Through the summer, one explanation competed with the lettuce for the largest Cyclospora outbreak in American history: that the parasite came out of American water — municipal sewage overflows around Detroit and Toledo, private wells, and Lake Erie. Below is every public instance of that argument I can document, what each one claimed, what each one cited, and the answer to each. Every source is linked. Check all of it.
A word on who IFPA is, because the name is newer than the organization.
The International Fresh Produce Association began operations on January 1, 2022, formed by the merger of the two largest produce trade groups in the country, the United Fresh Produce Association and the Produce Marketing Association. It is the largest association serving the fresh produce and floral supply chain, with offices in Washington and Newark, Delaware, and it speaks for growers, shippers, processors, retailers and foodservice operators. Max Teplitski is its chief science officer.
One disclosure first. My firm represents people sickened in this outbreak and we have filed cases naming Taylor Farms entities. Everything below is public either way.
July 15 — the statement that started it
The original IFPA statement, attributable to Chief Science Officer Max Teplitski, is still posted beneath the July 19 revision. Its central claim is that a statement released by the Michigan Department of Health and Human Services defies reputable science in saying recreational waters are not a recognized risk factor. Three footnoted citations carry that charge, with a fourth to a federal advisory report. Around it runs the argument that everything pointing at produce rests on patient recollection, that no contaminated product sample has been found, and that naming a product without one destroys consumer trust and puts thousands of jobs in jeopardy.
July 16 — the interview version, with the Flint line
Teplitski expanded it in The Packer. He said it was puzzling that state public health officials still rebuilding trust with consumers after Flint would rush to dismiss water as the source. He said the first documented U.S. Cyclospora outbreak, in 1990, was traced to contaminated water, and that a 2023 USDA and FDA advisory report lists recreational water among the water types where the parasite has been detected, including chlorinated water. He added that if Michigan officials are outright refusing to test water, they will never have a positive sample.
July 19 — the claim comes out of the statement
The updated statement, timestamped 8:15 p.m. ET on July 19 at the same address, contains no reference to recreational water, to Michigan, or to the three citations. The association preserved the original beneath it, which is more than most trade groups do and should be said plainly. The argument did not stop. It moved.
August 3 and 4 — the claim is filed with CDC
IFPA submitted formal comments on CDC’s proposed extension of the Cyclosporiasis National Hypothesis Generating Questionnaire and announced them on August 4. The filing recommended structured questions on drinking water sources, recreational water activities, wastewater exposures and septic systems, asked CDC to revisit unresolved historical investigations, and urged evaluation of wastewater surveillance. Teplitski framed it as a need to “expand our thinking about this parasite and seriously consider other routes of exposure.” The nine attached questions asked about main drinking water source, water drunk away from home, awareness of flooding or sewer overflow or a boil-water advisory, septic systems and last service date, occupational contact with sewage treatment, stormwater, portable toilets and biosolids, camping on private water, and recreational water contact — with a checklist running pool, hot tub, splash pad, lake, river, pond, irrigation canal or ditch. The Packer and the rest of the trade press carried it.
August 5 onward — the geographic version, on the consumer page
The claim is stated most plainly on IFPA’s consumer FAQ, which was last updated September 4, 2026 and is still live. Under what else could be a source of infection, it says the association continues to raise concerns about recent municipal sewage overflows in the greater Detroit and Toledo region and the likelihood of those events affecting drinking water, especially wells, and recreational water in Lake Erie and surrounding bodies of water, and that Cyclospora is fundamentally a waterborne parasite. A later answer cites documented rainfall and combined sewer overflow activity in parts of the affected region in mid-June. An August 6 member alert carried the same recommendations to the membership.
August 19 — a scientist quoted, and then a correction
Trevor Suslow, a food safety consultant and emeritus professor at the University of California Cooperative Extension, was quoted in Ag Alert on how hard this parasite is to trace. As first published, the article attributed to him the view that it was possible and even likely that the outbreak stemmed from the parasite surviving wastewater treatment rather than from fresh produce. That attribution no longer stands. The article now carries an editor’s note recording it, along with a clarification Suslow provided after publication: several possible sources could explain the totality of cases, different sources of produce remain among them, and exposure from treated wastewater and possibly drinking water should be investigated before it is either included as plausible or ruled out. His own summary is that “neither fresh produce nor treated wastewater can likely explain all” the illnesses this summer. The article also discloses that he has consulted on food safety for Taylor Farms, among other produce companies, and visited the implicated Mexican farms and processing facility the month before. I relied on the earlier uncorrected version of that article in an August post, and I am correcting it here.
