CDC and FDA Count 1,947 Cyclospora Cases Tied to Taco Bell Lettuce; Mexico Disputes the Source

1,947 people sick, and no lab has found the parasite on the lettuce

By July 26, 2026, the CDC and FDA counted 1,947 people sick across nine states: Illinois, Indiana, Kansas, Kentucky, Michigan, Ohio, Oklahoma, Pennsylvania and West Virginia^[1][4]. People started getting sick as early as June 22 and as recently as July 20^[3]. Federal investigators say the cause is shredded iceberg lettuce, grown in central Mexico by Taylor Farms de Mexico and served at some Taco Bell restaurants^[1][2][8].

The parasite is Cyclospora, and the illness it causes is not a bad afternoon. It brings weeks of watery diarrhea that can return in waves, and clearing it usually takes a specific antibiotic. By the time the outbreak count reached 1,644, about 94 people had been hospitalized^[13][26]. Across all cyclosporiasis cases reported nationally this year, not the Taco Bell outbreak alone, the CDC has logged 308 hospitalizations in 41 states and no deaths^[11].

Taylor Farms pulled all of its central-Mexico iceberg lettuce from the U.S. market on July 17, one day after the CDC and FDA named it^[2][7]. The recall reached grocery shelves too: Marketside-brand iceberg salad and shredded lettuce sold at Walmart, with best-by dates from July 18 through August 3, 2026^[9].

Here is the fact that has driven nearly everything written about this outbreak since: no laboratory test has ever confirmed Cyclospora on the lettuce. The FDA reported one positive sample on July 18. A day later, it re-reviewed that result and called it a false positive^[2][5][6]. Mexico's government ran its own tests at the Taylor Farms plant and got negative results, too^[12]. The federal case rests entirely on other kinds of evidence, and understanding what that evidence is, and why it can be strong even without a positive test, is the key to the whole dispute.

Why the parasite points investigators to a field, not a kitchen

Cyclospora is not a bacterium, and that difference shapes the entire investigation. When the parasite leaves a sick person's body, it cannot infect anyone else right away. It has to spend days or weeks maturing in soil or water first^[22][23].

That means the illness cannot spread from person to person, and it cannot spread from a cook to a customer at the counter. Contamination has to happen far upstream, almost always in a field or in the water used to grow or rinse the crop. That single fact is why investigators went looking at a farm in Mexico instead of at Taco Bell kitchens, and it is a fact both the U.S. and Mexican governments agree on.

It also explains why federal officials feel confident naming a source without a positive lab test. They rely on two other tools instead. One is patient interviews: when hundreds of sick people in different states are asked what they ate, and Taco Bell comes up far more often than chance would allow, that pattern counts as evidence. The other is called traceback, meaning investigators follow purchase and shipping records backward from each restaurant, through distributors, to specific farms and lots. When many separate chains of evidence all lead to the same grower, investigators call that convergence, and the FDA says its data "continue to converge on shredded iceberg lettuce from Taylor Farms locations in central Mexico^[2]."

The test that came back positive, then didn't

The FDA argues a negative product test proves less than it seems to here, and its reasoning is about timing. Lettuce spoils in days, not months, and the contaminated heads were eaten weeks before anyone knew to look for them^[2][5]. The parasite also tends to sit in scattered patches on a leaf at very low levels, so one clean sample from one bag doesn't rule out contamination elsewhere in the same shipment^[2][5].

That is the backdrop for the July 18 positive test. When the FDA re-reviewed it and called it a false positive on July 19 and 20, the agency insisted the walk-back changed nothing about its conclusion. It said the reversal "does not change the basis for FDA's ongoing outbreak investigation or the overwhelming epidemiological data supporting the current voluntary recall^[2][5]." Health and Human Services Secretary Robert F. Kennedy Jr. put it more bluntly, saying the outbreak is "under control" and that the government has "done extensive forensics, epidemiological forensics, and we've identified the source^[10]."

Mexico's health ministry sees the same set of facts and draws a very different conclusion. After inspecting the Taylor Farms plant in central Mexico, its tests of lettuce and water samples came back negative for Cyclospora^[12]. Water matters especially here, since irrigation water is the most common way the parasite reaches a crop in the first place. Health Minister David Kershenobich said there is no evidence yet that Mexican lettuce caused the U.S. outbreak^[12].

