Medically reviewed by Dr Zeeshan Afzal (MBBS, General Practitioner) — Medical Content Reviewer, Welzo.
Written by: The Welzo Longevity Editorial Team | Last updated: July 2026
Declared interest: Welzo sells supplements. This article is written about the category as a whole, not about our range, and deliberately contains no product links. Several compounds criticised below are widely sold, including by retailers like us. Where evidence is genuinely unsettled rather than negative, we say so instead of overstating the case. See our editorial policy.
This article is the counterpart to our longevity supplements guide: it works through the compounds where the evidence is genuinely weak, absent or negative. The market runs on a specific kind of argument — a compound does something interesting in a petri dish or a mouse, the finding is explained persuasively, and the explanation is then sold as though it were a result.
A few compounds here are not merely ineffective but carry real risk at high doses. Others are simply "not known yet" — NMN is the clearest case, and our NMN benefits, side effects and dosage guide sets out exactly where its evidence stands. That is a different category from "disproven", and should not be treated the same way. Most of the time, the human trial has either not been done, or has been done and did not deliver.
The short answer
Avoid outright: high-dose vitamin E and beta-carotene supplements. A Cochrane review of 78 trials in 296,707 participants found beta-carotene and vitamin E significantly increased mortality, with higher-dose vitamin A possibly doing the same [1]. Beta-carotene increased lung cancer incidence in smokers and asbestos-exposed workers [2]. Probably not worth buying for longevity: multivitamins (no mortality benefit across 390,124 adults followed over two decades [3]), resveratrol at typical doses, spermidine at supplement doses, oral "NAD⁺" itself, greens powders and telomere products. Genuinely unsettled rather than disproven: NAD⁺ precursors, urolithin A, taurine, Ca-AKG. The honest verdict there is "early", not "works" — and not "doesn't".
Table of contents
A supplement can fail in three distinct ways, and lumping them together produces bad advice in both directions.
| Failure mode | What it means | What to do |
|---|---|---|
| Tested and negative | Adequate human trials found no benefit, or found harm | Stop. This is settled enough to act on |
| Tested at the wrong dose or form | The compound may do something; the product does not deliver enough of it | The product fails even if the compound might not |
| Never properly tested in humans | Mechanism and animal data only | Not disproven — but you are funding a hypothesis, not buying a result |
Most of the longevity aisle sits in the third category. A smaller and more important group sits in the first. Understanding which is which is most of the skill. The mechanisms themselves are usually real — they map onto the twelve hallmarks of ageing [4] — but a real mechanism is a reason to run a trial, not evidence that a capsule works.
These are the ones that matter most, because the downside is not just wasted money.
In a Cochrane systematic review of 78 randomised trials with 296,707 participants, restricted to trials at low risk of bias, vitamin E significantly increased mortality (RR 1.03, 95% CI 1.00–1.05 across 46 trials) [1]. Separately, in the SELECT trial of 35,533 men, vitamin E at 400 IU daily was associated with a significant increase in prostate cancer risk on extended follow-up [2].
The same Cochrane review found beta-carotene significantly increased mortality (RR 1.05, 95% CI 1.01–1.09 across 26 trials) [1]. In smokers and asbestos-exposed workers, the ATBC and CARET trials found beta-carotene supplementation increased lung cancer incidence [2]. This is the clearest example in nutrition of an antioxidant that looked protective in observational data and proved harmful in randomised trials.
The Cochrane analysis found dose of vitamin A significantly associated with increased mortality in meta-regression [1]. Vitamin A is fat-soluble and accumulates, which makes casual high-dose use a poor idea.
Not dangerous — simply not doing the job people buy them for. Multivitamin use showed no mortality benefit across 390,124 US adults followed for more than two decades [3], and the US Preventive Services Task Force recommends against beta-carotene and vitamin E for the prevention of cardiovascular disease or cancer [5].
An important qualification
None of this applies to correcting a diagnosed deficiency under medical guidance, or to vitamin A and E at ordinary dietary levels. The problem is high-dose isolated antioxidant supplementation in people who are not deficient. If you have been prescribed a vitamin by a clinician, do not stop it on the strength of a web article — take the question to them.
The most heavily marketed longevity compound of the last twenty years, and the one whose foundations have eroded most.
In a randomised, double-blind, placebo-controlled trial, 12 weeks of resveratrol in non-obese postmenopausal women with normal glucose tolerance did not change body composition, resting metabolic rate, plasma lipids or inflammatory markers, and did not improve insulin sensitivity in liver, muscle or fat. Notably, it also failed to affect its supposed molecular targets — AMPK, SIRT1, NAMPT and PGC-1α — despite plasma resveratrol levels rising as expected [6].
That last detail is the important one: the mechanism the entire resveratrol story rests on did not show up. Add poor oral bioavailability and the fact that human trials in metabolically healthy people have been largely unimpressive, and the longevity case is weak.
