The 2024 Romaine E. coli Outbreak Had a Name the Whole Time. The FDA Just Would Not Say It. Transparency Should Not be an Option

By Bill Marler on July 21, 2026

Eighty-nine people in fifteen states got sick in November 2024. Thirty-six of them went to the hospital. Seven developed hemolytic uremic syndrome, which is the polite clinical way of saying that a child’s kidneys shut down. One person died. On January 15, 2025, the CDC and the FDA closed the file, recorded the vehicle as romaine lettuce supplied by a “common supplier,” and said nothing else. No processor. No grower. No point of service. The public notice the agencies had issued for every comparable romaine outbreak going back to 2010 was simply never issued for this one.

I wrote about that silence at the time and called the post “CDC and FDA — Romaine Silent.” I have now spent a year and a half on the other side of it, and the outbreak has become the clearest illustration I have of why transparency is not a courtesy the government extends to consumers when it is convenient. It is the thing that makes everything else — choice, accountability, deterrence — possible.

Start with what secrecy actually purchased. Because my firm represented families in several states whose infections were whole genome sequencing matches to one another, our staff epidemiologist did what the agencies would not publish: she found the common denominator. It was Taylor Farms romaine. In April 2025 we filed suit, and the company answered with a flat denial — its product was not the source, and any reporting connecting it to those illnesses was dangerous and irresponsible. That denial was only available to them because the FDA’s own traceback summary had been released with nearly every name that mattered blacked out under the (b)(4) exemption. The processor was a gray box. The grower was a gray box. The ranch, the distribution centers, the brokers, the lot codes — all gray boxes. A company cannot be contradicted by a document nobody is allowed to read.

I stayed on the agency, and in stages it took the boxes off. What had been hidden was this: Taylor Farms of California in Salinas was the sole processor, and the FDA’s summary says it supplied all the romaine available at every point of sale during the timeframe of interest. Anthony Costa & Sons of Soledad was the single grower. The records also named Andre’s Banquets and Catering of St. Louis as the caterer at the center of the largest cluster — three events over November 6 to 8, 2024, including a marching band banquet and a Veterans Day luncheon at a high school, with twenty-two illnesses from that one point of service.

None of that is a trade secret. A harvest date is not a proprietary formula. The identity of the company that processed lettuce that killed somebody is not confidential commercial information. It is the basic factual record of a public health disaster, and the only thing the redactions accomplished was to hand the responsible companies a period of deniability they had not earned.

If this were the first time, I would call it a lapse. It is not the first time, and I can trace the same habit across my entire career because I have been on the receiving end of it since the beginning. In the 1993 MMWR that described the hamburger outbreak that made me a food safety lawyer, Jack in the Box is called “Restaurant A.” In medical journals it still is. In 2010, two concurrent Salmonella outbreaks sickened 155 people in twenty-one states, and in 2011 a Salmonella Enteritidis outbreak sickened sixty-eight more across ten states — both attributed by the CDC to a “Mexican-style fast food restaurant chain, Restaurant Chain A.” I worked the elimination publicly, matching chain footprints against the ten outbreak states, and reporters at Food Safety News called every candidate. The name finally came out in February 2012, not from the CDC, but from a document the Oklahoma State Department of Health released after the agency let a deadline pass without telling the state what to do. It was Taco Bell.

In 2025 the mystery company was “a common supplier.” Last week, in the Cyclospora outbreak still going on, the FDA’s own recall notice listed the brands as “CV, JB, Mark and more” — internal customer codes, not brands, which no shopper has ever seen on a bag of lettuce. Restaurant A, Restaurant Chain A, a common supplier, CV and JB. Four decades of the same redaction in different clothes.

What is striking about the Taco Bell episode is that the agency said out loud why it was doing it. The policy, as described at the time, weighed public health against the business concerns of firms that could be hurt by bad publicity, and officials explained that the outbreak was over and that they wanted to keep relations with the chain cordial while it cooperated. I wrote then that this has it exactly backwards, and I have not moved an inch since. If a company will help find the source of an outbreak only in exchange for permanent anonymity, that tells you what its commitment to safe food is actually worth. Cooperation purchased with secrecy is not cooperation. And the government does not exist to protect a company’s bottom line from the consequences of selling food that hurts people. The reputational hit is not a bug in transparency. It is the point of it.

There is a further cost, and it is the one consumers never see happening. Naming a company is not only about stopping today’s illnesses — in most outbreaks the food is long eaten before anyone knows the source. It is about building a track record, so that a school district, a distributor, a restaurant group, or a mother in the produce aisle can see which firms keep turning up in the same sad story. That ledger only exists if somebody writes names in it. Taylor Farms’s entry would read: the 2013 Cyclospora outbreak traced to salad mix from its Mexico operation, the 2015 celery-and-onion mix tied to E. coli in Costco chicken salad, the 2024 slivered onions behind the McDonald’s Quarter Pounder outbreak that sickened more than a hundred people and killed one, this romaine outbreak, and now the Cyclospora outbreak of this summer. Four of those five entries exist because an agency, a court, or a reporter published a name. The romaine entry exists because this lawyer pried it loose eighteen months after the funeral.

I have heard the rest of the arguments too, and I have answered them one by one: that bad publicity is unfair economic harm, that the agencies cannot risk a tomatoes-then-peppers mistake, that surveillance is too underfunded to finish the job. That last one is not an argument at all. It is a confession, and the answer to it is to fund the work rather than redact our way out of admitting it is not getting done. As for the sky falling, we already ran the experiment: USDA has named contaminated-meat manufacturers for twenty years and the retailers who sold the product since 2008, and the beef industry’s trade secrets are still intact. I asked back in 2012 whether the 1993 outbreak happens at all if the 1982 one had carried McDonald’s name instead of a letter of the alphabet. I still do not know. Neither does anyone else, which is rather the problem.

Here is the part that has stayed with me longest. When the government declines to write the record, the regulated company writes it instead. I watched that happen again this month. In the middle of the Cyclosporaoutbreak now sickening more than ten thousand Americans, Taylor Farms’s public statement was rewritten on the same web address inside of about forty-eight hours — no version history, no correction note. Concern for the sick moved from the first paragraph to the fourth. The company’s grievance against the FDA moved to the lead. And the sentence acknowledging that the agency’s traceback pointed at a specific farm simply disappeared. That is what a record looks like when its author has an interest in the outcome. It is exactly the record consumers are left with whenever the agency holding the traceback data decides the names belong behind gray boxes.

I would put the principle as plainly as I can, and it has not changed since I was a young lawyer reading about Restaurant A. When people are hospitalized and someone dies, the public’s interest in knowing who grew and processed the food is at its highest and the commercial interest in anonymity is at its lowest. Redaction practice in cases like this one inverts that balance completely, treating the company names as the secret most in need of protecting when they are the single most important thing the public is entitled to know. In every one of these fights the name came out eventually — through a reporter, a state records officer, a FOIA request, or a lawsuit. Never once through the agency that had it first. The families at that Veterans Day luncheon should have been able to read it in a government notice in January 2025, in plain text, with no gray boxes at all.

There is something to do about it, and it is sitting at the FDA right now. In February, Stop Foodborne Illness filed a citizen petition asking the agency to stop treating a company’s name as confidential commercial information and, failing that, to disclose it anyway under its own statutory duty to keep food safe. The petition is right, and I said so at the time. Its argument is straightforward: the Trade Secrets Act protects sales figures and proprietary processes, not the factual identity of a firm that sells food to the public, and the D.C. Circuit has held that a company name is not commercial information in and of itself. The agency can grant it tomorrow with a memorandum — no rulemaking, no years of waiting. I would ask for one thing more. The petition speaks of the consumer-facing company, and in the 2024 E. coli outbreak, there was no such thing: the lettuce moved from a grower to a processor to a distributor to a caterer and onto a banquet plate with no brand on it anywhere. Disclosure has to reach the processor and the grower too or produce outbreaks will keep ending in gray boxes for the simple reason that nobody’s logo was on the bag. The petition’s own lead example is the outbreak in this post, and the child it describes is my client — Colton George, nine years old, eighteen days in the hospital, nine days of continuous dialysis and five more of intermittent, five transfusions, still seeing a nephrologist. His father spent months trying to learn the name of the company that sold the lettuce that did that to his son. He should not have had to.

I Was Wrong About the Positive Sample. IFPA Was Right. Now Let’s Both Go Fix the Traceability Rule.

By Bill Marler on July 20, 2026

Posted in Case News

On Sunday I graded every trade association in this outbreak on what it said and on how it handled being overtaken by events. I need to start by grading myself, because a paragraph of mine came apart within hours of publication. I wrote that there was a positive lot and that it was detained, and that we had gone from first naming a supplier to laboratory confirmation in seventy-two hours. That is not true. FDA has since concluded that the July 18 result on Taylor Farms de Mexico shredded iceberg does not represent true amplification and should be treated as a false positive. There has never been a confirmed positive product sample in this outbreak. I wrote about the retraction the same night, but I did not go back and fix the sentence, and that is on me.

It matters most because of who I used it against. I said Max Teplitski of the International Fresh Produce Association had claimed no contaminated product had been identified and that his claim was overtaken within two days. It was not overtaken. It is true today. He is entitled to hear me say so in the same place I said the opposite.

What has not changed is everything that was actually holding the case up. Michigan analyzed food-exposure details from 190 people who ate at Taco Bell and ninety percent of them ate iceberg lettuce. Traceback converged on a single supplier. FDA said on Sunday that the corrected laboratory result does not change the basis for the investigation or the overwhelming epidemiological data supporting the voluntary recall, and it went further on Monday, calling the epidemiology and traceback evidence very strong and saying that traceback and outbreak data continue to converge on shredded iceberg from Taylor Farms locations in central Mexico. An FDA official explained why the lab work is so thin and gave my own argument back to me better than I make it: with the time this takes and the epidemiology that has to happen first, the product may already have been eaten or thrown away, which turns testing into a chasing game. The word FDA is still using is implicated. The recall stands, and on Monday Taylor Farms said it will stop sourcing central Mexican iceberg for the rest of the growing season. As for the apology the company announced, an FDA official told Newsweek there was no official apology — the agency notified the company of the corrected result and explained how such a result can happen. However, my sentence about the positive sample was still wrong.

And Teplitski was half right about something else, which I should say now rather than after somebody makes me. His July 15 statement argued that patient recall explained at most about half the cases. FDA said Monday that of more than two thousand people interviewed — 2,025 of them in Michigan — many did not report eating lettuce at a chain restaurant, though lettuce showed up on the questionnaires again and again whether or not fast food did. He was right that the Taco Bell interviews do not explain everyone. But look at where that leads. It does not shrink this outbreak, it enlarges it. It means the five-state Taco Bell cluster is a subset of something bigger, that lettuce is showing up in the food histories of people who never went near a Taco Bell, and that the honest count is now above eight thousand across those five states with Michigan alone past six thousand one hundred.

