Guide
Ingredients avoided for the wrong reason
A list of corrections, each one traced to the document that settled it, and each one still circulating anyway.
Part of Banned elsewhere · Sensitive skin
This site spends most of its time saying an ingredient is more complicated than its reputation. Occasionally the honest answer is simpler: the reason people avoid it turned out to be wrong, and the correction never travelled as far as the alarm did.
Here they are together, each with the document that settled it. Every row links to a page that does the work properly, including the parts that remain unresolved.
| Ingredient | What circulates | What the record says |
|---|---|---|
| Saccharin | A cancer warning from the 1970s | Delisted in 2000. The rat bladder tumours depended on conditions that “would not be expected to occur in humans” |
| BHT | Banned in Europe | Permitted in the EU as E321, and GRAS in the US at 0.02 percent of fat content |
| Parabens | A 2004 study found them in breast tumours | It found them present, not causal. Reviews call an oestrogen-mediated effect at cosmetic exposures “biologically implausible” |
| Azodicarbonamide | The yoga mat chemical | True and irrelevant. Shared industrial use is not a fact about a 45 ppm food dose |
| Monosodium glutamate | Causes a syndrome | The controlled trials did not reproduce it; the name has to be declared, which is the useful part |
| Dimethicone | Suffocates skin and clogs pores | Hair and skin do not respire. Its comedogenic reputation comes from a rating derived from rabbit ears |
| Maltodextrin | Hidden gluten | US maltodextrin is corn, potato or rice by regulation. European wheat-derived maltodextrin is exempt because the protein is removed |
| Carrageenan | Settled harm | Genuinely unsettled. Permitted, with regulators asking for better data, which is a third answer |
The two that are not corrections
Two rows there deserve to be read differently, and putting them in the same table is deliberate.
Carrageenan is not a debunking. It sits here because the popular version and the careful version are both wrong: it is neither proven harmful nor cleanly cleared. The honest status is unresolved, and a list of corrections that quietly converted it into good news would be doing the thing this page exists to complain about.
Erythritol went the other way. For years it was the uncomplicated sugar alcohol. Then a 2023 study linked blood levels to cardiovascular events, and the question is now open rather than closed. Evidence moves in both directions and a page that only ever moves toward reassurance is not tracking evidence, it is tracking a preference.
What the good corrections have in common
Read the saccharin story and the parabens story next to each other and the same shape appears twice.
In both cases a real finding was reported accurately. Rats fed sodium saccharin did get bladder tumours. Parabens were found in breast tumour tissue. Neither report was fraudulent or careless.
What was missing in both cases was the mechanism. Once mechanistic work showed the rat tumours depended on urinary precipitate at dietary concentrations of 3 percent and above, conditions that do not arise in people, the same data supported the opposite conclusion. Once exposure was measured, an oestrogenic effect from cosmetic paraben doses became implausible rather than merely unproven.
Presence is not cause, and a finding without a mechanism is a question rather than an answer. That is the argument of the guide on hazard and exposure, and it is the single most useful idea on this site.
Why the correction never catches up
Because the alarm is a better story. Found in tumours fits in a headline; the mechanism is specific to rat urine chemistry at 3 percent dietary concentration does not.
There is also a structural asymmetry. Nobody builds an audience by telling people something is fine. The correction has no constituency, and the alarm has a large one.
That is not a reason to be reflexively reassuring, which is its own failure and a more comfortable one to fall into. Red 3 is on this page’s cluster and its listing was genuinely revoked. Titanium dioxide genuinely lost its EU food authorisation. The point is not that the worries are always wrong. It is that each one has a document behind it, and the document is usually more interesting than either the alarm or the reassurance.
Where the difference between countries fits
Several of these travel with a “banned in Europe” attached, and that phrase records at least four different things: a regulator assessed and found a problem, a statute forced an outcome, nobody filed an application, or the claim is simply false.
BHT is the fourth kind and it circulates in the company of the first kind, which is most of why it is believed. The full sorting is on the page about ingredients banned somewhere else.
Sources
- Hazardous Waste Management System; Removal of Saccharin and Its Salts From the Lists of Hazardous Constituents, Hazardous Wastes, and Hazardous Substances US Environmental Protection Agency, 75 FR 78918, 17 December 2010, via the Federal Register · federalregister.gov Records the 1981 NTP listing of saccharin, its delisting in 2000, and the finding that the factors contributing to tumour induction in rats would not be expected to occur in humans.
- 21 CFR 182.3173: Butylated hydroxytoluene US Food and Drug Administration, via eCFR · ecfr.gov Generally recognised as safe when the total content of antioxidants is not over 0.02 percent of the fat or oil content. BHT is also permitted in the EU as E321.
- A perspective on the safety of parabens as preservatives in wound care products Torfs and Brackman, International Wound Journal, via PubMed Central · pmc.ncbi.nlm.nih.gov States that it is biologically implausible that parabens could increase the risk of any oestrogen-mediated endpoint at the exposures cosmetics deliver.
- 21 CFR 172.806: Azodicarbonamide US Food and Drug Administration, via eCFR · ecfr.gov Permits it in cereal flour and as a dough conditioner at not more than 45 parts per million by weight of the flour used.