lands, the grade changes and this record says so.