Now to the reason I am writing again. On Sunday evening at 8:15 p.m. Eastern, IFPA updated its statement. Same web address. But the new text sits above a horizontal rule and the entire July 15 original sits below it, each one separately dated and separately attributed to Teplitski. Nothing was deleted. Nothing was quietly rewritten. A reader arriving today can see exactly what the association said before, what it says now, and to the minute when it changed its mind.

Hold that against the rest of the week. LGMA took down the truest paragraph any trade group wrote about this outbreak and left no trace at the same URL, and as of this morning its page still says nothing about the retraction, or about anything else since July 16. Taylor Farms overwrote its own statement, moving sympathy for the sick from the first paragraph to the fourth and dropping its acknowledgment that FDA’s traceback pointed at a specific farm. FDA, by its own account, removed the sample information from its July 18 update rather than striking it through. Three parties edited the record. One party corrected it. The one that corrected it is the produce association whose founding board chair was the CEO of Taylor Farms. Credit where it is owed.

Here is what I missed the first time, buried in the last paragraph of the original that IFPA had the nerve to leave standing. IFPA writes that it has long been a champion of end-to-end traceability, that IFPA and its members are on the record against continued delays with the implementation of the Traceability Rule, and that it has offered technological solutions that would bring the entire supply chain into compliance. On Sunday I wrote that I could find no evidence IFPA lobbied for the delay. That was an understatement. The largest produce trade association in the country is on the record against the delay, in writing, twice, during the outbreak, while the rule it wants is sitting on a shelf.

So let me take them at their word.Today is July 20, 2026. Congress set July 20, 2028 as the date before which FDA may not enforce the Food Traceability Rule. Two years, to the day. FSMA 204 requires lot-level records producible in twenty-four hours, and leafy greens are on the list, and compliance was supposed to begin this past January. Instead, FDA published a recall notice in the middle of the largest Cyclospora outbreak in American history with no lot codes and customers identified only as CV, JB, MARK, MKTSD, PK, SUB, SY and TF. IFPA’s own consumer FAQ explains why that is intolerable better than I can: one bag of salad can combine lettuce from several growers in different states or countries, washed and packed at a single facility, and once commingled there is no way to tell whose product is in the bag.

IFPA has a government relations shop, a lobbying budget north of a million dollars a year, and a political action committee. I have a blog that no one reads. IFPA can move that date and I cannot. Ask Congress to repeal the non-enforcement rider and put the January 2026 compliance date back where it was, and say so with the same specificity you used when you told Michigan its epidemiology was not robust enough.

One association that was silent on Sunday is not silent now. The National Grocers Association has published guidance for independent grocers, and it is current through the retraction, plainly written, and useful — it tells retailers what the parasite is, what FDA has and has not confirmed, and to stay in contact with their suppliers. That is more than most produced this month and I am glad to see it.

But there is a sentence in it that has to be fixed today. NGA tells its members to reassure customers that there is currently no broad retail grocery recall associated with this investigation. That is not right. The Taylor Farms recall includesMarketside Iceberg Salad and Shredded Lettuce sold at Walmart across fifteen states with best-if-used-by dates running into August, and CDC is telling retailers and consumers to throw recalled lettuce out or take it back to the store. An independent grocer reading NGA’s page could reasonably conclude there is nothing on their shelf to check. There is. And I would note, without making more of it than it deserves, that NGA is one of the organizations that applauded the thirty-month delay of the traceability rule, arguing it fell hardest on smaller grocers. Smaller grocers are exactly the members who cannot afford to guess whether a recalled bag is in the cooler. Correct the sentence and then ask yourselves whether the rule you opposed is the one that would have answered the question for them.

And now the part I want people to watch for, because it started within a day. A radio station in Michigan — in the state with six thousand sick people — reported the retraction as good news: a routine supply-chain surveillance program, advanced sequencing confirming no active parasite, proof of the rigorous multi-tiered protocols that catch threats before they reach a plate, and relief for regional growers. Nearly every clause of that is wrong. The sample was not routine; it was taken because of this outbreak. Nothing was confirmed clean; a test was found unreliable and withdrawn, which is not the same thing. And no protocol caught anything before it reached a plate, because more than eight thousand people had already eaten. That is the story this becomes if nobody says otherwise, and by August it will be the version people remember.

The scorecard from Sunday needs a correction, a concession and an addition. The correction is mine: there is no positive product sample and there never was. The concession is that Teplitski was right about that and half right about the interviews, and the half he was right about should frighten all of us, because it means we cannot account for where the rest of this outbreak came from. The addition is NGA, which showed up late and got one sentence badly wrong, and IFPA, which published a new version and left the old one standing — more than the grower group, the supplier, or the government managed. There is now an offer on the table from a trade association that says it wants the same traceability rule I want. I am taking them up on it.

Want to put me out of business? Move the date.

One Test. No Apology. Same Recall. And More Than 10,000 Sick with Cyclospora.

By Bill Marler on July 20, 2026

Posted in Case News

Taylor Fresh Foods told the public on Sunday that the FDA apologized to it – seemingly trying to sell the public that it has been exonerated as being the cause of some part of the Cyclospora Outbreak tearing through the East and Midwest.

The company led with that sentence. On Monday morning FDA answered in its own voice, and the answer contains no apology. What it contains is this: the false-positive lab sample does not change the basis for the outbreak investigation, the epidemiological data supporting the Taylor Farms recall is overwhelming, traceback and outbreak data continue to converge on shredded iceberg lettuce from Taylor Farms locations in central Mexico, and the public should keep avoiding every product on the recall list. FDA also said it will keep working with federal and state partners to ensure products implicated in this outbreak have been removed from the market.

So let us be precise about what actually happened, because there was only one test. A single sample of shredded lettuce from Taylor Farms de Mexico was pulled during import examination at the Southern Border, as part of the targeted surveillance FDA stood up because of this outbreak. It screened positive for Cyclospora. FDA notified the company. FDA’s own lab experts then re-reviewed the result, concluded the amplification was an artifact rather than a true detection, and notified the company again the same weekend. One sample, one correction, handled in about a day. I have no quarrel with any of that. Detecting this parasite in food is genuinely hard, and a lab that catches its own error and says so out loud is doing the job right.

A corrected PCR result is not an apology, and a company does not get to convert one into the other on its own website.

Notice also what quietly changed in the language. On July 17 Taylor Farms said FDA’s traceback pointed at a specific independent farm representing less than one percent of the U.S. iceberg supply. That sentence is gone from Sunday’s version. The FDA, meanwhile, says the data converge on Taylor Farms locations in central Mexico — plural. Sympathy for the sick, which opened the July 17 statement, now sits in the fourth paragraph.

And here is the part that no amount of sampling was ever going to fix. The people in this outbreak got sick beginning in mid-May. The recalled product was distributed from June 29 to July 16. Lettuce is perishable and the lag between the first illness and the first traceback was measured in months. The lots that sickened more than a thousand confirmed cases — against state-reported numbers many times higher — were eaten and thrown away long before anyone thought to swab anything. A negative product sample in a Cyclospora outbreak, or the absence of a positive one, proves nothing. The victims ate the evidence in May. I wrote about why that is, and what it would take to change it, on Sunday night.

FDA’s recall notice is still up. FDA is still telling people not to eat the product. The word implicated survived the false positive. If anyone at Taylor Farms believes otherwise, the company is free to say so plainly, publish the lot codes it promised, and name the customers behind those brand codes. That would be worth more to the public than an apology it did not receive. My first take on Sunday’s statement is here.

And, as of today, here is the updated list of the 50 State Cyclospora count:

State	2026 cases reported	Part of the outbreak?	Source
Alabama	Not separately reported	Not reported as part of the outbreak	CDC
Alaska	5 (typical annual total); 3 travel-related	No – AK DOH: no link to the outbreak	AK DOH / Epoch
Arizona	19	Cases reported; not in cluster	AZ DHS / Epoch
Arkansas	26 year-to-date	Under investigation for outbreak link	AR DOH / Epoch
California	41 provisional (Jul 14); in line with typical year	No – CDPH: monitoring, no local acquisition	CDPH
Colorado	150; 4 hospitalized; 8 of 120 domestic	No – CDPHE: not part of the Midwest outbreak	CDPHE / Epoch
Connecticut	35 (up from 19 in 2025)	Links its cases to the national outbreak	CT DPH
Delaware	Not separately reported	Not reported as part of the outbreak	CDC
Florida	106	No – FL DOH: seasonal disease Florida sees yearly	FL DOH / Epoch
Georgia	11-30 domestic since May 1	No – GA DPH: no Georgia-specific cluster	GA DPH / Epoch
Hawaii	Not separately reported	Not reported as part of the outbreak	CDC
Idaho	Not separately reported	Not reported as part of the outbreak	CDC
Illinois	277 (Jul 16); 23 hospitalized	No – IDPH: higher than average, no single cause	IDPH / TODAY
Indiana	327 (Jul 17); up from prior years	Yes – one of the 5 Taco Bell outbreak states	IN DOH / Epoch
Iowa	57; about half domestic (49 at this point in 2025)	No – IA HHS: none connected to national investigations	IA HHS / Epoch
Kansas	55 (Jul 15); 37 domestic; 30 since Jun 24	Domestic cases rising sharply; no Kansas source	KDHE
Kentucky	192 (Jul 15); 108 confirmed; 7 hospitalized	Yes – one of the 5 Taco Bell outbreak states	KY DPH / Epoch
Louisiana	1-10	Not declared part; not in cluster	LA DOH / Epoch
Maine	Not separately reported	Not reported as part of the outbreak	CDC
Maryland	69; some travel-related	No – no common link identified	MD DOH / Epoch
Massachusetts	18 (normal seasonal amount)	No – MA DPH: not affected by the Midwest outbreak	MA DPH / Epoch
Michigan	6,148 (Jul 20); 102 hospitalized – leads the nation	Yes – leads the 5-state outbreak; iceberg lettuce named	MDHHS
Minnesota	41 since May 1; some travel-related	No – MDH: no increase over expected, no outbreak	MDH / Epoch
Mississippi	Not separately reported	Not reported as part of the outbreak	CDC
Missouri	Not separately reported	Not reported as part of the outbreak	CDC
Montana	Not separately reported	Not reported as part of the outbreak	CDC
Nebraska	49	No – NE DHHS: none related to the wider outbreak	NE DHHS / Epoch
Nevada	Not separately reported	Not reported as part of the outbreak	CDC
New Hampshire	1-10	Cases reported; not in cluster	NH DHHS / Epoch
New Jersey	46 (May 1-Jul 11); within typical 30-50 range	No – NJDOH: no clusters or outbreaks	NJDOH
New Mexico	Not separately reported	Not reported as part of the outbreak	CDC
New York	517 (May 1-Jul 13), including New York City	No – NYSDOH: not a major deviation from the norm	NYSDOH / Epoch
North Carolina	307 (Jul 14); 13 hospitalized	Investigating; not in the 5-state cluster	NCDHHS
North Dakota	Not separately reported	Not reported as part of the outbreak	CDC
Ohio	1,316 (Jul 16); 96 hospitalized; Lucas Co. 278	Yes – one of the 5 Taco Bell outbreak states	ODH / Ohio Capital Journal
Oklahoma	57; 6 hospitalized	Part of the national outbreak; no Oklahoma source	OSDH / Epoch
Oregon	Not separately reported	Not reported as part of the outbreak	CDC
Pennsylvania	28 (40 in all of 2025); reporting is voluntary	No – mostly travel-related	PA DOH / Epoch
Rhode Island	4 since May 1	Cases reported; not in cluster	RI DOH / Epoch
South Carolina	Not separately reported	Not reported as part of the outbreak	CDC
South Dakota	Not separately reported	Not reported as part of the outbreak	CDC
Tennessee	11-30	Investigating; not in cluster	TN DOH / Epoch
Texas	68 (Jul 13); 15 hospitalized; all domestic	Meets CDC case definition; no source identified	TX DSHS
Utah	1-10	No – UT DHHS: not related to the wider outbreak	UT DHHS / Epoch
Vermont	Not separately reported	Not reported as part of the outbreak	CDC
Virginia	37 through Jul 4; more than double the 5-yr average	No – VDH: no evidence of a Virginia outbreak	VDH
Washington	27 since May 1; 5 domestic or possibly domestic	No – WA DOH: not associated with other states	WA DOH / Epoch
West Virginia	69 (Jul 13); 8 hospitalized	Yes – 5-state outbreak; statewide outbreak declared	WV Dept. of Health
Wisconsin	61 (May 1-Jul 15); 11 domestic	No – WI DHS: no common food item identified	WI DHS / Epoch
Wyoming	Not separately reported	Not reported as part of the outbreak	CDC