The lead citation says the opposite of what it was cited for
Start with the first footnote, because it carries the charge that Michigan defied reputable science. It is Beshearse and colleagues, Emerging Infectious Diseases, 2021 — a CDC-led structured expert judgment study of transmission pathways for thirty-three pathogens. Its finding for this parasite: Cyclospora cayetanensis was the only protozoan estimated to have majority transmission through the foodborne pathway. The numbers are 83 percent foodborne, with a 95 percent uncertainty interval of 59 to 99, against 6 percent waterborne. Of that 6 percent, the recreational subpathway takes 39 percent — roughly two percent of all domestically acquired cyclosporiasis, carrying an uncertainty interval that runs from 0 to 99, which is another way of saying the experts could not put a floor under it. The same paper’s comparison table puts earlier U.S. work at 99 percent foodborne and the Canadian expert elicitation at 83.1 percent. A document offered to prove that recreational water is a recognized risk factor is in fact the strongest American statement on record that this parasite is a food problem.
Finding the parasite in water is not finding a case of illness
The second and third footnotes are detection studies. Galván and colleagues, Applied and Environmental Microbiology, 2013, is a study of human-pathogenic microsporidia in Madrid; Cyclospora turned up in 9 percent of 223 samples from four drinking water plants, seven wastewater plants and six river locations across 2008 and 2009. No Spanish outbreak followed. Bilung and colleagues, 2017, is modified Ziehl-Neelsen microscopy on two rivers and two recreational sites in Malaysian Borneo, finding the parasite in two of twenty-four source-water samples. It measured no human illness and was not conducted in this country. Both papers establish that oocysts can be present in water. Neither establishes that anyone got sick from swimming.
The fourth citation is the 2023 NACMCF report on Cyclospora in produce, and it is the one IFPA should have read hardest. Teplitski chaired the NACMCF subcommittee that wrote it. The association’s own 2023 press release announcing the report says two things that cut against everything above. First, that Cyclospora is only carried by humans, and that individuals with a recent history of travel to areas where it is highly prevalent are likely the main source of contamination along the entire production, processing and food service continuum. Second, that current testing procedures do not adequately differentiate the species that sickens humans from others, referencing reports of 90 percent false-positive rates in the methodologies underlying some common tests. That is IFPA’s own summary of the federal report, and it is a warning about exactly the kind of detection study being offered as proof that the parasite is in the lake.
The 1990 outbreak was a building’s water tank, and it may not have been waterborne at all
The first reported U.S. outbreak, published by Huang and colleagues in the Annals of Internal Medicine in 1995, involved housestaff physicians and administrative staff at a Chicago hospital. CDC’s own reviewdescribes it as eleven confirmed infections among twenty-one people with diarrhea, associated with drinking tap water in a residents’ dormitory possibly contaminated by stagnant water from a rooftop storage reservoir after a pump failure. It is a plumbing story in one building. A 2019 review notes that examination of the water samples never revealed a single oocyst and that the incubation timing did not fit, so it is not clear the outbreak was waterborne at all. Offering it as support for a lake is offering a doubtful exception in the wrong category. The same CDC review records the only swimming-associated report in the American literature: one eight-year-old in Chicago who passed oocysts a week after swimming in Lake Michigan. One child, one case, thirty years ago, never an outbreak.
The parasite’s own life cycle puts mid-June overflows after the illnesses
This is the answer that uses their biology against their geography, and Michigan’s own provider bulletin states the premise: oocysts require days to weeks in the environment to become infectious, which is why direct person-to-person spread is unlikely. IFPA’s own consumer page puts the figure at seven to fifteen days in a warm, moist environment. Add roughly a week from ingestion to symptoms. CDC’s investigation page puts the first outbreak-linked onset at June 14. Work backward through a week of incubation and one to two weeks of sporulation and the contamination event lands in late May. Combined sewer overflows in mid-June are downstream of the illnesses they are offered to explain.
Humans are the only host, so the theory needs the outbreak to explain the outbreak
Nothing reaches a Detroit or Toledo sewer that a person did not shed first. This parasite infects no animal, multiplies in no lake, replicates in no soil — and that is not my characterization. IFPA’s own consumer page says humans are the only known host, and its 2023 release says the parasite can enter a farm environment only if human waste is not properly managed. For an overflow to carry it in quantity, a large population in southeast Michigan and northwest Ohio had to be infected already in late May. CDC’s national case data put the comparable 2025 figure at 1,180 domestic cases for the entire country and the entire season. That is not a reservoir. Whatever primed those sewers, if anything did, is the thing that needs explaining, and the arrow points toward the food rather than away from it.
