Skip to content
PurityIQAll posts

Pesticides/Residues

USDA Tested Nearly 10,000 Foods for Pesticides Last Year. Two True Numbers Tell Two Different Stories.

Published August 21, 2026

Every year, the USDA quietly buys thousands of fruits, vegetables, and other foods from the same distribution centers that supply your grocery store, washes them the way a shopper would, and tests them for hundreds of pesticide compounds at some of the lowest detection levels a lab can achieve. The results are published. Almost nobody reads them. And the headline number and the number two paragraphs down tell two different, equally true stories.

The most recent complete report, the Pesticide Data Program's Annual Summary for calendar year 2023, published November 7, 2024, tested 9,832 samples. The USDA's own stated topline: "over 99 percent of the samples tested had residues below the tolerances established by the EPA," and 38.8 percent had no detectable pesticide residue at all. That's the number that shows up in most coverage of this report, and it's accurate.

Here's the number that usually doesn't make the headline, from the same government report: only 38.8 percent had zero residue. The rest split further, with the USDA's own count showing the remainder carrying either a single detected pesticide or, in a substantial share of samples, more than one. Both facts, the overwhelming compliance rate and the frequency of multiple simultaneous residues, sit in the same document, and treating either one alone as the whole picture is where most coverage of this program goes wrong in one direction or the other.

The violations that did occur have a clear pattern. Residues exceeding the legal tolerance turned up in 0.49 percent of samples, 48 out of 9,832. Of those 48, the USDA's own breakdown shows 7 were domestic, 39 imported, and 2 of unknown origin — meaning imported produce accounted for the large majority of over-tolerance findings that year. Separately, 198 samples, 2.0 percent, carried a residue for which no U.S. tolerance has been established for that specific crop at all; again, the majority of those were imported. The USDA reports every one of these findings to the FDA and EPA, because the testing program itself has no enforcement authority — it measures, the FDA and EPA police.

The most current data available supersedes even that report. The 2024 Annual Summary, published January 6, 2026, tested 9,872 samples across 19 commodities of fresh and processed fruits, vegetables, nuts, and fish, and repeated the same topline: more than 99 percent below EPA benchmarks. Notably, 2024 marked the return of salmon to the testing program after roughly a decade's absence, a meaningful expansion of what the government is actually checking, not just how much.

One category of testing in this program deserves its own explanation, because it's the part most people have never heard of. Beyond active-use pesticides, USDA's labs also screen for what the program calls environmental contaminants: pesticides whose U.S. registrations were cancelled decades ago, including DDT, but which persist in soil and can still be taken up by crops grown in that soil today. This is a real, ongoing part of the testing program, run every year, on multiple commodities. We're not going to hand you a specific commodity-level percentage for this category in this article: the exact appendix tables containing that commodity-by-commodity breakdown weren't something we could independently confirm against USDA's own primary document with the sourcing tools available to us for this piece, and PurityIQ's standard is that a specific number goes in an article only when we can verify it ourselves against the primary source, not when a secondary source tells us what the primary source supposedly says. If you want to check a specific commodity yourself, USDA's own searchable database is public.

Worth flagging on data currency: strawberries, one of the most frequently cited items in pesticide-residue conversations online, were not tested by this program between 2016 and 2025 — a roughly nine-year gap. Testing resumed in 2025, but a 2025 annual summary won't publish until roughly early 2027. Any strawberry-specific statistic you've seen circulating traces back to that older 2014-2016 testing cycle, not anything current.

About 9.2 percent of the 2023 samples carried an organic label, and the program's public database lets anyone filter results by that claim directly rather than take anyone's word for what organic testing shows.

What can you do with a report like this? Less of a shopping trick than you might want, and more of a genuinely useful habit: wash your produce, since every sample in this program was rinsed the same way before testing, meaning the residue figures already reflect what survives a normal wash, not what's on an unwashed item. If import-versus-domestic sourcing matters to you, the over-tolerance data gives a specific, documented reason to pay attention to country-of-origin labeling rather than a vague unease about produce generally.

This is what PurityIQ is built to put in front of you at the shelf: the government's own residue counts, by commodity and by year, cited to the primary report, both the reassuring number and the one that complicates it, printed in the same place. No score picking one number over the other for you.

Sources:

As of August 21, 2026.

Sources

PurityIQ puts this same public record in front of you at the shelf. Cited to source. No score. Join the early-access list.