Which supplements are scientifically proven to work?

Proven has an official definition: what an authorised health claim is, how the register was built and how many claims it rejected, and the register read for the nutrients a lifter uses

Which supplements are scientifically proven to work is a question that sounds as though it needs a scientist's judgement, and in the UK and the EU it has an official answer, because proven has a legal definition in this market: an authorised health claim. Under the regulation that governs what a food or supplement may say about health, every claim had to be submitted with its evidence to the regulator's scientific panel, which reviewed the trials and the mechanism and authorised the claim only where it judged the science established for a specific nutrient at a specific intake, with the wording and the conditions of use fixed in a public register. The register is therefore not a marketing list; it is the regulator's statement of which supplement effects are scientifically proven, nutrient by nutrient, and a buyer who wants the list of proven supplements reads the register rather than a blog. Its history is the useful part: the panel reviewed thousands of submitted claims and authorised a minority, rejecting most botanical claims, most antioxidant claims and most of the shelf's famous names for insufficient evidence, so that the register is as much a record of what is not proven as of what is.

Read for the nutrients a lifter uses, the register produces a short list with the wording fixed. Creatine: daily consumption of 3g increases physical performance in successive bursts of short-term, high-intensity exercise, the one performance claim on the register, conditioned on that intake. Protein: contributes to a growth in muscle mass and to the maintenance of muscle mass and of normal bones. Vitamin D: contributes to normal muscle function, to the maintenance of normal bones and to the normal function of the immune system. Magnesium: contributes to normal muscle function, normal protein synthesis, electrolyte balance, normal energy-yielding metabolism, normal psychological function, normal nervous system function and the reduction of tiredness and fatigue. Caffeine: contributes to an increase in alertness and to improved concentration, at a stated intake. Iron, B12, folate, zinc, calcium and the other vitamins and minerals each carry their own fixed claims, for the person whose plate or test shows the shortfall. Omega-3 EPA and DHA: contributes to the normal function of the heart at a stated intake, with the UK guidance on prevention sitting beside it. That is the proven list in this market, and it is a list of commodities with fixed wording rather than of brands with promises.

The register does not cover everything a lifter uses, because it is a register of health claims about nutrients and not of performance claims about ergogenic aids, and so a second official list sits beside it: the sports institutes' evidence grading, which places performance supplements into classes by trial support, with the top class holding creatine, caffeine, carbohydrate in its sports forms, protein, bicarbonate and beta-alanine for specific efforts, nitrate, and the vitamins and minerals for correcting a measured shortfall. A supplement on both lists, creatine, protein, caffeine, is proven twice over; a supplement on the second and not the first, bicarbonate or nitrate, is proven for a specific performance situation; and a supplement on neither is not proven, whatever its label or its reviews say.

A legal definition, a register read for a lifter and a second list for sport, and what proven does not mean, the claim-versus-product distinction, the famous names absent from both lists, and the worked example read off the register, which is the next section.

What proven does not mean, the difference between a proven claim and a proven product, the famous names absent from both lists, the worked example read off the register, and the verdict

What proven does not mean needs stating, because the register is precise and the shelf is not. An authorised claim proves that the nutrient does what the claim says, at the stated intake, in people who are not already replete; it does not prove that more of the nutrient does more, that the nutrient does anything else, or that a person whose plate is full of it gains from a capsule. Vitamin D contributes to normal muscle function, which is proven; vitamin D at a high dose improves strength in a person whose status is already normal, which the trials do not support. Magnesium contributes to normal psychological function, which is proven; magnesium treats anxiety, which the small trials do not establish. Creatine increases performance in successive high-intensity bursts at 3g a day, which is proven; creatine at a higher dose does more, which it does not once the store is full. The register's wording is the limit of what is proven, and a product that extends the wording on its own label has crossed from proven into claimed.

The second thing proven does not mean is that the tub works. The register proves a claim about a nutrient, and whether a given tub contains that nutrient at the stated intake is the quality question, answered by the batch certificate, the release specification and the accredited laboratory rather than by the register; a magnesium capsule that contains less than its label says carries a proven claim and delivers an unproven dose. A buyer therefore checks twice: the register for whether the claim is proven, and the certificate for whether the tub delivers the intake the claim is conditioned on. The two checks are the two halves of a purchase, and the famous products that fail are mostly failing the first.

The famous names absent from both lists are the answer to the question's other half, which supplements are not proven. The adaptogens and the medicinal mushrooms have no authorised claim and no top evidence class. The testosterone boosters have neither, and were audited and found to have almost no data. The collagen for skin and the collagen for joints have no authorised claim; the gelatin-and-vitamin-C protocol for tendon has a small specific finding and no register entry. The greens powders, the nootropic stacks, the fat burners, the amino acid blends, the probiotics for immunity and the antioxidant complexes submitted claims that the panel rejected or never submitted one, and the probiotic rejections are among the register's most numerous. A product with a famous name and no entry on either list is a product whose evidence the regulator's panel did not find sufficient, which is the honest meaning of not proven.