Added up, what the fifty state health departments are actually reporting for 2026 comes to roughly 10,200 cases – more than six times the 1,644 the federal government has confirmed. Michigan alone added 1,146 cases between Friday and Monday. States count differently, and some of these totals include travel-associated illness, so this is a floor and not a precise figure. It is still the only number anyone has that reflects what the states themselves are seeing. Where a state has published no 2026 count, the entry reads “not separately reported” – which is itself the point. A state that is not counting cannot tell you whether it has any.

Taylor Farms Says FDA Apologized. FDA Says, “Product Implicated in This Outbreak.”

By Bill Marler on July 20, 2026

Posted in Case News

Taylor Fresh Foods opened its Sunday statement this way: “Today, FDA apologized to us.” That is the first line. Not the second, not the fourth. The first.

I do not know what anyone said on a phone call, and I am not going to comment on a private conversation I was not part of. But FDA’s update is public, it is four sentences long, and anyone can read it. Here is the whole of what the agency actually said. Because of the complexity of detecting this parasite, its laboratory experts re-reviewed the sample results and concluded the finding does not represent true amplification and should be treated as a false positive. FDA notified Taylor Farms. And then this: FDA continues working with the firm to ensure product implicated in this outbreak has been removed from the market. FDA and its state partners continue to collect and analyze samples.

There is no apology in there. There is a laboratory correction, and there is a sentence that ought to be read slowly by anyone tempted to treat the last two days as vindication.

Implicated. That word survived the false positive. In the same breath in which it withdrew its own test result, FDA restated that this lettuce is the implicated product in an outbreak that has sickened at least 1,644 people across five states and hospitalized 94 of them after eating Taylor Farm lettuce at Taco Bells, and said the job of getting it off the market is not done. FDA gave up a sample. It did not give up the case.

Which brings up the other thing about Sunday. Taylor Farms says it completed a voluntary recall. Past tense. Closed. Undertaken in an abundance of caution. FDA, the same day, about the same lettuce, says removal is ongoing and that sampling is ongoing. One of those sentences tells customers the matter is behind them. The other says it is not. Both were published on July 19, 2026, and only one of them is the government’s.

And the timeline has not moved either. The recall came on the seventeenth. The reported positive came on the eighteenth. The retraction came on the nineteenth. Taylor Farms pulled that iceberg a full day before the test it now calls FDA’s mistake ever existed, and its own statement on Friday said it was acting on information FDA had provided and acknowledged that FDA’s traceback was pointing at a specific farm. A false positive cannot retroactively undo a recall that preceded it.

The rest of it, why nobody was ever going to find Cyclospora on lettuce that was eaten in May, why a negative sample on this parasite is close to no information at all, and why the traceability rule and a funded CDC and FDA and State Health Departments, and a genuinely independent investigative board are the only things that would have answered any of this by now, I wrote last night.

Today I only want to make the small point. An apology is something a person owes a person. Whatever FDA said privately about its own laboratory, it, Taylor Farms and Taco Bell have apologized to no one who is lying in a hospital bed in Indiana, Kentucky, Michigan, Ohio or West Virginia, and neither, so far, has anybody else. If you are going to lead a press release about a five-state outbreak with the word apology, there are 1,644 people with a better claim to it than Taylor Farms.

FDA’s Cyclospora Positive Yesterday Is a False Positive Today. The Victims Ate the Evidence in May.

By Bill Marler on July 19, 2026

Posted in Case News

Yesterday the FDA told the press, and then its own advisory, that a sample of shredded iceberg lettuce supplied by Taylor Farms de Mexico had tested positive for Cyclospora. Today the agency says it was not a positive at all. Citing the complexity of detecting this parasite, FDA laboratory experts re-reviewed the result and concluded that what they were looking at “does not represent true amplification.” As of today, there are no confirmed positive product samples in this outbreak.

Taylor Farms will likely make something of this, and so will the trade associations that spent last week insisting the case against this lettuce rested on nothing more than patient interviews. So be precise about what changed today, which is very little, and about why nobody was ever going to find Cyclospora on the lettuce that made 1,644 people sick.

Start with the calendar. Illness onsets in this outbreak begin on May 13. The recall covers product distributed between June 29 and July 16. The sample FDA thought was positive was pulled at the border under targeted import surveillance, off product still crossing into the country this week, and Taylor Farms confirmed that lot was not part of the recall at all. Three different sets of lettuce. The leaves that sickened people in Indiana, Kentucky, Michigan, Ohio and West Virginia were eaten in May and June, and the incubation period, the diagnosis, the reporting and the interviews consumed the weeks after that. By the time anyone knew what to sample, the implicated lots had been eaten or thrown out. The victims ate the evidence. That is not a failure of diligence. It is what a seven-day incubation period does to a product with a shelf life measured in days.

What the outbreak rests on was complete before FDA collected a single sample. There are 1,644 laboratory-confirmed infections in people who reported eating at Taco Bell across five states, with 94 hospitalizations. Michigan pulled ingredient-level exposure detail from 190 of those cases and 90 percent reported eating iceberg lettuce. FDA’s traceback converged on one supplier, Taylor Farms de Mexico, and the company responded by pulling all of its central-Mexico iceberg from the American market, while Taco Bell removed the ingredient nationwide. None of that came from a test tube, and none of it moved today.

It could not have come from a test tube. Cyclospora cayetanensis cannot be grown. There is no culture system and no animal model, so nothing can be enriched before testing; detection in food is washing the produce and running PCR on whatever DNA comes out. Canadian government scientists who verified FDA’s method found leafy greens spiked with two hundred oocysts ninety-three percent of the time and greens spiked with ten oocysts thirty percent of the time, and contamination on a field-grown leaf is patchy rather than uniform. A clean sample is not exoneration; it is close to no information at all. The false positive is that same weakness running the other direction, and it is the more revealing half. Because the assay reads DNA and nothing else, there is no isolate to go back to and no plate to look at, so confirming a signal means re-reading the amplification curve and deciding whether you believe it. With Listeria or E. coli O157:H7 the organism is cultured from the patient and from the food, both genomes are sequenced, and PulseNet and GenomeTrakr set them side by side until the link is microbiology instead of inference.

I know I have been beating this drum for a bit.

Which leaves exactly two things: what sick people remember eating, and whether the supply chain can be reconstructed backward from a restaurant to a field. That is the whole toolbox for this pathogen. And both halves of it are things this country has spent the last several years defunding and delaying.

Take the paper trial first. The Food Traceability Rule was finalized in November 2022 after the Center for Food Safety went to court to force FDA to issue it. Leafy greens sit on the Food Traceability List. The rule requires lot-level key data elements at each critical tracking event, and it requires that a company be able to hand the government an electronic record within twenty-four hours. Twenty-four hours. That is the difference between knowing on May 20 which fields fed which restaurants, and standing here on July 19 with FDA unable to say where the rest of the lettuce went. Because that is where we actually are. The company recall notice lists twenty-seven states of distribution against a five-state advisory, identifies its customers by brand codes rather than names, and omits the lot codes FDA’s own update says it contains. FDA’s advisory concedes the product may have reached additional states. CDC and FDA are both still describing the scope of this outbreak in terms of what the firm has chosen to disclose, five days after the supplier was named. That is not a mystery. That is a records problem, and Congress and FDA have between them made sure it stays one.

The original compliance date was January 20, 2026. In March 2025 FDA announced it would push that out thirty months, to July 20, 2028, and in November 2025 Congress made the delay binding in the continuing appropriations act. Industry asked for it. The grocery association called the rule unachievable given available technology; the grocers’ association called the timeline nearly impossible; a bill in the last Congress would have stripped the lot-code requirement from restaurants and retailers outright. The capability exists. What was delayed was the obligation to use it on behalf of the public rather than on behalf of inventory.

Now the other half, the counting, which is worse. On July 1, 2025, CDC cut FoodNet from eight pathogens to two, keeping Salmonella and Shiga toxin-producing E. coli and dropping Cyclospora, along with Campylobacter, Listeria, Shigella, Vibrio and Yersinia. The reason given was funding. One year later this parasite produced what is on any honest reading the largest foodborne outbreak in the United States in decades, and the surveillance system built to see it coming had been switched off twelve months earlier. It gets more specific than that. States do roughly half of the food processing facility inspections in this country, ninety percent of the produce safety inspections and essentially all of the retail inspections, and in December 2024 four senators had to write to FDA objecting to cuts in the state and local funding that pays for it. The DOGE federal layoffs of 2025 then hit FDA’s foods program and CDC directly. None of this was a plot. It was a series of budget decisions, taken under two administrations, each one defensible in isolation, that together dismantled the only apparatus capable of telling Americans how many of them were poisoned this summer.

And it is bipartisan. In 2012 the Obama administration zeroed out USDA’s Microbiological Data Program, the modest five-million-dollar-a-year effort that tested high-risk produce and reported positives straight to FDA, CDC and the states, and that had triggered twenty-three produce recalls in its last two years alone. Dan Flynn credited the produce industry’s lobbyists with the kill at the time. The people who wanted less produce testing got less produce testing.