Resveratrol is not useless across the board. Trials in postmenopausal women have reported benefits for cognitive and cerebrovascular measures and for bone density. That is a legitimate, narrower finding in a specific population — it is not evidence that resveratrol slows ageing in a healthy 45-year-old, which is what it is usually sold for. See does resveratrol work? and resveratrol vs pterostilbene.
Marketed explicitly as an autophagy inducer on the strength of animal data. The best test so far randomised 100 older adults with subjective cognitive decline to a wheat-germ-derived spermidine supplement or placebo for 12 months, and found no significant benefit on memory performance or secondary outcomes [7].
The authors noted the supplement raised daily spermidine intake by only around 10%, so the honest reading is "not demonstrated at the doses commonly sold" rather than "definitively useless" [7]. Either way, the current products are not supported. See spermidine in the UK and autophagy explained.
Products selling NAD⁺ as the molecule rather than a precursor are selling a chemistry problem. NAD⁺ is a large, unstable molecule that is broken down in the digestive tract rather than absorbed intact — which is precisely why the research uses precursors such as NMN and nicotinamide riboside instead. If a label sells you NAD⁺ directly in a capsule, it is worth asking what evidence exists that it reaches your cells. See our NAD supplement guide.
Products claiming to lengthen telomeres rest on very thin human evidence, and the underlying premise is more complicated than the marketing suggests — unchecked telomere elongation is a feature of cancer cells, not an unambiguous good. Treat "telomere support" as a marketing category rather than a clinical one.
Two problems. First, they typically contain small amounts of many ingredients, none at a studied dose. Second, "detoxification" in the commercial sense is not a recognised physiological process requiring supplementation — the liver and kidneys handle it. Fibre and vegetables have real evidence behind them; a scoop of powder is not the same intervention.
Worth separating two claims. Collagen peptides have reasonable evidence for skin measures such as elasticity and hydration. That is not the same as an anti-ageing or longevity effect, and collagen is not a complete protein for building or preserving muscle, where total protein intake matters more. See astaxanthin vs collagen and supplements for skin ageing.
Fairness cuts both ways. These compounds are frequently lumped in with the failures, and that is not accurate. Their trials are small, short and mixed rather than negative.
| Compound | Where the evidence stands | Honest verdict |
|---|---|---|
| NMN and nicotinamide riboside | Reliably raise blood NAD⁺ in humans; outcome data is short-term and inconsistent, with a meta-analysis finding no effect on glucose or lipid metabolism [8] | Mechanism confirmed, benefit unproven. See is NMN a scam? |
| Urolithin A | Improved muscle endurance and mitochondrial biomarkers in a randomised trial of 66 older adults — but missed its primary endpoint of six-minute walk distance [9] | The best-evidenced compound in the category, with a real caveat. See urolithin A |
| Taurine | Extended healthspan and median lifespan in mice with supporting primate and human cohort associations; human randomised outcome trials pending | Promising, unproven in people. See taurine for longevity |
| Ca-AKG | Strong mouse healthspan data; human findings come from small, uncontrolled studies | Too early to judge. See Ca-AKG |
| Fisetin and quercetin | Plausible senolytic mechanism; human trials small and short | Research-stage. See fisetin vs quercetin |
The distinction matters practically. "Disproven" means stop. "Unproven" means you are making a personal judgement about spending money on an open question — which is a legitimate choice, provided you know that is what you are doing.
Most longevity compounds have their headline result in rodents, often at doses that would be impractical or unsafe scaled to a human. Mice are short-lived, genetically uniform and housed in controlled conditions. The history of ageing research is largely a history of mouse findings that did not replicate.
A compound has to survive digestion, reach the bloodstream and get into the relevant tissue. Resveratrol is the classic example of a molecule with poor oral bioavailability being sold as though absorption were solved.
A "proprietary longevity matrix" listing nine ingredients totalling 700 mg cannot contain a clinically studied dose of any of them. Blends exist partly to obscure this. See supplement fillers.
Raising a biomarker is not the same as improving health. NAD⁺ precursors reliably raise NAD⁺ — that part works — but a meta-analysis of randomised trials found no effect on glucose or lipid metabolism [8]. The intermediate marker moved; the outcome did not follow.
Researchers analysed 18 NMN and five urolithin A products and found actual content deviating from the label by between −100% and +28%, with three NMN products containing no detectable active ingredient at all [10]. A perfectly good compound fails if the capsule is empty. See third-party tested supplements and how to read a certificate of analysis.
Dr Zeeshan Afzal, MBBS: "The pattern I see most often is someone taking eight or nine products, unable to say what any individual one is for, and spending more than they can comfortably afford. The conversation that helps is not which to add — it is which three to keep. Almost nobody feels worse for stopping the rest, and the money is usually better spent elsewhere."
Removing things from a routine tends to be more valuable than adding them. In rough order of return:
On realistic timescales and costs, see how long supplements take to work, longevity stack cost in the UK and why supplements are expensive.
