What supplements have the most scientific evidence?

Evidence counted three ways: trials, pooled participants and consistency, why one count misleads, where each count is read, and the supplements at the top by all three

What supplements have the most scientific evidence is a question that gets a different answer depending on what is counted, and the shelf counts the thing that flatters its stock. A product page that says backed by hundreds of studies has counted trials, and a supplement can have hundreds of trials that disagree with each other, which is not evidence for anything except that it has been studied; a page that says a study of ten thousand people has counted participants, and one large trial can be wrong in a way that twenty small ones that agree are not. The honest count is three counts taken together: the number of randomised controlled trials, which says how often the question has been asked properly; the number of participants pooled in the meta-analyses of those trials, which says how much weight the pooled answer carries; and the consistency of the result, the degree to which the trials point the same way with a similar size of effect, which the meta-analyses report as heterogeneity and which is the count the shelf never mentions. A supplement with the most scientific evidence is one that scores high on all three, and that rule sorts the shelf in a way the first count alone never could.

The counts are read in the evidence hierarchy's top tier, the systematic reviews and meta-analyses of randomised trials, which pool the trials, report the number pooled, and report whether they agree. Creatine monohydrate scores at the top on all three: hundreds of trials, thousands of participants pooled across the meta-analyses of strength and lean mass with resistance training, and a result that points the same way almost every time, which is why it holds an authorised claim and a class A entry and sits at the top of every consensus statement. Caffeine scores the same way for performance, with hundreds of trials and consistent small gains in strength and larger ones in endurance at a few milligrams per kilogram. Protein scores the same way for muscle mass with resistance training, with the dose-response meta-analysis pooling dozens of trials and finding a consistent rise in lean mass gain up to a daily total near one point six grams per kilogram. Those three have the most scientific evidence by the only count that means anything, all three counts at once.

The next tier is specific rather than general. Beta-alanine has dozens of trials, a pooled meta-analysis of over a thousand participants, and a consistent small gain in efforts lasting one to four minutes, which is strong evidence for a narrow claim; nitrate from beetroot has a similar profile for endurance in the less trained. Both have the most evidence for the thing they do and no evidence for the things the shelf extends them to, which is the difference between a consistent result and a broad one.

Three counts and the top of the ranking, with the corrections that have large consistent evidence in the deficient, the supplements with volume and no agreement, the consensus check, the worked example and the verdict in the next section.

The corrections with large consistent evidence in the deficient; the supplements with the most trials and the least evidence; the sports-institute and Olympic consensus as the check; the worked example on its own evidence for a shortfall; and the verdict

A separate group scores high on all three counts in a population the shelf does not mention: the deficient. Vitamin D has hundreds of trials and tens of thousands pooled, and the consistent finding is that it corrects muscle function and bone outcomes in people who were low and does nothing measurable in people who were not, which is a large consistent body of evidence for a correction and none for an enhancement. Omega-3 has the same shape for triglycerides, with consistent reductions at a few grams a day in people whose levels were raised, and a far less consistent record for everything else sold under its name. Iron, B12 and folate have the same profile in their deficiencies. These supplements have the most scientific evidence for the person a test finds short and the least for the person it finds replete, so that the question which has the most evidence has, for this group, a second question behind it, evidence for whom.

Then there is the group with the most trials and the least evidence, which the first count alone would put at the top. The antioxidant vitamins, C and E and beta-carotene, have hundreds of trials and tens of thousands pooled in the meta-analyses, and the pooled result is nothing, or in the training literature a blunting of the adaptation the session was meant to produce, with the trials disagreeing in direction. The multivitamins have the same shape, large trials over years finding no benefit on the outcomes that matter. Those are the supplements the shelf describes as the most studied, and they are, and the studies agree on nothing useful; volume without agreement is not evidence, and a buyer who counts trials alone will buy exactly the products that count condemns.

The check is that the people whose job is to read all three counts arrive at the same ranking: the sports-institute classification puts creatine, caffeine, protein, beta-alanine, nitrate and the corrections in its evidence-backed class, and the Olympic committee's consensus statement on supplements for the high-performance athlete names the same few, with the rest of the shelf in the categories below. A buyer who wants the ranking without the reading can take theirs, because they counted the same three things.

The worked example is placed on its own evidence, which is evidence for a shortfall and not for an effect in the replete: {FL}, 1500mg magnesium bisglycinate stated as 300mg elemental at 80% NRV, two capsules with the evening meal, holds authorised claims for normal muscle function and the reduction of tiredness and fatigue that were assessed on the mineral's evidence, and the magnesium trials in replete lifters are small, few and inconsistent, which this page states rather than inflates; it is taken for the share of adults the survey finds below the reference intake, batch tested for lead, mercury, arsenic and cadmium, UK made under GMP and BRC, 180 capsules for £19.99 at 22p a day. The full label breakdown is here.