Look at what that costs in this outbreak. The official count is 1,644 in five states. Cyclosporiasis has been reported this year in thirty-four states, with more than 9,000 confirmed or probable cases nationally against roughly twenty-seven hundred in all of 2025, and Michigan alone reported more than five thousand confirmed cases as of Friday with a hundred and two hospitalizations. I do not know how many of those people ate this lettuce. Neither does FDA, neither does CDC, and neither does Taylor Farms. Nobody built the machine that would know. A confirmed case requires a physician who suspects the parasite, a laboratory that runs the right test, a state health department with staff to interview the patient, and a federal system willing to count the result. Cut any one of those and the illness still happens; it simply stops appearing in the number that recalls and enforcement and, eventually, juries are built on.

Which brings me back to where this started. FDA collected the sample, FDA ran the assay, FDA announced the positive, FDA graded its own result, FDA reversed itself, and FDA then edited its own public record.

Which brings me to the case Frank Yiannas has been making. Frank ran food policy and response at FDA under two administrations, and before that spent a decade running food safety at Walmart and Disney. When he left the agency in 2023 he warned that its fragmented structure was undermining its ability to protect the public, and in November 2025, after a Listeria outbreak in prepackaged pasta meals killed six people and the original investigation failed to find a root cause, leaving the company to keep operating and keep making people sick for the better part of a year, he laid out the remedy: an independent National Foodborne Outbreak Investigation Board, modeled on the NTSB, investigating across every food category, outside the agency silos and outside the political pressure, publishing what it finds so the lessons are banked instead of lost. His argument is short and hard to answer. Foodborne outbreaks in this country are investigated by the very regulators charged with preventing them, responsibility is scattered across FDA, USDA, CDC and the states, and no one is finally accountable for any of it. He renewed the call this month, on this outbreak, arguing that we have been treating Cyclospora like hurricane season when it is a preventable problem of sourcing, agricultural water and controls in the field rather than at the shopper’s sink.

Set the two systems side by side, because Frank does and the comparison is devastating. When an airplane goes down, an independent board with no stake in the outcome takes the wreckage, publishes a probable cause and issues recommendations in public, and accidents per mile flown have fallen to almost nothing. When a bag of lettuce goes bad, the agency that regulates the lettuce investigates the lettuce, and by CDC’s own FoodNet data the incidence of Salmonella,E. coli and Listeria has been flat or rising for twenty years. GAO has found that FDA and USDA missed their own illness-reduction targets, one of them badly. An NTSB does not let the airline write the accident report.

The ask is not complicated, and it is cheaper than what we are paying now. Fund FDA’s human foods program and the state and local inspection workforce that does most of its actual work. Fund CDC and FDA and restore FoodNet to the eight pathogens it was designed to watch, this one included. Fund state epidemiologists, who are the people who found this outbreak in the first place, in Michigan, by interviewing sick people one at a time. Put the Food Traceability Rule into force instead of moving it again, because when microbiology cannot deliver a match, the records are the evidence and everyone in the supply chain already knows it. And build Frank’s board, so that the next time an outbreak sickens thousands of Americans, somebody whose budget and reputation are not on the line gets to say what happened and why.

Weekend Wrap-Up: The Source Finally Got a Name — and Then FDA Found It on Lettuce That Was Never Recalled

By Bill Marler on July 19, 2026

Posted in Case News

Sunday, July 19, 2026

Last Saturday this column was called Four Outbreaks, No Source, and a System That Would Rather Not Know. This week one of those four finally got a source, a company name, and a recall — and within seventy-two hours the government demonstrated exactly how little that accomplishes when there are no lot codes behind it, no customer list, and no obligation to produce either. A parasite most people still can’t pronounce is now the largest foodborne outbreak in this country in years. Here’s what moved.

The lettuce got a name: Taylor Farms de Mexico. On Thursday the FDA and CDC linked a five-state Cyclosporaoutbreak to shredded iceberg lettuce served at Taco Bell in Indiana, Kentucky, Michigan, Ohio and West Virginia. On Friday FDA named Taylor Farms de Mexico outright, and the company announced it was pulling every iceberg lettuce it sources from central Mexico. Taco Bell, to its credit, had already moved on its own — pulling product in select states before any federal advisory required it, and completing a nationwide removal by Friday. The operator acted before the regulator named the supplier, while the National Restaurant Association said nothing at all.

Then the recall came out, and it went to twenty-seven states, not five. The recall notice covers product distributed June 29 through July 16 across AL, AR, CT, FL, GA, IA, IL, IN, KS, KY, LA, MA, MD, MI, MO, MS, NC, NH, NJ, OH, OK, PA, SC, TN, TX, VA and WI — twenty-two states beyond the outbreak advisory. It includes Marketside-brand 12- and 24-ounce iceberg salad and 8- and 16-ounce shredded lettuce sold at Walmart, with best-if-used-by dates running to August 3. Which is difficult to square with the reassurance that grocery store lettuce wasn’t affected. I wrote about the recall Friday, and about FDA reprinting it word for word with no lot codes and no customer names on Saturday. FDA’s brand-name field literally reads “CV, JB, Mark and more.”

And then the part that should stop everyone cold. On Saturday FDA reported that a sample of shredded iceberg lettuce supplied by Taylor Farms de Mexico had tested positive for Cyclospora — collected at the border during targeted import surveillance, as a result of this very investigation. Taylor Farms confirmed the positive product is not part of its current recall. The lot is detained, and the company is still working out whether any of it reached commerce or people’s kitchens. Read that twice. The parasite turned up on product still being shipped into this country, outside a recall the company itself drew, after that same company told the public this traced to one independent farm and represented under one percent of the U.S. iceberg supply. Either the recall is scoped too narrowly or the contamination is broader than one farm. Credit where it’s due for finding it — now tell us where the rest of that lot went.

The counting problem got worse, not better. CDC’s confirmed outbreak number is 1,644 cases in five states with 94 hospitalizations and no deaths. Michigan alone has reported more than 5,000 cases and 102 hospitalizations. Cyclosporiasis has now been reported in thirty-four states this year — more than 6,700 confirmed-or-probable cases against roughly 2,700 in all of 2025. CDC now tells the public it can take as long as six weeks to determine whether a sick person belongs to a Cyclospora outbreak, up from “several weeks” a few days earlier. That isn’t a footnote. That is the whole reason the official number and the real number are living on different planets.

Four more Cyclospora investigations are sitting right next to it with nothing in the product column. The FDA’s outbreak table is now carrying fourteen active investigations, eight of them with no food identified — last week it was thirteen and eight. Four of the eight are separate Cyclospora clusters, with federal counts of 18, 8, 8 and 2 beside blank product columns. A Salmonella Oranienburg cluster climbed from 51 to 69 cases with no food named. A SalmonellaEnteritidis cluster climbed from 80 to 91, same story. An E. coli O157:H7 investigation still sits at fourteen. Here’s the tell: the row for the Taco Bell outbreak itself, reference number 1390, still reads “Not Yet Identified,” because the table hasn’t been updated since Wednesday — four days after FDA named Taylor Farms de Mexico on its own website. When the agency’s tracking document contradicts the agency’s advisory, the problem isn’t a missing rule. It’s that nobody is required to keep the ledger current.

Chilean blueberries, Maryland cheese, and four hospitalized babies. The E. coli O145:H28 outbreak tied to frozen blueberries sold at Publix moved this week — FDA has now initiated sampling. The Listeria monocytogenes outbreak in requesón and soft ricotta has a recall, an inspection and sampling all running, with a death in Maryland and a count that keeps climbing past what the table shows. Salmonella in moringa leaf powder is now the fourth separate moringa investigation in twelve months. And the infant botulism outbreak tied to Nara Organics whole-milk powdered infant formula is still open — four babies, all hospitalized, in California, Pennsylvania and Washington. I represent families in these cases. FDA sent the formula industry a letter on July 13 telling manufacturers to actually know where their ingredients come from. It should not take two botulism outbreaks in two years to get that letter written.

Recalls beyond the lettuce. Fayus Inc., doing business as Yusol International Foods, expanded its recall of OLA-OLA pounded yam over undeclared milk, across two-, four-, five- and ten-pound bags with expirations running to 2029. Canada pulled various brands of meat products over Listeria monocytogenes. PepsiCo recalled Doritos Chilli Heatwave in Ireland and the U.K. for undeclared milk. Phyllis Entis has the full international roundup, as she does every week, more reliably than the agencies whose notices she is reposting.

And the fix that Congress pushed to 2028. FDA had to run border surveillance to find this parasite on a bag of lettuce, and even then the company got to decide whether that bag fell inside or outside its own recall. The Food Traceability Ruleaddresses exactly this — lot-level records, key data elements at every critical tracking event, electronic production within twenty-four hours, leafy greens squarely on the Food Traceability List. It was finalized in November 2022. Compliance was to begin January 20, 2026. FDA moved it thirty months, and last November Congress made the delay binding through July 20, 2028. Walmart didn’t wait — it imposed its own supplier traceability requirements in August 2025 and has been assessing chargebacks since. So the technology exists and retailers can demand it. What’s missing is a legal obligation, and this week is what its absence looks like. I wrote back to Western Growers about that, and then read what everyone else in the industry had to say.

That’s the week — a parasite with a name at last, a recall that stops short of the product that tested positive, and a traceability rule that would answer the only question anyone is asking, sitting on a shelf until 2028. Naming the company was the easy part. Telling seven thousand sick Americans where the lettuce went is the part nobody is required to do. Check back next Saturday.

About Bill Marler

William “Bill” Marler has spent more than thirty years as a food safety lawyer and advocate—work that began with the 1993 Jack in the Box E. coli outbreak and has never really stopped since. In the years that followed, he has represented victims in nearly every major foodborne illness outbreak in the United States. That case, and the movement it launched, is the subject of the book “Poisoned” and the Emmy Award–winning Netflix documentary of the same name. Bill’s work has been profiled in The New Yorker (“A Bug in the System”), the Seattle Times (“30 years after the deadly E. coli outbreak, a Seattle attorney still fights for food safety”), the Washington Post (“He helped make burgers safer. Now he’s fighting food poisoning again”), and many others.

Dozens of times a year, Bill speaks to industry, regulators, and universities across the United States, Canada, Europe, Africa, China, and Australia about a simple idea: outbreaks are preventable. He has testified before Congress on the Food Safety Modernization Act and teaches food safety at institutions including the Harvard T.H. Chan School of Public Health. He writes regularly about food litigation and food safety at Marler Blog, and in 2009 he founded Food Safety News, which he continues to publish.

During an Active Cyclospora Outbreak, What do We all Need to Know?

By Bill Marler on July 19, 2026

Posted in Case News

WHAT IS CYCLOSPORA?