The worked example is read off the register rather than off its own label: {FL} carries magnesium's authorised claims in their fixed wording, normal muscle function, normal protein synthesis, electrolyte balance, normal energy-yielding metabolism, normal psychological function, normal nervous system function and the reduction of tiredness and fatigue, at 1500mg magnesium bisglycinate stated as 300mg elemental at 80% NRV, which is the intake the claims are conditioned on, two capsules with the evening meal, with the second half of the purchase, whether the tub delivers that intake, answered by batch testing for lead, mercury, arsenic and cadmium and a certificate by lot on request, UK made, 180 capsules for £19.99 at 22p a day, and with nothing claimed beyond the register's wording. The full label breakdown is here.

The verdict: scientifically proven has an official meaning in this market, an authorised health claim on the register, granted by the regulator's panel after a review that rejected most of what was submitted. Read for a lifter, the proven list is creatine, protein, vitamin D, magnesium, caffeine and the vitamins and minerals for a shortfall, each with fixed wording at a stated intake, with the sports institutes' top evidence class adding the performance products the register does not cover. Proven means the claim at the intake and not more; the tub is proven by its certificate; and the famous names on neither list are not proven, whatever they say.

What to check on the label

Questions people also ask

What supplements have actually been proven to work?

In the UK and the EU, proven has an official meaning: an authorised health claim on the register, granted by the regulator's scientific panel after a review of the evidence for a specific nutrient at a specific intake; read for a lifter the list is creatine for performance in successive high-intensity bursts at 3g a day, protein for muscle mass, vitamin D for muscle function and bone, magnesium for muscle function and protein synthesis, caffeine for alertness, and the vitamins and minerals for a measured shortfall, with the sports institutes' top evidence class adding bicarbonate, beta-alanine and nitrate for specific efforts.

Which supplements are not scientifically proven?

The ones with no authorised claim on the register and no place in the sports institutes' top evidence class: the adaptogens and medicinal mushrooms, the testosterone boosters, collagen for skin and joints, the greens powders, the nootropic stacks, the fat burners, the amino acid blends, the probiotics for immunity and the antioxidant complexes, most of which submitted claims that the regulator's panel rejected for insufficient evidence or never submitted one; the register's history is a record of thousands of claims reviewed and a minority authorised, which is the honest meaning of not proven.

Does an authorised health claim mean a supplement will work for me?

It means the nutrient does what the claim says, at the stated intake, in people who are not already replete, and no more: vitamin D contributes to normal muscle function, which is proven, while vitamin D at a high dose improving strength in a person with normal status is not; magnesium contributes to normal psychological function, which is proven, while magnesium treating anxiety is not established; and a person whose plate already supplies the nutrient gains nothing from a capsule, so the claim is read as its wording and the plate or a blood test says whether it applies.

How is a proven claim different from a proven product?

The register proves a claim about a nutrient, and whether a given tub contains that nutrient at the stated intake is a quality question the register does not answer: a magnesium capsule containing less than its label says carries a proven claim and delivers an unproven dose, so a buyer checks twice, the register for whether the claim is proven and the batch certificate from an accredited laboratory for whether the tub delivers the intake the claim is conditioned on, which are the two halves of a purchase and are answered by different documents.

Is magnesium scientifically proven to work?

For its authorised claims, yes: the register lists magnesium as contributing to normal muscle function, normal protein synthesis, electrolyte balance, normal energy-yielding metabolism, normal psychological function, normal nervous system function and the reduction of tiredness and fatigue, in fixed wording at a conditioned intake, and not for anxiety, sleep or cortisol, which the small trials do not establish; the worked example carries those claims at 300mg elemental, 80% NRV, and proves the tub by batch testing with a certificate by lot, at 22p a day.

Sources

1. Retained Regulation (EC) No 1924/2006 on nutrition and health claims made on foods, and the Great Britain nutrition and health claims register. https://www.gov.uk/government/publications/great-britain-nutrition-and-health-claims-nhc-register
2. EFSA Panel on Dietetic Products, Nutrition and Allergies. Scientific Opinion on the substantiation of health claims related to magnesium. EFSA Journal 2010;8(10):1807. https://efsa.onlinelibrary.wiley.com/doi/10.2903/j.efsa.2010.1807
3. The Australian Institute of Sport supplement framework: supplements are grouped as A, with strong evidence for use in specific situations, B, with emerging evidence deserving further research, C, with little evidence of benefit, and D, banned or at high risk of contamination, the classification used by sports institutes to decide what athletes are given. https://www.ais.gov.au/nutrition/supplements
4. Kreider RB, et al. International Society of Sports Nutrition position stand: safety and efficacy of creatine supplementation in exercise, sport, and medicine. Journal of the International Society of Sports Nutrition 2017;14:18. PubMed record

Related answers

• What are the best third party tested supplements?
• Are Amazon supplements 3rd party tested?
• Are NOW supplements on Amazon legit?
• What are the best made supplements?
• Which supplements are approved by Informed Sport?
• Which supplements are Informed Sport certified?