I learned this morning, the way we learn most things now, from a LinkedIn post. Keith Warriner, a food science professor at the University of Guelph, wrote that Doug Powell had died over the weekend. Ben Chapman, Doug’s former student and his partner in barfblog for twenty years, confirmed it.

Keith’s post was short and exactly right. Doug, he wrote, “was a character, for sure.” He established the Food Safety Network at Guelph. He was among the first to put the idea of food safety culture in front of the people who run food companies. He will be missed.

Who he was

Doug Powell grew up in Brantford, Ontario, trained in molecular biology, worked as a journalist, and as a student at Guelph in the 1980s started an alternative campus newspaper because he did not like the official one. He earned his PhD in food science at Guelph in 1996, but by then he had already been running the Food Safety Network for three years. Beginning in 1993, FSnet, AgNet, AnimalNet and FFnet went out by email every day, before Google Alerts, before RSS, before Twitter. He pulled stories off the AP wire through CompuServe at a time when CDC’s MMWR could take months to arrive by mail. If you worked in food safety in the late 1990s, in industry, in government or at a university, FSnet was in your inbox before your coffee. After the Walkerton E. coli O157 drinking-water outbreak in 2000, which killed seven people and sickened more than 2,000, he served on an expert committee on how to warn the public when the water is not safe.

He wrote one book, and it holds up. In 1997 he and William Leiss, the Canadian risk-communication scholar, published Mad Cows and Mother’s Milk: The Perils of Poor Risk Communication with McGill-Queen’s University Press. The case studies were Britain’s mad cow fiasco, hamburger disease, meaning E. coli O157, the bug that brought me into this work, genetically engineered growth hormone in milk, silicone breast implants, dioxins and PCBs, and agricultural biotechnology. The lesson in every chapter was the same: institutions that will not tell the public what they know, and what they do not know, about a risk end up paying for it, in money and in trust. Paul Slovic, the risk-perception researcher, called it “the best book I have seen on risk communication.” It went to a second edition in 2004. Doug talked for years about writing another one and never got to it.

In 2005, on an airplane, he and Ben Chapman decided to start a blog. An undergraduate named Christian came up with the name, and Doug insisted it stay lowercase: barfblog. It was to be evidence-based opinion about the things that make people sick, with references, mixed in with hockey, music and whatever else was on Doug’s mind that morning. By the time he announced in March 2019 that he was stepping back, barfblog had published 15,287 posts, and Doug had written 13,070 of them. He worked that out as a goaltender’s save percentage of .8549, which he conceded would not be great in hockey, but this was something different.

Food safety culture

The phrase “food safety culture” is everywhere now. In 2011, when Doug, Casey Jacob and Ben Chapman published “Enhancing food safety culture to reduce rates of foodborne illness” in Food Control, it was not. Two years later came “Audits and inspections are never enough.” The argument in both was the one Doug had made in a 2009 interview with Food Safety News: there is no evidence that more inspection makes food safer, and plenty of evidence that food gets safer when the people inside a company take it seriously every day and can prove it. He had been pushing the industry to market food safety directly to consumers since the 2006 spinach outbreak. The 2011 paper has been cited hundreds of times and still is.

He practiced what he preached about communication. The barfblog infosheets, one page, plain language, often funny, in English, Spanish and French, were built to be taped up in restaurant kitchens, and he and his students put cameras in real kitchens to measure whether they changed what food handlers did.

Kansas and Brisbane

Doug moved to Kansas State University in 2006 as a professor of food safety and scientific director of the International Food Safety Network. When Amy Hubbell took a position at the University of Queensland, he followed her to Brisbane and kept teaching and publishing from the other side of the world. In 2013 K-State declined to renew his contract. His own headline for the news was that he had been fired for bad attendance. Eight years later, when every university on earth had gone to distance learning, he noted the irony and did not spare anyone the barb. He kept writing from his kitchen in Brisbane, where, as his byline put it, he shopped, cooked and fermented.

