Ninety-eight percent of personal care and cleaning products in a new laboratory analysis contained at least one chemical that did not appear on the label, and 85 percent contained at least one unlabeled chemical with known or suspected health hazards.
Researchers tested 113 products across 12 categories, including sunscreens, baby lotions, shampoos, soaps, and household cleaners, using non-targeted chemical analysis. On average, each product contained 51 detected chemicals, more than three times the average of 16 ingredients listed on its label. The study was published in the journal Environment and Health by a team from the Million Marker Research Institute in Berkeley working with researchers at the Southwest Research Institute in San Antonio.
The finding that undercuts the most common consumer strategy is this: products marketed as natural, non-toxic, organic, or free of specific chemicals frequently contained the very substances the labels claimed to exclude.
The Contradictions Between Label and Contents
About 26 percent of the products contained chemicals that directly contradicted claims on the label or the company website.
Researchers found phthalates in products marketed as phthalate-free, synthetic chemicals in products marketed as 100 percent natural, and known allergens in products labeled hypoallergenic. More than half the products, 57.5 percent, failed the researchers' testing standard due to undisclosed contaminants.
The most striking result may be what happened to the products that looked best on paper. Before any testing, the team scored all 113 items on their ingredient lists alone and sorted them into three ranks. Twenty-seven earned the top rank. After testing, 59 percent of those failed anyway. Failure rates for large brands, regional brands, and small direct-to-consumer brands all landed between 50 and 60 percent, meaning neither a higher price nor a smaller company predicted a cleaner result.
"The label can't protect you when it doesn't show what's really in the bottle," said Jenna Hua, senior author and founder of the Million Marker Research Institute, in the announcement accompanying the study.
The Meaning of Undisclosed in This Context
Undisclosed does not automatically mean deliberately concealed, and the distinction matters for interpreting the result.
The authors describe the unlabeled compounds as falling into several groups: chemicals umbrellaed under the terms fragrance or parfum, chemicals occurring naturally in botanical ingredients, and synthetic contaminants. Contamination during manufacturing, migration from packaging, and breakdown of listed ingredients all produce compounds a formulator may not have added on purpose. Fragrance is a separate matter, since fragrance blends are treated as proprietary and can encompass dozens of individual chemicals under a single word.
The specific substances identified include endocrine-disrupting compounds such as phthalates and parabens, fragrance chemicals, silicone contaminants, and butylated hydroxytoluene, a common preservative. The authors note that endocrine-disrupting chemicals as a group have been linked to outcomes including cancer, metabolic disease, and infertility.
The evidence limits belong here rather than at the end. This study detected chemicals in products. It did not measure how much of any chemical entered a person's body, did not measure health outcomes in anyone, and cannot establish that using these specific products causes disease. Detection is not dose, and dose is what determines risk.
One technical point is often garbled in coverage. The analysis detected 3,141 features, meaning chromatographic peaks that include both identified and unidentified compounds, rather than 3,141 confirmed distinct chemicals.
The Regulatory Gap the Authors Are Pointing At
The authors' central policy argument is that American oversight depends too heavily on ingredient disclosure and manufacturer self-reporting.
The Modernization of Cosmetics Regulation Act of 2022 expanded the Food and Drug Administration's authority over cosmetics, including requirements for facility registration and product listing. What it did not do, as coverage of the study noted, is generally require routine testing of finished, packaged products for undisclosed chemical contaminants.
That is the gap. A company can fully comply with labeling requirements while the bottle on the shelf contains unlisted compounds, because nobody is required to test the bottle.
One practical limitation for shoppers: the specific brands and products tested have not been released, and the authors have said they will appear in an upcoming report. Until then, no one can act on this study by avoiding a particular item, which limits its immediate usefulness at the drugstore, even as it strengthens the case for policy change.
The products were purchased from major U.S. retailers in 2021 and 2022, so their formulations may have changed since then. That gap does not undercut the regulatory argument, since the testing requirement the authors propose would apply to whatever is on shelves today.
Proportionate Steps Without Overhauling the Bathroom Cabinet
Nobody needs to throw out their shampoo. The reasonable response is to reduce overall exposure where it is easy, not to pursue a chemical-free household that does not exist.
Fewer products mean fewer exposures. A person using four products daily encounters less than a person using twelve, and that arithmetic holds regardless of what any label claims. Fragrance-free versions genuinely reduce one category of exposure, though the study found that even those claims are not always accurate, as Gizmodo's account of the failure rates makes clear.
Ventilation matters more for cleaning products than for label reading. Opening a window while cleaning and avoiding aerosol sprays in enclosed spaces directly reduces inhalation exposure.
Pregnant people, infants, and young children are the groups where caution is most warranted, because endocrine-disrupting compounds are of greatest concern during developmental windows. Reducing the number of scented products used on a baby is a low-cost adjustment.
Anyone with a specific concern about a chemical exposure and a health condition should raise it with their clinician rather than a product blog. And HealthDay's coverage captures the study's honest bottom line: a careful shopper reading ingredient lists can still end up exposed to the chemicals they were trying to avoid, which is a case for changing the regulatory system rather than shopping harder.
Key Questions Answered
What did the study test? Researchers analyzed 113 personal care and cleaning products across 12 categories using nontargeted chemical analysis, comparing detected chemicals against label ingredient lists.
What does undisclosed mean here? A chemical detected in the product that does not appear on the label. It can arrive through manufacturing contamination, packaging migration, ingredient breakdown, naturally occurring botanical constituents or proprietary fragrance blends, not necessarily through deliberate concealment.
Does this mean these products caused harm? No. The study detected chemicals in products. It did not measure how much entered anyone's body, nor did it measure health outcomes. Detection alone does not establish risk.
Were natural and non-toxic products cleaner? No. Failure rates were similar across large, regional, and small brands, and 26 percent of products contained chemicals contradicting their own label claims.
Which brands were tested? The specific brands and items have not been released. The authors say they will appear in an upcoming report.
Why does existing regulation not catch this? The 2022 cosmetics law expanded FDA authority over cosmetics but generally does not require routine testing of finished packaged products for undisclosed contaminants.
What can I reasonably do now? Use fewer products overall, choose fragrance-free where possible, ventilate when cleaning and avoid aerosols in closed spaces, and be most cautious with products used on infants and during pregnancy.