Cyclospora is a parasite composed of one cell, too small to be seen without a microscope. The organism was previously thought to be a blue-green alga or a large form of Cryptosporidium. Cyclospora cayetanensis was long regarded as the only species of this organism found in humans; however, genomic analyses reported by CDC scientists in 2023 indicate that what had been identified as C. cayetanensis actually comprises at least three genetically distinct, reproductively isolated species that infect people—C. cayetanensis, C. ashfordi, and C. henanensis.^[1]

Cyclosporiasis is an intestinal illness caused by the parasite Cyclospora cayetanensis, which is transmissible by ingestion of fecally contaminated food or water.^[2] Cyclosporiasis is most common in tropical and subtropical regions of the world. In the United States, foodborne outbreaks of cyclosporiasis have been linked to various types of imported fresh produce (e.g., basil, raspberries, and snow peas). For years, the lack of validated molecular typing tools hampered efforts to link cases and trace outbreaks. Since about 2018, however, CDC has developed and begun using a C. cayetanensisgenotyping system—based on targeted deep sequencing of mitochondrial and nuclear DNA markers—to complement outbreak investigations and connect geographically scattered cases, and work continues to refine it into a nationwide typing network.^[3]

Outbreaks of cyclosporiasis in humans have been reported mostly from North America, from the infection sources of contaminated fresh food products, such as soft fruits (raspberries), leafy vegetables (coriander, basil, and mixed salad), and herbs. Soil is another possible infection source, particularly in areas with poor environmental sanitation.^[4]

The Centers for Disease Control and Prevention (CDC) has been conducting national surveillance for cyclosporiasis since it became a nationally notifiable disease in January 1999. As of 2026, cyclosporiasis is reportable in 47 states, the District of Columbia, and New York City (NYC).^[5] Health departments voluntarily notify CDC of cases of cyclosporiasis through the National Notifiable Diseases Surveillance System and submit additional case information using the CDC cyclosporiasis case report form or the Cyclosporiasis National Hypothesis Generating Questionnaire (CNHGQ).^[6]

While cyclosporiasis cases are reported year-round in the United States, cyclosporiasis acquired in the United States (i.e., “domestically acquired,” or cases of cyclosporiasis that are not associated with travel to a country that is considered endemic for Cyclospora) is most common during the spring and summer months. The exact timing and duration of U.S. cyclosporiasis seasons can vary, but reports tend to increase starting in May. In 2020, multiple outbreaks of cyclosporiasis were identified and found to be linked to different produce items. As of September 23, 2020, the CDC documented 1,241 laboratory-confirmed cases of cyclosporiasis in people who had no history of international travel during the 14-day period before illness onset.^[7] Reported cases have climbed markedly over the past decade—from 537 in 2016 to 3,519 in 2018 and 4,703 in 2019—an increase that reflects both a genuine rise in illness and the growing use of rapid multiplex molecular stool panels (such as the FDA-cleared BioFire FilmArray gastrointestinal panel) that now detect Cyclospora as a matter of routine.^[8] The seasonal pattern has continued in the years since; early in the 2026 season, for example, CDC reported 145 domestically acquired cases across 17 states between May 1 and June 16.^[9]

What are the typical symptoms of Cyclospora infection?

Cyclospora infects the small intestine (bowel) and usually causes watery diarrhea, bloating, increased gas, stomach cramps, and loss of appetite, nausea, low-grade fever, and fatigue. In some cases, vomiting, explosive diarrhea, muscle aches, and substantial weight loss can occur. Some people who are infected with Cyclospora do not have any symptoms. The time between becoming infected and becoming ill is usually about one week. If not treated, the illness may last from a few days up to six weeks. Symptoms also may recur one or more times (relapse). In addition, people who have previously been infected with Cyclospora can become infected again.^[10]

Where does Cyclospora come from?

The modes of transmission of C. cayetanensis are still not completely documented, although fecal–oral transmission is the major route. Direct person-to-person transmission is unlikely. Indirect transmission can occur if an infected person contaminates the environment, the oocysts sporulate under the right conditions, and then contaminated food and water are ingested. The role of soil in transmission has also been proposed. The relative importance of these various modes of transmission and sources of infection is not known.^[11]

The dissemination of infective Cyclospora oocysts via water, soil, and unprocessed foods (e.g., fruits and vegetables, including ready-to-eat salads) is enabled by their small size (8–10 μm), low specific gravity, and high infectivity. Such oocysts can survive for weeks to months in water and food, depending on the environmental temperature, and are resistant to the routine sanitization or chemical disinfection procedures used in irrigation systems, recreational waters, or drinking water treatment plants.^[12]

How is Cyclospora diagnosed?

Cyclosporiasis is usually diagnosed symptomatically in clinical settings, including the presence of watery diarrhea, abdominal cramping, and bloating. In untreated, immunocompetent people, the diarrhea can last from days to weeks to a month or more, and can wax and wane, with variable oocyst shedding. Oocysts can continue to be shed (intermittently or continuously) by non-symptomatic people, and symptoms can also persist in the absence of oocysts in feces. In a clinical context, conventional diagnosis usually involves microscopic examination of intestinal tissue biopsy sections, stool samples for the presence of developmental stages of Cyclospora, or advanced molecular testing for DNA. Improved specificity and sensitivity have been possible largely through the use of PCR, which enables the specific amplification of genetic loci from tiny amounts of genomic DNA of Cyclospora. Because of the intermittent nature of oocyst shedding and the low numbers of this stage in feces, it is recommended that multiple stool samples be collected at 2–3-day intervals over a period of more than a week, to increase the likelihood of identifying the disease microscopically.^[13]

Why is Cyclospora so difficult to detect, in patients and in food?

Cyclosporiasis is easy to miss in the laboratory, and the reasons matter to every outbreak investigation. The routine ova-and-parasite examination performed on most stool specimens, a concentrated wet mount and trichrome stain, does not reliably detect Cyclospora, because the oocysts stain poorly with the stains that examination uses. Microscopic detection requires that the laboratory be asked to do something additional: a modified acid-fast or modified (“hot”) safranin stain of a concentrated specimen, or examination under ultraviolet light, which takes advantage of the fact that Cyclospora oocysts autofluoresce blue or green against a black background. Even then the oocysts stain inconsistently, and some appear only as unstained “ghost cells.” The American Society for Microbiology advises clinical laboratories to tell clinicians affirmatively that a routine ova-and-parasite result does not rule out Cyclospora.^[14] Head-to-head evaluation of ultraviolet fluorescence against modified acid-fast staining reflects how long this organism went underdiagnosed for want of the right stain and the right suspicion.^[15]

Shedding compounds the problem. Even symptomatic patients often do not pass enough oocysts to be found in a single specimen, and shedding is intermittent, which is why CDC and ASM both recommend that several specimens be collected on different days, generally at least three collected more than twenty-four hours apart.^[16] A patient who submits one sample on a day the diarrhea has eased may test negative while still infected. The spread of FDA-cleared multiplex molecular gastrointestinal panels has changed this substantially, because those panels detect Cyclospora DNA from a single specimen without anyone having to suspect the parasite in advance. But not every panel includes Cyclospora, and not every laboratory has adopted one. The practical consequence is that reported case counts track testing practice at least as closely as they track illness, and a state that tests aggressively will appear to have more disease than a state that does not.

Testing the food is harder still, for a reason that has no counterpart in bacterial outbreaks: Cyclospora cannot be grown. There is no in vitro culture system and no animal model, so nothing can be biologically enriched or amplified before testing.^[17] Detection instead depends on washing the produce, concentrating the wash, and running a real-time PCR assay on whatever DNA is recovered, the method set out in the FDA Bacteriological Analytical Manual, Chapter 19b. That method was originally validated for cilantro and raspberries and has since been extended by matrix studies to romaine lettuce, shredded cabbage, carrots, basil, parsley and berries; in December 2024 FDA replaced the assay’s 18S ribosomal RNA target with a more specific mitochondrial target for use by its regulatory laboratories.^[18]

The reported detection limit is on the order of five oocysts in a twenty-five gram sample, but that figure describes ideal laboratory conditions with a known quantity of oocysts deliberately applied. When Canadian government scientists verified the FDA method for use in their own national produce survey, leafy greens spiked with two hundred oocysts were detected ninety-three percent of the time, and leafy greens spiked with ten oocysts only thirty percent of the time.^[19]Complex prepared foods interfere further; the fat content of dishes such as salsa and guacamole required a separate set of method modifications before as few as five oocysts could be recovered.^[20] Contamination is also patchy rather than uniform, so a negative twenty-five gram sample says very little about the leaf next to it. And because the assay detects parasite DNA, a positive result does not by itself establish that the oocysts were sporulated and infectious, just as a negative result does not establish that a lot was clean.

Timing finishes the job. Leafy greens have a shelf life measured in days, the incubation period for cyclosporiasis is roughly a week, and diagnosis, reporting and interview take weeks more. By the time investigators know which product to sample, the implicated lot has almost always been eaten or thrown away. The absence of a positive product sample in a Cyclospora outbreak is therefore close to meaningless as evidence of anything, a point worth remembering whenever a supplier offers clean test results as exoneration.

Why genetic fingerprinting of Cyclospora does not work the way it does for E. coli and Listeria

In a bacterial outbreak the laboratory evidence can be close to conclusive. E. coli O157:H7 and Listeria monocytogenes can be cultured, grown from a patient’s stool, from a package of food, from a swab of a drain or a slicer on a processing line. Once the organism is in hand its entire genome can be sequenced, and public health laboratories compare those genomes through PulseNet and, on the food and environmental side, GenomeTrakr, using core-genome multilocus sequence typing and high-quality single-nucleotide polymorphism analysis.^[21] Isolates from patients hundreds of miles apart that differ by only a handful of alleles are for practical purposes the same strain, and when the strain from a sick person matches the strain recovered from a food or from a plant environment, the connection between the illness and its source is established as a matter of microbiology rather than inference.^[22]

None of that machinery exists for Cyclospora. Because the parasite cannot be cultured, there is no way to enrich it before sequencing; DNA must be recovered from oocysts purified directly out of raw stool by density gradient centrifugation and flow cytometry, and even then most of the DNA in the preparation belongs to something other than Cyclospora.^[23] The genome is large, with draft assemblies running to roughly forty-five megabases, so obtaining adequate coverage from a clinical specimen is impractical, and whole genome sequencing is simply not a feasible routine typing strategy for this organism.^[24] CDC has therefore had to build something different: targeted deep amplicon sequencing of eight genotyping markers, six nuclear and two mitochondrial,^[25] with the resulting haplotype data run through a custom bioinformatic clustering workflow that groups genetically related infections into what are called temporal genetic clusters.