Speak to your GP or pharmacist before stopping any supplement that was recommended or prescribed by a clinician — particularly vitamin D, B12, iron or folic acid, which are often prescribed for a specific reason. This article is about self-selected longevity products, not treatment.
Speak to your GP if you take prescription medication alongside supplements, especially anticoagulants, blood pressure or diabetes medication, or if you are pregnant, trying to conceive, breastfeeding or under 18. Your community pharmacist can answer interaction questions without an appointment.
Do not use supplements to self-treat symptoms. Persistent fatigue, unintended weight loss, breathlessness, chest pain or a change in bowel habit require assessment — delay matters for several conditions that present this way.
The clearest failures are high-dose vitamin E and beta-carotene, which increased mortality in a Cochrane review of 78 trials covering 296,707 participants [1], and multivitamins as a longevity intervention, which showed no mortality benefit across 390,124 adults [3]. Resveratrol, spermidine at typical supplement doses, oral NAD⁺, telomere products and greens powders all lack supporting human outcome evidence.
For general longevity purposes, the evidence is weak. A randomised trial in non-obese postmenopausal women found no effect on body composition, lipids, inflammatory markers or insulin sensitivity — and, crucially, no effect on its supposed molecular targets including SIRT1 and AMPK [6]. Trials in postmenopausal women have reported narrower benefits for cognition and bone, which is a different and more specific claim.
High-dose isolated antioxidants can be. In low-risk-of-bias trials, beta-carotene (RR 1.05) and vitamin E (RR 1.03) significantly increased mortality, and higher doses of vitamin A may do the same [1]. Beta-carotene increased lung cancer incidence in smokers and asbestos-exposed workers [2]. Antioxidants from food do not carry the same signal.
It reliably raises blood NAD⁺ in humans — that part is established. Whether that produces meaningful health benefits is unresolved: a meta-analysis of randomised trials found no effect on glucose or lipid metabolism [8]. That makes it unproven rather than disproven, which is a genuinely different position from resveratrol. Our NMN evidence guide sets out the trials in detail.
Mice are short-lived, genetically uniform and kept in controlled conditions, and doses used in rodent studies are often far higher relative to body weight than anything sold to humans. Compounds also have to survive digestion and reach the right tissue in a human, which many do not.
Not as a longevity strategy. It showed no mortality benefit in 390,124 adults followed for over two decades [3], and the USPSTF recommends against beta-carotene and vitamin E for preventing cardiovascular disease or cancer [5]. Correcting a specific, identified gap is a better use of the money.
Often much less than it implies. It may refer to a single ingredient rather than the finished product, at a different dose, in a different population, over a few weeks — or to a study in cells or animals. Ask which study, how many people, how long, and whether the primary endpoint was met.
Generally not. They typically contain small amounts of many ingredients, none at a studied dose, and "detoxification" in the commercial sense is not a physiological process requiring supplementation. Vegetables and fibre have real evidence; a scoop of powder is not equivalent.
Look for a batch-specific certificate of analysis matching the number on your tub, independent third-party testing rather than in-house, and a stated purity figure. This is not a theoretical concern: an analysis of 18 NMN and five urolithin A products found deviations from the label between −100% and +28%, with three containing no detectable active ingredient [10].
Yes, but the list is shorter and duller than the marketing suggests: vitamin D through UK autumn and winter, omega-3 if you rarely eat oily fish, magnesium, creatine, and correction of any tested deficiency. Beyond that, exercise, sleep, blood pressure and glucose control outperform everything in the supplement aisle.
Medical disclaimer
This article is for general information and does not constitute medical advice, diagnosis or treatment. It discusses the general evidence base for supplement categories and is not an assessment of any specific product. Food supplements are not a substitute for a varied, balanced diet and a healthy lifestyle, and should not be used to treat, prevent or cure disease. No supplement has been shown to extend human lifespan. Do not stop taking any vitamin or supplement that has been recommended or prescribed by a healthcare professional on the basis of this article — discuss it with them first. Consult your GP or pharmacist before starting or stopping supplements if you take prescription medication, have an existing condition, or are pregnant, trying to conceive, breastfeeding or under 18. The findings on high-dose vitamin E, beta-carotene and vitamin A concern high-dose isolated supplementation, not dietary intake or clinically supervised correction of a diagnosed deficiency. Do not use supplements to self-treat symptoms: persistent fatigue, unintended weight loss, breathlessness, chest pain or a change in bowel habit require prompt medical assessment. Reviewed for medical accuracy by Dr Zeeshan Afzal. See the full Welzo medical disclaimer.
All photography in this article is owned by Welzo and served from the Welzo media library. Research positions correct at the time of publication.
Rated Excellent by 26,523+ Reviews
100% Money Back Guarantee
Our money back guarantee ensures you receive your items or get a full refund.
