What Counts As Good Evidence in Nutrition

The supplement industry has a habit of using the word "evidence" the way a student uses the word "research" in an essay written the night before. Loosely. Impressively. Without much behind it. "Clinically proven," "backed by science," and "evidence-based formula" all sound reassuring and tell you almost nothing about what was actually studied, on whom, at what dose, and whether the findings survived peer review.

The Evidence Hierarchy in Nutrition Science

Not all evidence is equal. The hierarchy, roughly from weakest to strongest:

Anecdotal Evidence and Testimonials

"I took this and felt amazing." You have seen these on Instagram, on product pages, sometimes from people you trust. But a testimonial tells you what one person experienced, which could be the supplement, the placebo effect, a coincidental dietary change, or the passage of time. It tells you nothing about whether the product will do anything for you.

In Vitro and Animal Studies

"This compound reduced inflammation in mouse liver cells." Useful for understanding mechanisms, but not evidence that taking a supplement will produce the same effect in your body. The gap between a cell culture and a human digestive system is enormous. We covered this gap in our clinical research assessment for phytoplankton specifically.

Small Human Trials

A study of 20 to 50 people showing a positive result is a signal, not a conclusion. Small studies are more vulnerable to statistical flukes and may not represent the broader population. They are worth reading. They are not worth treating as settled science.

Randomised Controlled Trials

The gold standard: randomised, placebo-controlled, adequately powered, pre-registered, and published in peer-reviewed journals. This is the level EFSA requires before authorising a health claim. When we went through the EFSA register for our own products, we were struck by how few supplement claims of any kind have cleared this bar. It is also the level most supplement marketing claims have never reached.

Systematic Reviews and Meta-Analyses

These aggregate the results of multiple trials to draw broader conclusions. They are the strongest form of evidence when done well and the most misleading when done poorly. A meta-analysis that pools five weak studies does not produce one strong conclusion. It produces a well-organised collection of weak evidence.

How to Spot Weak Evidence Claims on Supplement Products

"Studies show..." without naming the study, the journal, or the year. This is the nutritional equivalent of "people say." Which studies? Where? When?

In vitro research presented as human evidence: "Shown to reduce oxidative stress" may mean "in a petri dish." That is not the same as "in you."

A single study cited as definitive proof: one study is a data point. Replication across independent groups is what makes it evidence. The supplement industry chronically cites single trials with dramatic results.

"Clinically proven" without a citation: proven where, by whom, and published in what journal? If the company cannot answer, the claim is decoration.

Sliding from authorised to unauthorised claims: saying EPA and DHA "contribute to normal heart function" is close to the EFSA wording, but then adding "boosts immunity" in the next sentence crosses from regulated territory into marketing. We listed the full authorised claims so you can check.

What Good Nutrition Evidence Looks Like in Practice

The EFSA health claims for omega-3 are good evidence in action: they survived independent scientific assessment and specify the nutrient, the health relationship, the intake condition, and the target population. They are bounded: "contributes to normal function," not "cures" or "prevents." That precision is what distinguishes a defensible claim from a marketing wish.

When we make claims about our products, we classify them by evidence level using our five-level classification system: established science, ingredient-level research, product-specific data, EFSA-authorised claims, or editorial assessment. The classification itself is a quality signal. A company willing to show you which tier its claims sit on is more trustworthy than one that presents everything as equal certainty.

Why Replication Matters More Than Single Breakthrough Studies

A single positive study is exciting. It is also preliminary. When we assessed the evidence behind marine phytoplankton claims, we found several that rested on a single trial. The honest response was to describe those findings as preliminary rather than proven, even though "proven" would have sold more product.

The history of nutrition science is littered with findings that looked compelling in one trial and evaporated on replication. Antioxidant supplements reducing cancer risk. High-dose vitamin E preventing heart disease. All showed promise initially. All failed on larger follow-ups. When you see a brand citing one study as settled science, ask whether anyone else has found the same thing.

What Counts as Good Evidence FAQs

How can I check whether a health claim is EFSA-authorised?

The EU Register of Nutrition and Health Claims is publicly searchable. Look up the specific nutrient and health relationship. If the exact wording appears in the register with its conditions, the claim has been independently assessed. If it does not, the brand is making an unauthorised claim regardless of how scientific the language sounds.

What does "clinically proven" usually mean on a supplement label?

Often less than it implies. Check whether the clinical evidence is for the exact product or the ingredient category, whether the trial was dose-matched and placebo-controlled, and whether it was published in a peer-reviewed journal. In many cases, the phrase refers to a single small study or to general ingredient research rather than product-specific evidence.

Is manufacturer-funded research automatically unreliable?

No. Manufacturer-funded research can be rigorous and well-designed. But it warrants the same scrutiny you would apply to any research where the funder has a financial interest in the outcome. Check whether it was pre-registered, independently peer-reviewed, and whether the results have been replicated by groups without a commercial stake.

Why do supplement companies cite animal studies as though they prove human benefits?

Because animal studies are cheaper, faster, and more likely to produce dramatic results than human trials. They are useful for understanding biological mechanisms but do not prove that a supplement will produce the same effect in your body. The translation gap between a mouse model and a human digestive system is the part the marketing conveniently omits.

Does Phytality classify the evidence behind its own claims?

Yes. Every claim on our website is assigned to one of five evidence levels: established nutrition science, ingredient-level research, product-specific data, EFSA-authorised health claims, or editorial assessment. The classification is published in our claims and evidence matrix so you can see which tier each statement sits on.

This article is for informational purposes only and is not a substitute for professional medical advice, diagnosis, or treatment. Consult your GP or a qualified healthcare professional before starting any supplement.

Methodology and Disclosure

The evidence hierarchy described reflects standard research methodology as applied in nutrition science. EFSA health claims cite Commission Regulation (EU) No 432/2012 and Regulation (EC) No 1924/2006. EFSA guidance cites EFSA Journal 2016. Examples of failed nutritional hypotheses (antioxidants, vitamin E, beta-carotene) reflect published outcomes from the cited trials.

Vendor disclosure: Phytality manufactures supplements and applies a five-level evidence classification to its content. Phytality has a commercial interest in evidence literacy being valued. The hierarchy and red flags described apply to the entire supplement industry, including our own products.