That system works, and it deserves credit. Evaluated against epidemiologically defined clusters from 2019, it showed roughly ninety percent sensitivity and ninety-nine percent specificity.^[26] But its limits are structural rather than incidental. A specimen must amplify at a minimum of five of the eight markers before it can be run through the clustering algorithm at all, and specimens with weak parasite signal are screened out before typing begins; whether a specimen sequences successfully turns in part on how it was stored before it reached Atlanta.^[27] Not every laboratory-confirmed case has a specimen submitted for genotyping, and not every specimen submitted yields a usable genotype, both of which cause genetic clusters to understate the true size of an outbreak.^[28]

The deeper obstacle is biological. Cyclospora reproduces sexually in the human intestine, which means that a single infection is genetically heterogeneous and that two people who ate from the same bag of salad will carry parasites that are typically similar but not necessarily identical.^[29] There is no equivalent of the near-identity match that bacterial sequencing delivers. What the analysis produces is a probabilistic statement that a group of infections is more closely related to one another than to background cases, which is genuinely useful for telling epidemiologists which patients to interview together, but which cannot, standing alone, name a food. As the investigators themselves have put it, genetic clustering results become actionable, in the sense of supporting a recall, only when an epidemiological investigation identifies a common vehicle.^[30]

Genotyping the product is harder yet, because the number of oocysts on a contaminated leaf is far lower than the number in an infected patient’s stool. FDA researchers have developed a targeted amplicon sequencing assay aimed squarely at that problem, covering fifty-two loci and several hundred known single-nucleotide polymorphism sites, and have shown that it can haplotype at least twenty-four markers from leafy greens carrying as few as ten oocysts in twenty-five grams, and that samples inoculated from the same oocyst source cluster together.^[31] That is a real advance. It is not yet a deployed national surveillance network, and nothing in the Cyclospora toolbox today can do what a Listeria genome match does routinely, which is to set the organism from the patient beside the organism from the product and declare them the same.

The consequence is that Cyclospora investigations rest on epidemiology and on paper: what sick people remember eating, and whether the supply chain can be reconstructed backward from a restaurant or a bag of salad to a particular field and a particular week. Where records are lot-level and electronic, that reconstruction takes days. Where they are not, it takes months, and by then the growing season has turned over and the product is long gone.

What are the serious and long-term risks of Cyclospora infection?

Cyclospora has been associated with a variety of chronic complications such as malabsorption, reactive arthritis, and cholecystitis (inflammation of the gallbladder). Since Cyclospora infections tend to respond to the appropriate treatment, complications are more likely to occur in individuals who are not treated or not treated promptly. Extraintestinal infection also appears to occur more commonly in individuals with a compromised immune system.^[32]

Although human cyclosporiasis is usually not fatal in developed countries such as the United States, protracted diarrhea often leads to dehydration, particularly in infants who are at greatest risk of severe dehydration and death, especially if cyclosporiasis is complicated by infections with other pathogens (viral, bacterial, or parasitic—e.g., Cryptosporidium and Giardia), malnutrition, or malabsorption, particularly in underprivileged communities.^[33]

According to the CDC^[34], the recommended treatment is a combination of two antibiotics, trimethoprim-sulfamethoxazole, also known as Bactrim, Septra, or Cotrim. It is advisable for people who have diarrhea to also rest and drink plenty of fluids.

How can Cyclospora infection be prevented?

There is no vaccine for cyclosporiasis, and no medication is available to prevent infection before exposure. Because the parasite is spread through food or water that has been contaminated with feces, the best available protection, according to the CDC, is to avoid consuming food or water that may be contaminated.^[35]

Standard food-safety and hygiene practices reduce the likelihood of infection. Public health authorities recommend washing hands with soap and water before and after handling or preparing raw produce; washing all fruits and vegetables thoroughly under running water before eating, cutting, or cooking (items labeled “prewashed” need not be washed again); scrubbing firm produce such as melons and cucumbers with a clean produce brush; cutting away any damaged or bruised areas; and refrigerating cut, peeled, or cooked produce within two hours.^[36]

These measures decrease but do not eliminate the risk of transmission. Rinsing or washing produce is not likely to remove Cyclospora oocysts, which adhere tightly to surfaces, and the parasite is resistant to routine chemical disinfection methods such as those using chlorine.^[37] The theoretical infective dose may be as low as a single sporulated oocyst, and there is no way to render contaminated produce reliably safe short of avoiding it altogether.^[38]

Travelers to tropical or subtropical regions where cyclosporiasis is endemic face a heightened risk and should follow the food and water precautions set out in the CDC’s Yellow Book, consuming only water known to be safe—such as sealed bottled water, boiled tap water, or carbonated beverages. In such settings, routine chemical disinfection or sanitization of food or water is unlikely to inactivate the parasite.^[39]

For retailers, restaurants, and other food-service operators, the FDA recommends washing and sanitizing utensils, cutting boards, surfaces, display cases, and refrigerators before and after handling potentially contaminated products, sourcing produce from reputable suppliers, and maintaining purchase records so that traceback investigations can identify a contaminated source when an outbreak occurs.^[40]

^[1] CDC. (2024, May 17). Three of a Kind: CDC Researchers Find Cyclospora Is Not Just a Single Species. Advanced Molecular Detection, Centers for Disease Control and Prevention. https://www.cdc.gov/advanced-molecular-detection/php/success-stories/cyclospora.html

^[2] Casillas, S. M., Hall, R. L., & Herwaldt, B. L. (2019). Cyclosporiasis Surveillance – United States, 2011-2015. Morbidity and mortality weekly report. Surveillance summaries (Washington, D.C.: 2002), 68(3), 1–16. https://doi.org/10.15585/mmwr.ss6803a1

^[3] Nascimento, F. S., et al. (2019). Mitochondrial Junction Region as Genotyping Marker for Cyclospora cayetanensis. Emerging Infectious Diseases, 25(7). https://doi.org/10.3201/eid2507.181447

^[4] Giangaspero, A., & Gasser, R. B. (2019). Human cyclosporiasis. The Lancet Infectious Diseases, 19(7), e226–e236. https://doi.org/10.1016/S1473-3099(18)30789-8

^[5] CDC. (2026). Surveillance of Cyclosporiasis. Centers for Disease Control and Prevention. Retrieved July 2026, from https://www.cdc.gov/cyclosporiasis/php/surveillance/index.html

^[6] Casillas, Ibid, Note 2 at Page 1.

^[7] CDC. (2020, September 24). Cyclosporiasis Outbreak Investigations – United States, 2020. Centers for Disease Control and Prevention. https://www.cdc.gov/parasites/cyclosporiasis/outbreaks/2020/seasonal/index.html

^[8] CDC. (2023). Notes from the Field: Doubling of Cyclosporiasis Cases Partially Attributable to a Salad Kit — Florida, 2021–2022. MMWR Morbidity and Mortality Weekly Report, 72(27), 748–749.

^[9] CDC, Ibid, Note 5.

^[10] Cyclosporiasis – Disease. (2018, May 11). https://www.cdc.gov/parasites/cyclosporiasis/disease.html

^[11] Almeria S, Cinar HN, Dubey JP. Cyclospora cayetanensis and Cyclosporiasis: An Update. Microorganisms. 2019; 7(9):317.

^[12] Giangaspero, Ibid, Note 4 at Page 1.

^[13] Giangaspero, Ibid, Note 4 at Page 3-4.

^[14] American Society for Microbiology. (2025). Cyclospora Detection and Reporting From Clinical Samples. https://asm.org/guideline/cyclospora-detection-and-reporting-from-clinical-s

^[15] Mathison, B. A., et al. (2024). Shedding new light on Cyclospora: how the use of ultraviolet fluorescence microscopy can improve diagnosis of cyclosporiasis. Journal of Clinical Microbiology. https://doi.org/10.1128/jcm.01084-24

^[16] CDC. (2025). Clinical Overview of Cyclosporiasis. Centers for Disease Control and Prevention. https://www.cdc.gov/cyclosporiasis/hcp/clinical-overview/index.html

^[17] McCaughan, C., et al. (2026). Current Knowledge and Future Directions for Cyclospora cayetanensis Research and Its Surrogates. Comprehensive Reviews in Food Science and Food Safety. https://doi.org/10.1111/1541-4337.70327

^[18] U.S. Food and Drug Administration. BAM Chapter 19b: Molecular Detection of Cyclospora cayetanensis in Fresh Produce Using Real-Time PCR (December 2024 modification adopting the Mit1C qPCR target). https://www.fda.gov/food/laboratory-methods-food/bam-chapter-19b-molecular-detection-cyclospora-cayetanensis-fresh-produce-using-real-time-pcr

^[19] Verification and Use of the US-FDA BAM 19b Method for Detection of Cyclospora cayetanensis in a Survey of Fresh Produce by CFIA Laboratory. (2022). Microorganisms, 10(3), 559. https://doi.org/10.3390/microorganisms10030559

^[20] Almeria, S., et al. (2021). Modifications of the U.S. Food and Drug Administration validated method for detection of Cyclospora cayetanensis oocysts in prepared dishes: Mexican-style salsas and guacamole. Food Microbiology, 96, 103719. https://pubmed.ncbi.nlm.nih.gov/33494896/

^[21] Validation of Core and Whole-Genome Multi-Locus Sequence Typing Schemes for Shiga-Toxin-Producing E. coli (STEC) Outbreak Detection in a National Surveillance Network, PulseNet 2.0, USA. (2025). Microorganisms. https://pmc.ncbi.nlm.nih.gov/articles/PMC12195096/

^[22] Chen, Y., et al. (2017). Whole Genome and Core Genome Multilocus Sequence Typing and Single Nucleotide Polymorphism Analyses of Listeria monocytogenes Isolates Associated with an Outbreak Linked to Cheese, United States, 2013. Applied and Environmental Microbiology, 83(15). https://pmc.ncbi.nlm.nih.gov/articles/PMC5514676/

^[23] Qvarnstrom, Y., et al. (2018). Purification of Cyclospora cayetanensis oocysts obtained from human stool specimens for whole genome sequencing. BMC Microbiology, 18, 165. https://pmc.ncbi.nlm.nih.gov/articles/PMC6182803/

^[24] Barratt, J. L. N., et al. (2022). Genotyping Cyclospora cayetanensis From Multiple Outbreak Clusters With an Emphasis on a Cluster Linked to Bagged Salad Mix – United States, 2020. The Journal of Infectious Diseases, 225(12), 2176-2180. https://doi.org/10.1093/infdis/jiab495

^[25] Nascimento, Ibid, Note 3 (eight-marker targeted amplicon deep sequencing scheme, six nuclear and two mitochondrial markers).

^[26] Barratt, J. L. N., et al. (2021). Investigation of US Cyclospora cayetanensis outbreaks in 2019 and evaluation of an improved Cyclospora genotyping system against 2019 cyclosporiasis outbreak clusters. Epidemiology and Infection, 149, e214. https://pmc.ncbi.nlm.nih.gov/articles/PMC8506454/

^[27] Assessing the sequencing success and analytical specificity of a targeted amplicon deep sequencing workflow for genotyping the foodborne parasite Cyclospora. (2025). Journal of Clinical Microbiology. https://doi.org/10.1128/jcm.01811-24

^[28] Retrospective evaluation of an integrated molecular-epidemiological approach to cyclosporiasis outbreak investigations – United States, 2021. (2023). Epidemiology and Infection, 151, e163. https://pmc.ncbi.nlm.nih.gov/articles/PMC10540164/

^[29] Barratt, Ibid, Note 24.

^[30] Ibid, Note 28.