Detroit does not drink Lake Erie water
The Great Lakes Water Authority, which supplies roughly four million people across southeast Michigan, draws from Lake Huron and from the Detroit River — a river that flows into Lake Erie. There is no Lake Erie intake. The drinking water half of the hypothesis points at a lake that is downstream of the people doing the drinking.
Water is local and this outbreak was not
CDC’s final outbreak update, posted September 11, puts the outbreak at 12,883 illnesses in 21 states, with onsets from June 14 through August 17 and at least 570 hospitalizations. The season behind it is larger: 19,595 laboratory-confirmed domestically acquired cases from May 1 through September 8, with 1,043 hospitalizations, across 49 states and the District of Columbia. Michigan’s own counts were never a Lake Erie pattern — West Michigan alone was reporting hundreds of cases by mid-July, including more than a hundred in Muskegon County, which is on Lake Michigan. Minnesota and Wisconsin share the same Great Lakes and never saw anything close. For contaminated American water to have done this, dozens of unconnected utilities and lakes would have had to fail in the same direction inside the same eight weeks, for the first time in the recorded history of the disease, and not one of them would ever have been found.
And the twenty-one states are not twenty-one places where people happened to fall ill. CDC defined this outbreak as people who reported exposure to processed iceberg lettuce from Taylor Farms de Mexico, so every one of those 12,883 illnesses is tied to one processor’s product. The last four states CDC added, named on its own page, were Georgia, Tennessee, Texas and Virginia. No body of water in Michigan or Ohio delivers a shared product exposure to Texas. A distribution list does.
The epidemic curve answered the question in July
The strongest single fact against the water theory is in CDC’s final update and it is one sentence long. At the peak, before the recall, more than a thousand infections were occurring in a single day. During August, after the lettuce came off the market, the average was fewer than two infections a day. Sewers and lakes do not respond to a produce recall. One contaminated commodity does.
No U.S. outbreak has ever been traced to swimming, and a Michigan water scientist said so
MDHHS’s July 13 statement is the sentence that drew the charge, and it is accurate: recreational water exposure is not a recognized risk factor for cyclosporiasis, and there is no evidence linking recreational water activities to this outbreak. Its consumer FAQ, revised July 24, gives the reasons rather than the conclusion — no documented U.S. outbreaks linked to pools or lakes, dilution across a lake volume against an infectious dose, the contrast with Cryptosporidium, the delay before an oocyst can infect anyone, and the fact that almost all U.S. outbreaks have been traced to fresh produce. The independent voice matters more than the department’s. Joan Rose, the Michigan State water microbiologist whose lab was developing wastewater testing for this parasite in August, said waterborne Cyclospora cases are rare, making it unlikely that people would be infected while swimming. That is the Flint-adjacent water expertise the argument reached for, answering it.
The genetics point at one source
CDC reported on August 20, on the same investigation page, that genotyping performed on samples from sick people showed the parasites that caused these illnesses are genetically related, which it called additional evidence that people in this outbreak got sick from the same source. Read it precisely: it ties patients to one another, not to any lettuce, and no product isolate exists to compare against. It is still the wrong result for the water theory. A summer of separate exposures across dozens of watersheds predicts genetic heterogeneity. None appeared.
The asymmetry in the ask
Nine proposed questions, and not one of them would improve produce traceback. No brand fields, no lot or date codes, no package type, no purchase location — nothing that moves an investigator faster from a sick person to a shipment. Anyone can read the questionnaire and see what is in it and what is not. If the goal is finding a source quickly and accurately, you want both halves. Asking only for the half that points away from your own sector is the tell, and it can be said without accusing anyone of bad faith.
What I grant, because the weak version of this loses
IFPA narrowed its claim, and the narrowing belongs in print. Its consumer page says that for the confirmed subcluster, FDA’s traceback points to imported product from central Mexico, a different pathway from the domestic water exposure it raises for the remaining clusters. It has scoped water to the cases nobody can explain and conceded the recall cluster. Those unexplained cases are real and there are a lot of them: on the two agencies’ own most recent figures, roughly six thousand seven hundred laboratory-confirmed illnesses this season sit outside the outbreak count with no source publicly identified. Adding water questions to a hypothesis-generating questionnaire costs almost nothing and might answer something. No product sample ever tested positive. Washington State and other departments published findings that cut against the simple version of the case. None of that is a reason to believe the parasite that sickened more than nineteen thousand Americans came out of Lake Erie.
IFPA knows it was not the water, and it should not be trying to carry it for some of its members.
Recent Comments