The Lettuce Cyclospora Outbreak Is Over: 12,883 Cases, 570 Hospitalized, 21 States. Here Is What the Final Count Says, and What It Leaves Open
Eight weeks after the count stood at 400 cases in four states, the CDC has posted its final update on the iceberg lettuce outbreak: “This outbreak has ended.” The eight updates in one table, the national count that runs nearly 7,000 higher than the outbreak count, the 1996 record this one passed nine times over, and the root-cause investigation that did not close with it.
The CDC declared the multistate Cyclospora outbreak linked to Taylor Farms de Mexico iceberg lettuce over on September 11, 2026, its final update: 12,883 laboratory-confirmed illnesses in 21 states, at least 570 hospitalisations and two deaths, both in Michigan, with illness onsets from June 14 to August 17. The count rose from more than 400 (July 14) to 1,644 on recall day (July 17), 6,358 (August 5), 9,481, 10,930, 11,458 and 12,883 as reporting caught up with June and July infections; CDC says more than 1,000 infections occurred on a single day at the peak before the recall and fewer than two a day on average in August. Because reporting can take up to six weeks, cases will still arrive after the count closed, and CDC says the true number was likely higher. It is separate from the national total: FDA cites 19,595 lab-confirmed US cyclosporiasis cases in 2026 as of September 8, and CDC's cases page listed 19,883 (1,064 hospitalisations, 49 states and D.C.) by September 15, so roughly 7,000 confirmed cases this year belong to other clusters and seasonal background. CDC calls the season "unprecedented"; its archive describes the 1996 raspberry outbreak, about 1,400 cases, as the largest ever reported, and the 2013 salad-mix outbreak, also linked in part to Taylor Farms de Mexico, at 631. Still open: FDA "continues to investigate how the product ... became contaminated," Mexican grower inspections have ended with samples pending, no product sample has tested positive, a letter to industry was issued September 11, and Taylor Farms, which disputes responsibility, has made no statement on the close-out. Symptoms begin about a week after exposure (2 days to 2 weeks or more): watery diarrhea, appetite loss, weight loss, cramping, bloating, gas, nausea, fatigue; untreated it may last from a few days to a month or longer with relapses; trimethoprim-sulfamethoxazole is the treatment of choice; cooking produce to 158°F kills the parasite. Educational, not medical advice.
On Thursday the CDC posted the sentence it had been building toward for eight weeks: “This outbreak has ended.” The multistate Cyclospora outbreak linked to iceberg lettuce from Taylor Farms de Mexico closed at 12,883 laboratory-confirmed illnesses in 21 states, at least 570 hospitalisations and two deaths, both in Michigan. It is the agency’s final update. The recalled lettuce is past its dates and gone from shelves, and the investigation page says the daily rate of new infections fell from “more than 1,000” on the worst day before the recall to “fewer than 2” a day on average in August.
We have followed this since the first week, when the count was 400 and the states were four. A final number invites two mistakes: treating it as the whole story, and treating it as the whole count. It is neither. Here is the ledger, the arithmetic behind a second number you will see quoted alongside it, and the parts of the investigation that did not close on Thursday.
The ledgerEight updates, one line
| CDC update | Cases | States | Hospitalised | Deaths |
|---|---|---|---|---|
| July 14 | “more than 400” | 4 | — | — |
| July 17 (recall day) | 1,644 | 5 | 94 | 0 |
| July 24 | 1,947 | 9 | 98 | 0 |
| August 5 | 6,358 | 15 | 278 | 2 |
| August 13 | 9,481 | 17 | 398 | 2 |
| August 20 | 10,930 | 17 | 454 | 2 |
| August 27 | 11,458 | 20 | 495 | 2 |
| September 11 (final) | 12,883 | 21 | 570 | 2 |
The shape of that table is the story. The recall came on July 17, when the count was 1,644 and every case was tied to Taco Bell restaurants in five states. Most of the 11,000 cases added afterwards were people who had already been infected before the recall and were still working their way through testing, diagnosis and interview; the biggest single jump, on August 5, came when CDC widened the case definition from “ate at Taco Bell” to anyone exposed to the recalled lettuce wherever it was sold, including Walmart’s Marketside brand. Illness onsets ran from June 14 to August 17. Nothing after mid-August was new spread; it was the paperwork catching up with June and July.
CDC states the caveats itself, and they cut in the same direction. “The true number of sick people in this multistate outbreak was likely higher than the number reported,” because many people recover without being tested. And the agency is stopping its count, not the reporting: “Because it can take up to 6 weeks after symptoms start for a case to be reported, infections acquired during this outbreak will still be reported to CDC as they are identified.” It adds that, “due to the unprecedented number of cases that occurred this season,” earlier cases may keep arriving beyond even that window. So 12,883 is where the counting stopped. The outbreak itself stopped weeks earlier, and its final total will only ever be an estimate.
More than 1,000 infections in a single day before the recall. Fewer than two a day in August. The recall worked; the count kept climbing because reporting takes six weeks.
