Unbranded glass and plastic food-storage containers held beside everyday kitchen items.

Did a Report Find Hazardous Chemicals in 93% of Household Items?

No. The underlying analysis classified chemicals in a database, not products in people's homes. It found that 685 of 738 chemicals with enough toxicity and exposure information met the Environmental Justice Foundation's broad high-risk screen. Most of the 3,552 chemicals selected for analysis could not enter that calculation because information was missing.

In the news
A September 30 CNN article ran under the headline “Over 90% of common household items contain hazardous plastic chemicals.” The percentage came from a new Environmental Justice Foundation report, but the headline changed the unit being measured.

Quick answer

The report did not test cookware, storage containers, utensils, toys or other household products. It analyzed chemical records drawn from the PlastChem database.

The 92.8% figure describes 685 of 738 chemicals that had both toxicity and exposure-category information. The full selected group contained 3,552 chemicals, and 2,814 lacked one or both types of information.

That does not make the chemical concerns disappear. It means the result cannot tell us that 92.8% of plastic products are hazardous or that using a product gives someone a 92.8% chance of harm.

Three things to know

  • The unit was chemicals, not products. No representative sample of household items was purchased or tested.
  • Most records were excluded from the 92.8% calculation. Missing information is not proof that those chemicals are safe or harmful.
  • “High risk” was a screening classification. It combined toxicity indicators with evidence of release, presence or marketed use. It did not measure a person's dose or health outcome.

What did the report actually analyze?

The Environmental Justice Foundation, or EJF, published Poison Plastics as an advocacy report. It reanalyzed information in PlastChem, a database of chemicals that may be used in or present in plastics.

The broader PlastChem project compiled 16,325 chemicals and was described in a peer-reviewed 2025 Nature paper. EJF narrowed that database to 3,552 chemicals assigned to five sectors: food-contact plastics, textiles, toys, household items, and hygiene and medical items.

Those are database categories, not counts of products on store shelves. A chemical can appear in more than one category, so the sector totals cannot be added together.

Where did 92.8% come from?

Here is the denominator behind the number:

  • 3,552 chemicals were assigned to the five selected sectors.
  • 2,814, or 79.2%, lacked toxicity information, exposure information or both.
  • 738 had both kinds of information.
  • 685 of those 738, or 92.8%, met EJF's high-risk rule.
  • 53 of 738 did not meet that rule or remained under review.

The 685 flagged chemicals are about 19.3% of the full 3,552-chemical group. That is not the report's estimate of how many chemicals are hazardous. Most of the full group could not be classified, so the data do not support calling the remainder safe.

Better-studied chemicals may be more likely to have both exposure information and documented hazards, so the assessed subset may not represent the missing records.

What did EJF mean by “high risk”?

EJF's rule required two things: high or suspected toxicity on at least one selected endpoint, plus high or potential exposure.

The toxicity screen covered five endpoints, including carcinogenicity, reproductive toxicity and endocrine disruption. Documented release from plastic counted as high exposure risk. Detection in plastic or marketed use counted as potential exposure.

That is a precautionary screen, not a conventional product-specific risk assessment. Presence or marketed use does not establish how much reaches a person. The analysis did not measure concentration in finished products, migration during ordinary kitchen use, absorbed dose, frequency of contact or disease outcomes.

What about household and food-contact products?

The household-items category contained 1,197 chemicals. Only 410 had enough toxicity and exposure information for EJF's combined screen, and 379 of those 410 met its high-risk rule. That is the source of the household figure of 92.4%.

For food-contact plastics, 456 of 498 chemicals with both kinds of information met the rule. The full food-contact category contained 2,109 chemicals, with 1,611 missing toxicity information, exposure information or both.

These results point to real questions and large information gaps. They do not show how many actual products contain a flagged chemical, how much might migrate or whether ordinary use produces a harmful dose.

How strong is the evidence base?

The EJF report is not peer reviewed. It is a secondary analysis from an environmental advocacy organization supported by the Paul Angell Foundation. The underlying PlastChem work was funded by the Norwegian Environment Agency, and its 2025 Nature paper was peer reviewed.

PlastChem's FAQ says some duplicates may remain and that the team did not manually decide whether every imported entry made sense as a plastic chemical. A 2026 Environmental Science & Technology analysis argued that about 30% of its Red List entries appeared questionable.

That criticism does not directly recalculate EJF's 685 chemicals or prove its result wrong. It is a reason to treat the database as a broad screening inventory rather than a finished census of product safety.

What does this mean in the kitchen?

This report does not rank materials or tell you which exact product to buy. It also concerns chemicals associated with plastics, which is a different evidence question from the release or ingestion of microplastic particles.

A practical approach is to prioritize recurring food and drink contact involving heat, wear, long storage and frequent use. Choosing a suitable lower-plastic option can remove one contact point, but it does not guarantee lower body burden or better health.

Our guide to what to replace first in a lower-plastic kitchen can help you choose where to begin without trying to replace everything at once.

PFK assessment: CNN's headline overstates the finding by turning a percentage of assessable chemical records into a percentage of household items. The corrected result still matters: many better-studied plastic-associated chemicals raise legitimate concerns, while most chemicals in the selected sectors lacked enough information for EJF's combined screen.

How PFK evaluated this headline

We traced the percentage from CNN to EJF's methodology and checked its numerator, eligible denominator, missing-data exclusions and unit of analysis. We also reviewed the underlying peer-reviewed PlastChem work, its limitations and current criticism.

Sources

Research reviewed September 30, 2026. This article explains a database-based chemical screening analysis. It is not a product-safety determination or medical advice.

Back to blog