The UK Health Security Agency published new figures on July 30. Sixty-seven people in England, Scotland and Wales were reported with cyclosporiasis between April 30 and July 15 of this year — thirty in England, twenty-seven in Scotland, ten in Wales. The comparison UKHSA itself draws is the one that matters. The annual average for the four years from 2022 through 2025 was ninety-three cases for an entire year. Britain is two-thirds of the way to a normal year by the middle of July.
Travel information was available for fifty-one of the sixty-seven. Forty-eight of those fifty-one had been to Mexico. One of them had also been to the United States, one had been to the United States only, and one to Kenya. The people who went to Mexico stayed at a range of different hotels in the Riviera Maya and Cancún regions and ate and drank as part of all-inclusive holiday packages. Read that last sentence slowly, because it is the whole finding. Different hotels. One region. That is a supply chain.
And then the line that ought to be getting more attention on this side of the Atlantic than it is. UKHSA says it anticipates a potential increase in cases linked to travel to the USA, where a widespread outbreak has been reported. A foreign government is telling its citizens that American travel now carries a parasite risk, and it is the second government to do so inside a week. The Windsor-Essex County Health Unit in Ontario has identified twenty-four cyclosporiasis cases in 2026 against a typical year of about three, has identified travel to Michigan as a common exposure among many recent cases, and is advising residents to take precautions when they cross.
Here is why the British data is worth more to this argument than another American press release. It comes from an independent surveillance system in another country, with no exposure to our supply chain and no stake in our fight. It is not counting the same people we are counting. And it has landed on the same country of origin that FDA’s traceback landed on. I am not going to overstate that — UKHSA has not named a farm, a shipper or a product, and Guanajuato is not Quintana Roo. But when British holidaymakers in one part of Mexico and American diners in Michigan are acquiring the same parasite in the same eleven weeks, the question of whether Mexican produce is the through-line stops being a question about who holds a positive sample.
Now the federal numbers, and a figure I have not put in front of you before. CDC’s surveillance page, last updated July 28 with data through July 27, reports 6,707 laboratory-confirmed domestically acquired cases across forty-five states, more than 11,500 additional cases still awaiting confirmation, and 423 hospitalizations. Four hundred and twenty-three.
FDA’s nine-state advisory says ninety-eight hospitalizations. It has said ninety-eight since July 24. Total Illnesses 1,947. Hospitalizations 98. Deaths 0.
That last line is now simply wrong. CDC’s own outbreak page carries an August 3 entry stating that it is aware of two cyclosporiasis-related deaths from Michigan as part of this outbreak. Michigan’s page now reads Deaths: 2. FDA’s page, on the same investigation, run by the same government, published to the same public, still reads zero.
How do the federal hospitalization and death numbers compare to what the states themselves are saying? I went and added the states up. Thirteen states publish a hospitalization figure of their own, and they come to 426. The CDC is at 423 with the Taco Bell portion at 98.
People keep calling this the Cyclospora outbreak, singular, and it is not one thing. The honest way to separate them is to describe what each investigation has actually established, rather than what it is suspected of. There are four, and then there is everything else.
One. The nine-state Taco Bell cluster, FDA reference number 1390. This is 1,947 people in Illinois, Indiana, Kansas, Kentucky, Michigan, Ohio, Oklahoma, Pennsylvania and West Virginia who were infected with Cyclospora and reported exposure to Taco Bell, with onsets running June 22 to July 20 and at least ninety-eight hospitalizations. It has a named supplier; Taylor Farms de Mexico. It has a named product; shredded iceberg lettuce sourced from central Mexico. It has the only recall of the entire 2026 season. Michigan interviewed 190 of the Taco Bell patients and ninety percent of them had eaten iceberg lettuce. The critical thing to understand about this cluster is that documented Taco Bell exposure is the entry ticket. That is why Michigan’s 11,234 produces a federal figure of 1,947. It is a description of who got interviewed in a particular way, not of who got sick.
Two. New Hampshire. The state has thirty-three confirmed cases as of July 28 with two hospitalizations, in a place that identifies between two and nine cases in a typical year. Seventeen of the thirty-three are people who ate lettuce at a location that served recalled Taylor Farms product, and all seventeen trace to one location. New Hampshire DHHS publishes the address: Cheshire Medical Center, 580 Court Street, Keene. It committed in writing to posting additional locations as it identifies them. Spokesman Jake Leon says the department is treating it as a foodborne cluster investigation and believes the cluster is part of the larger nationwide outbreak. What makes New Hampshire different from the nine is not the product, which is the same recalled lettuce. It is the establishment, which is not Taco Bell. Twenty-seven other states received that recalled product. New Hampshire is the only one that has told its residents where it was served.
Three. Alabama. Twenty-eight cases as of July 20, two hospitalized, four travel-related. Eight of the twenty-eight — more than a quarter of the state’s total — are tied to two separate foodborne outbreaks at Mexican-style restaurants, and ADPH says in its own release that these are not Taco Bell. Alabama has named a category of establishment. It has not named a product, and it has not connected any of its cases to Taylor Farms, even though Alabama is on the recalled-lettuce distribution list. This is the cleanest published example we have of a state finding restaurant-associated Cyclospora that is not the federal outbreak.
Four. North Carolina, which is the biggest of these and the least resolved. NCDHHS reports 718 cases since May 1 and twenty-four hospitalizations; its dashboard puts the calendar-year figure at 728 against 306 for all of 2025, with 432 of the cases in Wake County. The state says the increase does not appear to be associated with the Taco Bell outbreak. What it has named is a commodity, not a company: patient interviews show parsley and cilantro reported more often than population-wide survey data would predict. It has no supplier and no establishment. And the department is careful to add that many cases reported eating neither, which it says suggests more than one source.
There is a detail in FDA’s own records that belongs next to North Carolina. On the closed-investigations portion of the CORE outbreak table sits reference number 1325, posted August 2025: Cyclospora cayetanensis, parsley, sixty-nine cases, investigation closed. The federal government named parsley as a Cyclospora vehicle twelve months ago. North Carolina is reporting parsley now, at 718 cases and climbing, and there is no open federal investigation with parsley’s name on it.
And then everything else, which is the part almost nobody writes about. As of its July 29 update, FDA’s CORE table carries six Cyclospora entries. One is the lettuce outbreak. The other five say Not Yet Identified in the product column: ninety-three cases under reference 1392, ten under 1385, eighteen under 1384, two under 1381 and eight under 1375. Three of those five are marked ended with the investigation still open. That is 131 people across five files where the federal government opened an investigation, ran a traceback and named nothing.
Add it up. 1,947 plus 131 is 2,078. That is every Cyclospora illness this season that the federal outbreak-response system has attached to a numbered investigation. The fifty states are reporting 22,683. The investigative apparatus has a file open on roughly nine percent of the illness in this country and has named a food in one file out of six.
Missouri sits outside all four categories and is the reason I keep writing about it. It has 1,095 cases as of August 2. It is on FDA’s distribution list. St. Louis County says people in its area who got sick reported eating at Taco Bell. The state health department says it expects to be added to the CDC outbreak map soon. And a CDC spokesperson has told a Kansas City television station that cases in the remaining thirty-two states, Missouri included, are unrelated and are being investigated as separate clusters driven by different foods. That is not a fifth category. That is a disagreement between a state and a federal agency about which outbreak a thousand people belong to, being conducted in public, with nobody showing their work.
As to the count itself: I have been through all fifty states again this morning and the total holds at 22,683.