The recordCDC says “unprecedented.” Its own archive says why
No CDC page calls this the largest cyclosporiasis outbreak in American history; the agency’s word, on its cases page, is “unprecedented,” and the superlative belongs to CBS, CNN and the produce trade press. The agency’s own archive makes the comparison anyway. CDC’s journal described the 1996 raspberry outbreak as “the largest ever reported outbreak of cyclosporiasis, affecting more than 1,400 persons in North America.” The 2013 salad-mix outbreak reached 631 people in 25 states, and CDC’s record of it notes that restaurant illnesses in Iowa and Nebraska “were linked to salad mix from Taylor Farms de Mexico in Guanajuato, Mexico,” the same processor. The 2018 season, then considered severe, totalled 2,299 confirmed cases across all outbreaks combined; 2019, the worst prior year, recorded 4,703 for the whole year. This single outbreak is roughly nine times the 1996 record and nearly three times the worst full year on file.
The second numberNineteen thousand, and why it is not this outbreak
The same day CDC closed the outbreak, the FDA published a post-outbreak page with a bigger figure: “As of September 8, the CDC reported 19,595 laboratory confirmed cases of cyclosporiasis in the United States” this year, a total that “includes more than 12,000 cases from one large multi-state outbreak linked to iceberg lettuce.” The two numbers will be quoted next to each other for months, so it is worth being precise: 12,883 is one outbreak; 19,595 is every lab-confirmed case of the disease in the country in 2026, from every source. Subtract one from the other and roughly 6,700 confirmed cases this year belong to something other than this lettuce, and the gap is still widening: by September 15, CDC’s national cases page had 19,883 laboratory-confirmed cases acquired in the United States since May 1, with 1,064 hospitalisations and two deaths across 49 states and the District of Columbia, and a note that weekly updates are ending. Some of the difference is the other clusters FDA says it is still investigating; some is the ordinary summer background of a parasite that arrives on imported produce every year. Neither agency breaks it down, and neither can we.
That distinction was already blurry in July, when national seasonal totals and outbreak totals were being used interchangeably in coverage, and it matters more now that one of them has a full stop after it. The outbreak is over. Cyclosporiasis season, which CDC describes as running through the late spring and summer, winds down on its own calendar.
What did not closeThe cause, the company, and the growers
“Ended” is an epidemiological word. It means no one is still getting sick from this lettuce. It does not mean anyone knows how the lettuce got contaminated. The FDA says plainly that it “continues to investigate how the product linked to this multistate outbreak became contaminated with Cyclospora,” and its post-outbreak page lists what that now involves: onsite inspections and sample collection at the iceberg growers and the processing facility in Mexico, which began on August 20 and, per the investigation page, “have ended,” with samples “pending analysis”; a meeting the week of September 8 between senior FDA and Mexican public health and agricultural officials on what the agency calls “a prevention focused framework”; and a letter to industry, signed by the acting deputy commissioner for food, addressed “to every link in the supply chain” and calling recent outbreaks “a serious public health concern and a reminder of our shared responsibility to U.S. consumers.” Findings, the agency says, will go into an updated Cyclospora prevention and research action plan “in the coming months.” One fact has held all summer: the FDA says “there are no confirmed positive sample results for product testing for Cyclospora,” the single July positive having been ruled a false positive.
The company’s position has not changed either. In August, Taylor Farms told Food Safety News it did not accept responsibility, that it had cooperated fully, and that its own testing found no parasite; its information hub was last updated with the August 27 no-positive-samples line and, as of this writing, carries no statement on the close-out. CDC and FDA, in their final language, still describe iceberg lettuce from its central Mexico processor as the product that “was contaminated with Cyclospora and made people sick,” a conclusion drawn from epidemiology and traceback, and, since late August, from genotyping showing the parasites in patients were genetically related to one another. The two positions coexist in the record, as they did in July. What the close-out does not do is settle them; that, if it happens, is the root-cause report’s job.
Know the signsIf you were exposed in the summer
With onsets ending in mid-August, anyone who was going to get sick from this lettuce almost certainly already has. The reason the illness is still worth describing is that it is unusually slow, and some of the people counted in September were infected in July. Per the CDC, symptoms begin about a week after exposure, with a range from two days to two weeks or more, and the hallmark is watery diarrhea, often frequent and urgent, alongside loss of appetite, weight loss, cramping, bloating, increased gas, nausea and fatigue. Untreated, per CDC’s clinical page, “the illness may last anywhere from a few days to a month or longer,” and “symptoms may seem to go away and then return one or more times.” We have set out the timeline in detail; the short version is that a stomach bug that has lasted well past a week, especially with a lettuce history over the summer, is the kind a clinician can test for specifically, because the same guidance names a specific antibiotic, trimethoprim-sulfamethoxazole, as the treatment of choice, and the parasite is not looked for on a standard stool culture.
For the future, CDC’s own advice on the closed page is unglamorous and exact: wash hands and fresh produce under running water; and because “washing alone cannot guarantee Cyclospora removal,” cooking produce “to a temperature of at least 158 °F (70 °C)” kills the parasite. We have covered what a rinse can and cannot do, and why greens in particular keep turning up in these stories.
“Ended” means nobody is still getting sick from this lettuce. It does not mean anyone yet knows how the lettuce got contaminated.
The summer in outbreaksWhere this one sits
This was the largest of a crowded season, not the only entry. CDC closed the 134-case Midwest Poultry egg outbreak on September 3, a week before two smaller egg outbreaks surfaced in California and Vermont; two unrelated sprout outbreaks, alfalfa and broccoli, remain open across 19 states between them; and a frozen-berry E. coli investigation widened its recall to 26 states on September 11. Against that list the lettuce outbreak is the one with an ending, a number, and a still-open question about cause. We will report the root-cause findings when the FDA publishes them. This article is educational, describes what the cited agency pages say, and is not medical advice.