The verdict: the supplements with the most scientific evidence are the ones that score on all three counts, trials, pooled participants and consistency, and by that rule creatine monohydrate, caffeine and protein with resistance training lead, beta-alanine and nitrate follow for the specific efforts they help, and vitamin D, omega-3 for triglycerides, iron, B12 and folate have the most evidence for the deficient and the least for the replete. The antioxidants and the multivitamins have the most trials and no agreement, and a magnesium has evidence for a shortfall and nothing beyond it. The shelf counts one thing; the ranking needs three.

What to check on the label

Questions people also ask

What is the most scientifically proven supplement?

Creatine monohydrate, by the three counts that together mean evidence: hundreds of randomised trials, thousands of participants pooled across the meta-analyses of strength and lean mass with resistance training, and a result that points the same way almost every time, which is why it holds an authorised claim for physical performance in short bursts, a class A entry at the sports institutes and the top place in the Olympic consensus statement; caffeine and protein with resistance training score the same way, and the three together are the supplements with the most evidence for a person who trains.

Which supplements have the strongest evidence for building muscle?

Protein and creatine monohydrate, both with resistance training and both consistent across many trials: the dose-response meta-analysis pools dozens of protein trials and finds lean mass gain rising with total daily protein to near one point six grams per kilogram a day, and the creatine meta-analyses pool hundreds of trials and find strength and lean mass gains over training alone; caffeine adds a consistent performance gain to the session, beta-alanine a consistent gain in efforts of one to four minutes, and nothing else on the shelf has a consistent muscle result in trained people, with the antioxidants blunting the adaptation in some trials.

Are the most studied supplements the most effective?

Often the reverse, which is why trials are never counted alone: the antioxidant vitamins and the multivitamins are among the most studied supplements in existence, with hundreds of trials and tens of thousands of participants pooled, and the pooled result is nothing on the outcomes that matter, with the trials disagreeing in direction and the antioxidants blunting training adaptation in some; a supplement is studied heavily because a question was interesting, and the answer is read from consistency across the trials, not from how many there were, so that most studied on a label is a warning until the agreement is checked.

Does vitamin D have strong scientific evidence?

Strong and specific: hundreds of trials and tens of thousands pooled find that vitamin D corrects muscle function and bone outcomes in people who were low and does nothing measurable in people who were not, which is a large and consistent body of evidence for a correction and none for an enhancement, and the UK public health advice draws exactly that line by recommending ten micrograms a day through the autumn and winter for everyone, when the sun is too low for the skin to make it; the question which supplement has the most evidence has, for vitamin D and for omega-3, iron, B12 and folate, the second question evidence for whom, and a test answers it.

Does this site's magnesium have scientific evidence?

For a shortfall, and it says so: the authorised claims it carries for normal muscle function and the reduction of tiredness and fatigue were assessed on the mineral's evidence by the European authority, the national survey finds a share of adults below the reference intake, and the magnesium trials in replete lifters are small, few and inconsistent, which is stated rather than inflated; the product is 1500mg magnesium bisglycinate as 300mg elemental at 80% NRV in two capsules with the evening meal, batch tested for four metals, UK made under GMP and BRC, at 22p a day, taken for the plate the survey finds short and not for a performance effect in the replete.

Sources

1. The evidence hierarchy in clinical research: systematic reviews and meta-analyses of randomised controlled trials sit above single randomised trials, which sit above pilot and uncontrolled studies, which sit above animal and cell work, and a claim is proven in proportion to the highest level it reaches and the number and size of studies at that level. PubMed record
2. Rawson ES, Volek JS. Effects of creatine supplementation and resistance training on muscle strength and weightlifting performance. Journal of Strength and Conditioning Research 2003;17(4):822-831: creatine with resistance training produced around 8 per cent greater gains in maximal strength and around 14 per cent greater gains in repetitions at a given load than training alone. PubMed record
3. Saunders B, et al. Beta-alanine supplementation to improve exercise capacity and performance: a systematic review and meta-analysis. British Journal of Sports Medicine 2017;51(8):658-669: a small benefit, around 3 per cent, in exercise lasting 30 seconds to 10 minutes, with the effect concentrated in efforts of one to four minutes and no effect in shorter or longer exercise. PubMed record
4. Maughan RJ, et al. IOC consensus statement: dietary supplements and the high-performance athlete. British Journal of Sports Medicine 2018;52(7):439-455: the performance supplements with good evidence are caffeine, creatine, nitrate, beta-alanine and sodium bicarbonate, with most other products lacking support. PubMed record

Related answers

• Who makes the most trusted supplements?
• What is the best cheap supplements?
• Is it safe to buy supplements from Amazon?
• What are the best third party tested supplements?
• How to spot fake supplements on Amazon?
• Are Amazon supplements 3rd party tested?