Data study · Cat food · Clean Label Project, 2026
What an independent laboratory found when the Clean Label Project tested 100 cat foods from 62 brands, set out from the figures in its published report.
99 of 100
cat foods in the study contained detectable lead, and cadmium and arsenic were found in nearly as many.
In early October 2026 news outlets reported on a Clean Label Project study of 100 top-selling cat foods, bought at retail by the Colorado-based nonprofit. News coverage centered on the detection rates for heavy metals, pesticides and plastic chemicals. This page sets out the figures the report itself publishes, so that each one can be read next to the others and traced to its source.
I did not run any of the tests. The Clean Label Project has not released results for individual products, so everything here comes from the summary figures in its report and on its study page, and every figure is listed with its source in the table at the end.
01 · Detection
Each tick below is one of the 100 products. Lead, cadmium and arsenic turned up in nearly all of them, the weedkiller glyphosate in a little over a third, and the plastic chemical DEHP in just under half. The ticks show how many products had each substance and not which ones, because the report does not identify them.
Mercury is left out of this chart because the report gives it as comparisons between protein sources and not as a detection rate. Acrylamide was measurable in 4 products and present at very low levels in 24 more. Dibutyl phthalate was seen in 26 products but never at a level the laboratory could quantify, and dihexyl phthalate was not found in any product.
02 · Comparisons
The report's main comparison is between ingredients and formats. Fish and seafood foods carried about 18 times the mercury of other foods, and among wet foods alone the gap was about 28 times. Dry food carried about three times the lead of wet or air-dried food.
Each dot is the multiple the report states, on a log scale where 1× means no difference. The report converts results to micrograms per 1,000 calories before comparing foods, and each row names its own baseline, so the rows cannot be added together or ranked as if they measured the same thing. The glyphosate row is an association the report found, rising by about 40% with each additional plant ingredient among the first ten listed, which is plotted here as 1.4×.
03 · The standard
The report scores each product against the other cat foods in the study, and it says no regulatory limits apply to compare against. By that measure 84.8% of poultry products met the standard and 25.7% of fish and seafood products did. Dry and wet formats were within one percentage point of each other.
Meeting the standard is a ranking within these 100 products, set by the Clean Label Project, and not a safety threshold. I could not find the number of products in the poultry and fish groups in the report, so the percentages are not converted to counts here.
04 · The sample
The 100 products were split across three formats. Wet food was the largest group.
05 · Reading it carefully
Detection means the laboratory measured the substance in the product. It does not mean the amount was harmful to a cat, and the report does not compare its results with regulatory limits because it says none apply, so the figures cannot show whether any product is unsafe.
The standard is the Clean Label Project's own and is set relative to the products in this study. Newsweek reported that Elise Fennig, chief executive of the Pet Food Institute, the industry trade group, objected to the use of a proprietary standard in place of recognized scientific and regulatory benchmarks, and questioned whether the detected amounts could harm cats. Newsweek also reported that the Clean Label Project named only three brands that meet its certification.
The sample is 100 products, each tested from one batch. It is large enough to show clear differences between fish and poultry formulas and between dry and wet formats, and too small to say much about any single brand.
06 · The data
The table lists each figure used on this page with the places it was confirmed. Figures that appear only in the report are marked that way, and the 28× wet-food multiple is on the study page and in the press release but not in the report's running text. The same table is available as a CSV file.
Source codes. R: the Clean Label Project report (PDF). W: the Clean Label Project study page. P: the Clean Label Project press release. C: CNN or Scripps News coverage. N: Newsweek.
Sources