Every supplement industry has a favourite piece of paper. A certificate, a seal, a QR code on the tub, a line on the label saying the batch was tested by an independent laboratory. It is one of the few things in this sector that a reasonable person is entitled to find persuasive, and it is also the thing most often asked to carry more weight than a single page of results can bear.
Three surveys, three countries, and no contradiction between them: undeclared substances in supplements are common, and the scale is documented. The question this article is about is narrower and more useful — what an independent test result does and does not tell you.
If you want the mechanics of reading one of those reports, younglab.eu has a page written for exactly that: what third-party testing actually proves.
875 products out of 3 132
The 2022 review on unintentional doping searched four databases, with the last search in June 2021, and pooled 50 studies. The headline arithmetic is unusually clean: of 3 132 dietary supplements analysed, 875 were found to contain substances that were not declared on the label. The authors describe more than 28 percent of the analysed products as posing a risk to the consumer.
The two most frequently found undeclared substances were sibutramine and anabolic-androgenic steroids. Neither is a trace contaminant of a plant extract. Sibutramine is a withdrawn weight-loss drug; steroids are a deliberate choice with a pharmacological reason in it.
38 percent in a defined market segment
The 2021 Dutch survey is smaller but methodologically more interesting, because it describes its own sampling frame in detail. The authors selected 66 sports nutrition supplements — chosen precisely because they claimed to modulate hormone regulation, build muscle, reduce fat or boost energy — purchased from 21 brands through 17 web shops, and analysed them at LGC using an extended version of an ISO 17025 accredited screen.
25 of the 66 products, 38 percent, contained undeclared doping substances. The list is worth reading as a catalogue of what a supplement label can hide: the stimulants oxilofrine, BMPEA and NBDMPEA, DMAA, the steroids boldione and 17alpha-AED, the beta-2 agonist higenamine, and the beta-blocker bisoprolol.
A beta-blocker. In a fat-loss product.
The regulator’s own list
The third dataset is not a survey of products but of warnings, and it is the largest in absolute terms. Between 2007 and 2016, the FDA issued warnings covering 776 adulterated dietary supplements, implicating 146 different companies. The category split is the part people argue about least:
- Sexual enhancement: 353 products, 45.5 percent.
- Weight loss: 317 products, 40.9 percent.
- Muscle building: 92 products, 11.9 percent.
And 157 products, 20.2 percent, contained more than one unapproved ingredient. Multiple undisclosed drugs in a single capsule is not a contamination story; it is a formulation decision.
So what does a certificate prove?
Here is the honest version, and it is more useful than the marketing version. An independent test result establishes that a sample drawn from one batch, analysed for the substances on the panel, at one moment, came back as reported. That is a real and substantial claim. It is also narrower than the word certificate implies in conversation.
Three things a single report cannot tell you:
- Whether the next batch was made the same way. Results attach to a batch number, not to a brand.
- Whether the panel was the right panel. A screen that looks for a defined list of compounds reports on that list. Nothing in the format certifies what was not searched for.
- Whether the sample came from the tub you are holding. Sampling is the weakest link in most of this, and it is the one almost never documented on the page a consumer sees.
What this evidence does not show
- It does not show a failure rate for the certified segment. All three datasets study adulterated or high-risk categories. None of them estimates how often a batch carrying a third-party certificate is itself adulterated.
- It does not show that testing reduces harm in a population. The surveys are descriptive. No controlled study here compares outcomes in users of tested versus untested products.
- It does not separate contamination from deliberate addition in the pooled review. The 875 figure covers both mechanisms, and they have very different implications for how much a certificate can be expected to catch.
- The 2021 study is Dutch and its sample was deliberately high-risk. It measures a worst-case segment, not a market average.
- None of the three tells you anything about efficacy, which is a completely separate question from what is in the product.
In one line
Independent testing is the best tool this industry has and it is still a narrow one: it certifies a batch, against a list, on a day — and the three surveys above are the reason it exists.