^[31] Nascimento, F. S., et al. (2023). Development of a targeted amplicon sequencing method for genotyping Cyclospora cayetanensis from fresh produce and clinical samples with enhanced genomic resolution and sensitivity. Frontiers in Microbiology, 14. https://pmc.ncbi.nlm.nih.gov/articles/PMC10311907/

^[32] CDC. (2020, October 21). CDC – Cyclosporiasis – Resources for Health Professionals. Centers for Disease Control and Prevention. https://www.cdc.gov/parasites/cyclosporiasis/health_professionals/index.html

^[33] Giangaspero, Ibid, Note 4 at Page 2.

^[34] CDC. (2020, September 17). CDC – Cycloporiasis – General Information – Cyclosporiasis FAQs. https://www.cdc.gov/parasites/cyclosporiasis/gen_info/faqs.html

^[35] CDC. (2024, August 8). Preventing Cyclosporiasis. Centers for Disease Control and Prevention. https://www.cdc.gov/cyclosporiasis/prevention/index.html

^[36] New York State Department of Health. (2025). Cyclospora Infection (Cyclosporiasis) Fact Sheet. https://www.health.ny.gov/diseases/communicable/cyclosporiasis/fact_sheet.htm

^[37] U.S. Food and Drug Administration. Cyclospora. https://www.fda.gov/food/foodborne-pathogens/cyclospora

^[38] Goodrich Schneider, R., Schneider, K. R., et al. Preventing Foodborne Illness: Cyclosporiasis (FSHN0519/FS130). University of Florida IFAS Extension. https://ask.ifas.ufl.edu/publication/FS130

^[39] CDC, Ibid, Note 35.

^[40] U.S. Food and Drug Administration, Ibid, Note 37.

I Wrote Back to Western Growers about Cyclospora. Then I Read What Everyone Else in the Industry Said.

By Bill Marler on July 19, 2026

Posted in Case News

Earlier this week I wrote a response to Joelle Mosso of Western Growers, who argued that this summer’s Cyclosporaresponse exposed three major gaps in our public health system. I agreed with two of her three points and said so publicly, which is not a sentence I write often.

I then went looking for what everyone else had said. One of those statements has since been quietly deleted, and the restaurant chain at the center of this outbreak moved faster than the association that speaks for it. Here is what each group said, what I say back, and the three things I think we all actually want.

The calendar matters, because everything below turns on it. On July 16 the U.S. Food and Drug Administration and the Centers for Disease Control and Prevention named shredded iceberg lettuce served at Taco Bell in five states. On July 17 FDA named the supplier, Taylor Farms de Mexico, and the company pulled all iceberg from central Mexico off the U.S. market. On July 18 FDA reported that a sample of that lettuce, caught at the border, tested positive for Cyclospora. Seventy-two hours from first naming to laboratory confirmation.

The International Fresh Produce Association, or IFPA, speaks for growers, packers and processors — and for United Fresh and the Produce Marketing Association, which merged into it in 2022, so there is no separate statement from either to look for. Its founding board chair was Bruce Taylor, CEO of Taylor Farms. On July 16, Chief Science Officer Max Teplitski warned that “the response we are seeing will likely not slow this outbreak down.” He argued that officials were rushing to place blame rather than find a source. No contaminated product had been identified though samples were taken a week earlier. Questionnaires were a valuable signal, not robust evidence.

Every one of those claims was overtaken within two days, and I say that without pleasure. No contaminated product? There is a positive lot now, and it is detained. Only questionnaires? Michigan analyzed 190 people who ate at Taco Bell and found 90 percent ate iceberg lettuce — epidemiology, then traceback to a single supplier, then laboratory confirmation. All three legs in a Cyclospora case is close to unheard of, because this parasite will not grow in culture, a point Teplitski made himself and made correctly. Then someone went looking with the right method and found the thing he said nobody had found.

IFPA also published a consumer FAQ that is far better, and I want to give it credit and then turn it around. It explains why traceback is so hard: one bag of salad can combine lettuce from several growers across different states or countries, washed and packed at a single facility, and once commingled there is no way to tell whose product is in the bag. Exactly right — and the best argument anyone has made this month for the Food Traceability Rule, which exists to solve that precise problem. IFPA wrote the case for the rule while its industry was asking to delay it.

The California Leafy Greens Marketing Agreement, or LGMA, speaks for the lettuce farms, and there are two statements at the same web address — only one still there. The version live today extends sympathies, credits officials for moving quickly, accepts Mexican-grown lettuce as the likely source, and notes California and Arizona grow about 90 percent of America’s lettuce. The version captured on July 15 called lettuce a leading suspect, said investigators were relying primarily on interviews, said Cyclospora is “more commonly associated with imported fresh produce than with domestically produced leafy greens,” and told members CDC discontinued routine national reporting of cyclosporiasis in 2025 following federal budget reductions.

Same URL, no correction, no update stamp. And here is what I want LGMA to hear: the paragraph you deleted is the truest thing any trade group has said about this outbreak. Surveillance was gutted in 2025, which is exactly why CDC confirms 1,644 outbreak cases while Michigan alone reports more than 5,000, and why I have spent two weeks explaining the difference between a confirmed case and a sick person. Your members are hurt by that gap worse than anyone — a clean California grower cannot prove they are clean when nobody can count accurately or trace quickly. Do not delete that paragraph. Send it to Congress. I will co-sign it.

The National Restaurant Association has said nothing. USA TODAY asked the NRA and Yum! Brands, McDonald’s, Chick-fil-A, Jersey Mike’s, Burger King, Subway and Wendy’s what they were doing about a parasite moving through restaurant lettuce. Not one answered. Meanwhile Dipisa’s Pizza in Stevensville, Michigan pulled lettuce, tomatoes and onions off its menu on July 10. Nor can I find a word from the Fresh Produce Association of the Americas, which represents the importers who bring Mexican produce into this country — the very channel this outbreak came through.

Silence is a position. An outbreak centered on restaurants, in states full of NRA members, and the association that exists to speak for those members produced nothing an operator could act on — no sourcing guidance, no advice on what to ask a distributor, no word of support for franchisees about to lose a month of sales through no fault of their own. Independent operators figured it out alone. They deserved better from the organization they pay dues to.

FMI — The Food Industry Association speaks for the grocery stores, and FMI got it right. I will say so plainly, because it is true and nobody else will. Its cyclosporiasis backgrounder is operational rather than defensive. Asked whether there has been a recall, it answers “Yes, products containing shredded iceberg lettuce have been recalled” and links the notice. It explains the confirmed-versus-probable divergence accurately and without spin. It is the only one of these documents that would survive being read aloud in a deposition.

My one criticism is meant as a suggestion. FMI never tells member grocers to pull the product from the shelf — and FDA named Marketside-brand bags at Walmart, with best-by dates running to August 3. CDC now says the recall covers lettuce sold at retail and tells consumers, restaurants and retailers to throw it out. FMI is the organization best positioned to make that happen in ten thousand stores at once. Say it, and you will have done more for public health this month than any regulator.

And then Taco Bell, which is not a trade association but belongs in this accounting. On Thursday July 16, before the federal confirmation and with no advisory requiring it, Taco Bell said it was voluntarily removing potentially impacted lettuce and pulling the ingredient from its national supply chain indefinitely. On July 17 it confirmed removal was complete and said, “public health is a shared responsibility among restaurants, their suppliers, and authorities.”

Taco Bell moved ahead of the government instead of waiting to be told. Worth noting that Taco Bell is a Yum! Brands company, and Yum! was among those that did not answer USA TODAY’s questions a week earlier — so the brand at the center of this acted while its parent, which also owns KFC and Pizza Hut, stayed quiet. Compare both to Taylor Farms, which called the source a specific independent farm representing less than one percent of the U.S. iceberg supply — a sentence engineered to shrink a problem — and to the trade association they both belong to, which said nothing at all. When the operator with the most to lose acts faster than the association paid to protect it, the association should ask itself why.

Now back to where I started. I told Western Growers it was right about two of three, and I meant it. What I would say to all of you comes down to three things we already agree on. Rebuild national Cyclospora surveillance and extend it into the water — irrigation, post-harvest, wastewater — where this parasite lives; Mosso asked for that and so did Teplitski, and they are both right. Give FDA real import authority, because in 2013, after a Cyclospora outbreak traced to this Taylor Farms, FDA announced increased surveillance of Mexican leafy greens, and thirteen years later announced it again, and it worked both times. And put the Food Traceability Rule back on schedule.

That last one is where I need the industry, because you can move it and I cannot. FSMA 204 requires lot-level records producible in twenty-four hours, leafy greens are on the list, and compliance was set for January 20, 2026. FDA announced a thirty-month delay in March 2025, published it that August, and in November Congress made it binding through July 20, 2028. Then this outbreak arrived, and FDA published a recall notice with no lot codes, customers identified only as CV, JB, MARK, MKTSD, PK, SUB, SY and TF, and a positive lot FDA says may be “available in commerce or in consumers homes” because nobody is required to be able to find it. IFPA’s own FAQ explains why that is unacceptable better than I can.

And here I have to say something uncomfortable, because it is true and because leaving it out would make everything above dishonest. The organization that has fought hardest against this rule is FMI — the same FMI whose outbreak communication I just praised. FMI has called the Food Traceability Rule overly complex, said its requirements go well beyond what Congress directed and are so burdensome as to not be achievable with currently available technology, and when FDA announced the thirty-month delay, President and CEO Leslie Sarasin applauded it and urged the agency to reexamine parts of the rule. The National Grocers Association applauded too, arguing the rule disproportionately impacts smaller grocers and that the original timeline was nearly impossible to meet. The Global Cold Chain Alliance said it was encouraged by this delay. And in 2024, Representatives Scott Franklin, Sanford Bishop and Jimmy Panetta introduced the Food Traceability Enhancement Act, which would have delayed compliance further and stripped lot-code requirements from restaurants, retail stores and warehouses outright. FMI and NGA both supported it.

Someone did fight for it. The Safe Food Coalition filed formal opposition to the delay in September 2025, telling the Commissioner that January 2026 already gave industry ample time and that no one should need more than a decade and a half to implement a law passed in 2010. The Center for Science in the Public Interest said the delay would only keep the public in the dark. The Center for Food Safety is the reason the rule exists at all — when FDA blew its statutory deadline, CFS sued, and a court order forced the rulemaking. To its credit, the National Restaurant Association is not on the other side of this one: its own guidance page calls the rule a measure expected to result in fewer foodborne illnesses and deaths and tells restaurants to prepare now. I could find no evidence IFPA lobbied for the delay either.

So set the two things side by side. FMI wrote the clearest, most useful public document of this outbreak — and FMI is the most effective reason FDA could not trace a bag of shredded lettuce in twenty-four hours when it mattered. I do not think that makes the backgrounder less good. I think it makes this outbreak the strongest argument against FMI’s own position that anyone could have written, and I would rather FMI reach that conclusion than have me keep pointing at it.

Congress should hold a hearing — a working one, not a show trial. Energy and Commerce and Senate HELP both have jurisdiction, and four questions need answers on the record. Why was routine Cyclospora surveillance discontinued in 2025, who decided it, and what did it save? Why did FDA screen Mexican leafy greens at the border in 2013, stop, and only start again after people got sick? How does a lot of lettuce test positive and stay outside a recall? And with this outbreak showing precisely what the traceability rule was written to prevent, will FDA and Congress defend the delay under oath?