Mexico's regulators, Cofepris and Senasica, then made a narrower and harder-to-dismiss point: tracing a product back to a country tells you where it came from, not where it got contaminated^[12][24]. A truck, a distribution warehouse, or a restaurant somewhere in the supply chain could, in theory, be the actual source. The agencies call their inspections "strictly preventive" and say the investigation is still open^[12].

A supply chain that both sides want to protect

Taylor Farms and Taco Bell are making their own case, built on timing rather than lab results. The company pulled its central-Mexico lettuce the day after being named, before any test had come back positive, and while the only positive result on record was later withdrawn^[2][7]. In its own recall statement, Taylor Farms said the FDA traceback pointed to one independent farm supplying less than 1% of the entire U.S. iceberg lettuce supply, not to its broader operations^[7]. That is the company's own characterization of the scope of the problem, not an independently confirmed figure.

Taco Bell's position is that it bought a pre-washed, pre-shredded ingredient from a supplier, with no way to inspect it for a parasite invisible to the naked eye. But the recall still landed hardest on its brand, since its name appears in most headlines rather than the supplier's.

The stakes for the companies are not just reputational. A Michigan couple, Preston and Marie Parrish, has sued Taco Bell and Taylor Farms, seeking more than $75,000^[21]. Their attorney argued that the absence of a positive lettuce test "doesn't change the fact that there's thousands of people in Michigan with Cyclospora after they ate at Taco Bell^[21]." More lawsuits are expected to follow.

Mexico has its own stake in how this resolves. A large share of Mexico's economy depends on exporting produce to the United States, and accepting blame here risks more than one farm's reputation; it risks Mexico's standing as a reliable supplier and invites more U.S. scrutiny of Mexican farms going forward.

A fight over budget cuts that predates this outbreak

A second dispute running alongside the science is about whether the outbreak should have been caught sooner. Consumer advocates, some former FDA officials, and congressional Democrats argue that cuts to CDC disease surveillance and FDA lab capacity slowed detection^{[17][18][19][20]}. They point to specific numbers: in 2024, the FDA had 443 food-safety inspectors for every facility under its authority, against the roughly 1,500 the agency itself estimates it needs^[17]. In March 2025, HHS announced a plan to cut about 3,500 FDA jobs, roughly 20% of the agency's workforce^[17][20].

A former director of the FDA's food safety center made a related argument in a signed opinion piece: an administration that urges Americans to eat more fresh, unprocessed food is, at the same time, cutting the monitoring that keeps fresh food safe^[19]. Consumer Reports has made a more technical version of the same point, arguing that routine produce testing looks for bacteria, not parasites like Cyclospora, so this kind of contamination can slip through even functioning inspection systems^[18].

Secretary Kennedy's position is the opposite: that the outbreak is handled and the source is known^[10]. Whether this argument ultimately matters may come down to what happens next. If new cases keep appearing after the recall, the surveillance-cuts critique gains weight. If the outbreak winds down quickly, the administration's "under control" framing will look vindicated.

Numbers that don't agree, because they're not counting the same thing

Different groups have cited very different case counts in this outbreak, and none of them are lying. The CDC's official outbreak number, 1,947, counts only people who tested positive for Cyclospora and reported eating at Taco Bell^[1][4]. Michigan's state health department, by contrast, has reported more than 5,000 cases, because it counts every lab-confirmed Cyclospora case in the state, whether or not the person ate at Taco Bell^[17][26]. National surveillance data adds a third layer: the CDC has been tracking four separate Cyclospora outbreaks in the U.S. this year, so some illnesses in public debate may trace to other food entirely^[17].

Coverage of the story split along familiar lines. Fox News framed it as a matter of personal precaution, running a doctors' Q&A on washing produce, with little discussion of inspection budgets or staffing^[13]. CNN and The Fulcrum led with the outbreak's scale and tied it directly to the surveillance-cut argument^[15][16][17]. Al Jazeera described the U.S. finding as something officials "claim" rather than something established, and placed Mexico's denial high in its story^[11]. The Packer, a trade publication for growers, ran the story mostly as Mexico's exoneration, leading with "no evidence" rather than with how many people are sick^[12].

What remains unresolved is the same thing that has been unresolved since July 18: no test has found the parasite on any lettuce sample, from either country, and every party in this dispute has reasons to want that fact to mean something different.