One association argued the investigation could not work. One made the right argument and then deleted it. One said nothing. One did its job. A restaurant chain moved before it was asked. And one wrote something thoughtful enough that I answered it, which is how this whole thing started. I am not keeping score for its own sake. I care what happens next summer, because this parasite comes back every summer, and it always will until somebody fixes those three things. Want to put me out of business? Fix it.

FDA Found Cyclospora on Taylor Farms Lettuce at the Border. Credit Where It’s Due. Now Tell Us Where the Rest of That Lot Went.

By Bill Marler on July 18, 2026

Posted in Case News

Let me start with the part that deserves praise, because it is real and because I do not say it often enough. FDA tested. It stood up targeted import surveillance on shredded iceberg lettuce coming out of central Mexico, it pulled a sample, and it found Cyclospora. That is hard to do. This parasite cannot be cultured, it does not grow in a lab, and detecting it on produce takes a molecular method that most of the food safety world was not running on lettuce a decade ago. In 2013, when this same company’s same Guanajuato facility was tied to a Cyclospora outbreak, the case was built on epidemiology and traceback, and the company spent that summer telling reporters its own tests were all negative. This time somebody went looking with the right tool and found it. Whoever ordered that sampling and whoever ran that assay did their job.

Now the rest.

Read the sentences FDA wrote next, slowly. A sample of shredded iceberg lettuce supplied by Taylor Farms de Mexico has tested positive for Cyclospora. Taylor Farms has confirmed that this positive product is not part of their current recall. They are working to identify whether any part of this implicated lot is available in commerce or in consumers’ homes. The positive lot is currently detained.

In consumers’ homes. FDA has confirmed Cyclospora in a specific, identified lot of lettuce, does not know where that lot went, has told the American public it may be sitting in their refrigerators tonight, and has not recalled it. The lot is detained. A detention holds the portion FDA has its hands on. It does nothing whatsoever about whatever already cleared, already shipped, already got broken down at a distribution center and delivered to a restaurant or stacked in a cooler at a grocery store. Detention is not a recall. It protects the government’s evidence. It does not protect the family making tacos on a Saturday night.

And how is it not part of the recall? Taylor Farms told the country yesterday that it was voluntarily removing all iceberg lettuce sourced from central Mexico from the U.S. market. Not some. All. Either that statement was not true, or the recall is narrower than the statement, or this lot falls outside the June 29 through July 16 distribution window and that window is wrong. Those are the only three possibilities and every one of them is bad. FDA is repeating the company’s answer instead of demanding one.

There is a second question FDA has not answered, and it may be the more important one. When was that sample collected? FDA says only that it came from targeted import surveillance conducted as a result of this outbreak investigation. It does not give a date. FDA announced increased border screening on July 16. Taylor Farms announced its removal on July 17. If the positive was pulled after the company said it had taken all of this lettuce off the market, then contaminated product was still moving across the border after the removal was announced, and the public is entitled to know that. FDA should say when.

Two more things. FDA finally named a customer — in its advisory it identifies Marketside-brand product available at Walmart: twelve and twenty-four ounce Iceberg Salad and eight and sixteen ounce Shredded Lettuce, best if used by July 18 through August 3. Good. But go look at the recall notice itself, which is where a shopper would actually go. Those same items appear there under the code MKTSD. Not Marketside. Not Walmart. Four letters. The brand field at the top of that notice reads, in its entirety, “CV, JB, Mark and more.” FDA decoded exactly one of the eight customer codes on that list and left the other seven as initials, and it did the decoding on a different page than the one carrying the recall. And scroll up on that advisory and FDA still tells you Taylor Fresh Foods has not publicly provided distribution information or a list of customers — in the same document where FDA names one.

Then there is the lot codes. FDA’s update tells consumers that a complete list of affected products including lot codes is available on the company’s recall notice. I have read that notice. There are no lot codes in it. There are brand initials, a segment column, a product description and a best-by range. FDA is vouching for a disclosure that does not exist, on the very day it is telling people a contaminated lot may be in their kitchens.

So: thank you for testing. Genuinely, and without qualification. Now finish the job. Publish the lot number. Publish the collection date. Publish the customers behind the other seven codes — you ran the traceback, you have them. And recall the positive lot. Sixteen hundred people have an explanation and something close to seven thousand are still waiting for one.

FDA Finally Posted the Iceberg Recall. It Is Taylor Farms’ Announcement, Word for Word, With No Lot Codes and No Customer Names.

By Bill Marler on July 18, 2026

Posted in Case News

The FDA posted the iceberg lettuce recall this morning. Read the top of the page before you read anything else. It says COMPANY ANNOUNCEMENT, and then it says that when a company announces a recall, the agency posts the company’s announcement as a public service, and that FDA does not endorse either the product or the company. That is not the FDA weighing in. That is the FDA hosting. Everything below the disclaimer is Taylor Farms’ text, reproduced without a word of federal comment.

So here is what the federal government is now telling the public, in Taylor Farms’ voice. Shredded iceberg went out June 29 through July 16 to twenty-seven states — Alabama, Arkansas, Connecticut, Florida, Georgia, Iowa, Illinois, Indiana, Kansas, Kentucky, Louisiana, Massachusetts, Maryland, Michigan, Missouri, Mississippi, North Carolina, New Hampshire, New Jersey, Ohio, Oklahoma, Pennsylvania, South Carolina, Tennessee, Texas, Virginia and Wisconsin. The outbreak advisory that sits one link away still describes a five-state problem at Taco Bell.

Then comes the sentence that should have ended the guessing. A complete list of affected products, lot codes, use-by dates and instructions for returning or disposing of them, the announcement says, is available here. Look at the table. There are no lot codes. Four columns — brand, segment, description, best-if-used-by — and not one of them is a lot code. FDA published a promise of lot codes on top of a table that does not contain any, and nobody at the agency appears to have looked.

The brand column is worse. FDA’s own summary field lists the brand names as “CV, JB, Mark and more.” Those are not brands. They are Taylor Farms’ internal customer codes — CV, JB, MARK, MKTSD, PK, SUB, SY, TF. No one in America has ever seen a bag of lettuce labeled CV. I spent yesterday’s post working out who those codes most likely are, said plainly that it was inference, and invited anyone to correct me. Today one of them got confirmed. Walmart pulled four Marketside bagged iceberg products — iceberg salad in 12- and 24-ounce bags, shredded lettuce in 8- and 16-ounce bags — matching the MKTSD retail entries exactly, in all four sizes, and its notice carries the same twenty-seven states.

Credit where it belongs. Mike Snider at USA Today got Walmart on the record and put those four products and that state list in front of the public. Not the FDA, which had the recall on its own website. Not Taylor Farms, which wrote it. A trending-news reporter with a phone and the sense to call the retailer whose store brand was sitting in the table in code.

Snider also put to me the question the recall itself refuses to answer — who else got this lettuce — and I gave him what the codes point to: the national distributors US Foods and Markon, and the chains Subway and Jack in the Box. I told him what I wrote here yesterday, that this is an inadequate public notice for an outbreak that may have sickened seven thousand Americans. I will say the rest of it again, because nothing has changed since.

“Because somebody knows. Taylor Farms knows. FDA knows, because it ran the traceback. The family deciding what to put on the table tonight does not, and there is no defensible reason for that.”

And I told him the part I think actually explains the silence.

“I fear that most of the companies that are listed there are just hoping that no one ties them to illnesses.”

Notice what all of this does to the line that grocery store shredded lettuce is not affected, which CDC is still running. Marketside is grocery store shredded lettuce. It is on the recall list, labeled Retail, in consumer sizes, and it has now been pulled from shelves. The federal advisory has not been updated to say so.

Now the company’s own statement. Six sentences. Three of them are about what Taylor Farms is not: no Taylor Farms-branded salads or kits are associated with this outbreak, no Taylor Farms-branded salad kits contain iceberg lettuce, and the implicated farm represents less than one percent of the U.S. iceberg supply. The company then ran a banner across the top of taylorfarms.com repeating the first one. That is a brand-protection document with a condolence paragraph stapled to the front.

The one percent deserves its own moment. That is market share. It is not exposure. One farm’s iceberg, chopped and shredded and blended into four-by-five-pound foodservice cases and retail bags, shipped to twenty-seven states over eighteen days through Sysco and through the codes that look like US Foods, Performance Food Group and Markon, is not one percent of anything a person eating lunch cares about. The denominator is not the national lettuce supply. It is the number of bags.

There is one more thing neither document will say. Illnesses in this outbreak began in mid-May. The recall covers product distributed beginning June 29. The lettuce that made people sick was eaten weeks before the recall window opens and is long since gone. So, the recall does not answer the only question that matters to the people already sick — where did the lettuce go? — while the advisory answers it with five states and one restaurant chain and the company’s own list says twenty-seven states and at least eight customers nobody will name. Michigan alone reported more than 5,000 cases and 102 hospitalizations as of July 16. The confirmed outbreak count is 1,644.

None of this required heroics. It required lot codes and customer names on the day of the recall. That is exactly what the Food Traceability Rule under FSMA 204 was written to produce — lot-level records on leafy greens, handed to FDA within twenty-four hours — and it is exactly what got pushed to 2028. Walmart did not wait. It imposed its own supplier traceability requirements last August, which is very likely why it could identify and pull four items in a day while the federal advisory still reads like a Taco Bell story.

FDA had one job this week that it could not delegate: tell people where the lettuce went. Instead, it reprinted the press release of the company that sent it there, disclaimed any endorsement of the contents, and left the customers in code. It took a reporter to find the first name on the list.

Menu

Marler Blog

Marler Blog

The 2024 Romaine E. coli Outbreak Had a Name the Whole Time. The FDA Just Would Not Say It. Transparency Should Not be an Option

I Was Wrong About the Positive Sample. IFPA Was Right. Now Let’s Both Go Fix the Traceability Rule.

One Test. No Apology. Same Recall. And More Than 10,000 Sick with Cyclospora.

Taylor Farms Says FDA Apologized. FDA Says, “Product Implicated in This Outbreak.”

FDA’s Cyclospora Positive Yesterday Is a False Positive Today. The Victims Ate the Evidence in May.

Weekend Wrap-Up: The Source Finally Got a Name — and Then FDA Found It on Lettuce That Was Never Recalled

During an Active Cyclospora Outbreak, What do We all Need to Know?

I Wrote Back to Western Growers about Cyclospora. Then I Read What Everyone Else in the Industry Said.

FDA Found Cyclospora on Taylor Farms Lettuce at the Border. Credit Where It’s Due. Now Tell Us Where the Rest of That Lot Went.

FDA Finally Posted the Iceberg Recall. It Is Taylor Farms’ Announcement, Word for Word, With No Lot Codes and No Customer Names.

Blogs

Disease

Insight & Litigation