Doug and me

Doug and I quoted each other for years, not always approvingly. Two Doug-isms made it into this blog and into Food Safety News more than once. Ask anyone who works in food safety what one food they will not eat, and the answer is raw sprouts, always. And “stick it in,” his instruction to use a tip-sensitive digital thermometer rather than color, time or guesswork to decide when the chicken is done. He was immediately suspicious of anyone who said “trust me,” a category that included regulators, companies, celebrity chefs and, on occasion, lawyers. We did not always agree, and when we did not, he said so in print, which is what he did with everyone. That was Doug.

Stepping away

Doug wrote about his own health with the same bluntness he applied to everyone else’s. In the March 2019 post announcing he would step back, he disclosed that he had early-onset dementia and other health problems, and said someone else should be in charge. “I did my best, even when my best wasn’t good enough,” he wrote. Then, being Doug, he kept posting anyway, until the summer of 2021, when the posts stopped. He signed off that 2019 piece by describing the job as improving food safety one digital thermometer, one skipped serving of raw sprouts and one called-out piece of bad advice at a time. That is as good a summary of his career as anyone will write.

He held three passports, Canadian, American and Australian. He described himself as an OK goaltender in pickup hockey, taught kids to skate in Guelph and coached them in Brisbane, and raised five daughters who gave him grandchildren.

My heart goes out to Amy Hubbell, to the mothers of his daughters, to his five daughters and his grandchildren, and to Ben Chapman and the students Doug trained and bragged about. Food safety has one fewer honest voice this morning than it did last week.

Peace and love, Doug.

First, an apology.

In September and October I told you I would post seven citizen petitions for comment, take your edits, and file them. I posted them. You sent edits. I have not filed them.

Two reasons. The day job has been relentless this fall, between two infant formula botulism outbreaks, a chicken wrap outbreak, raw oysters, pumpkin patches and Cyclospora. And, you sent me more good thoughts and edits than I ever expected.

That second reason is the better one. Microbiologists, a law professor, a farm-animal advocacy group, Stop Foodborne Illness, and parents whose babies were on ventilators all read these drafts closely, and every one of them made the drafts better. Thank you. I mean it.

So here is the plan. I will file most, if not all, of them by the first week of December. Before I do, if you want another look at any one of them, email me at bmarler@marlerclark.com. Tell me which one, and I will send you the current draft.

Here is what they are, and why we need them.

Why petitions at all

A citizen petition is the most underused tool in food safety. Anyone can file one. It needs no vote, no sponsor and no committee. Under FDA’s rules the agency owes an answer within 180 days, which is a deadline no bill in Congress has.

An agency can still say no. But a denial with reasons can be challenged, and silence cannot. And every one of these petitions asks the agency to do something it has already said, in its own record, that it should do.

Each one is also paired with the legislation that would do the same thing if the agency will not. A new Congress arrives on January 3. I want it to find the bills already written.

1. Test infant formula for botulism

Fifty-two babies were hospitalized with infant botulism after drinking powdered formula: 48 in the ByHeart outbreak across seventeen states, and four more in the Nara outbreak. Whole genome sequencing tied them to one supply of milk powder. Every one was treated with an antitoxin that California prices at $69,300 a dose.

Federal rules require powdered formula to be tested for exactly two organisms, Cronobacter and Salmonella. Clostridium botulinum is not one of them.

The petition asks FDA to:

  • add botulinum neurotoxin-producing clostridia to the finished-product testing table, with a sample large enough to find low-level contamination
  • require testing of the milk powder and other dairy ingredients before they are used, because that is where the spores came in
  • require notice to FDA within one business day of a positive
  • finish validating the test method it presented in July
  • require lot-level records tracing each ingredient back to the dryer and the farm

Stop Foodborne Illness, Consumer Reports, EWG and forty-four parents of twenty-three of those babies are co-petitioners so far.

The legislative version is H.R. 7867, with one sentence added to require testing of the dairy ingredient. The Senate already passed a narrower bill, S. 272, without objection in April. It has been sitting at the Speaker’s desk since May.

2. A company’s name is not a trade secret

FDA will not tell you the name of a company in an outbreak unless there is a recall. Its theory is that the name is “confidential commercial information.”

Here is what that looks like in practice. In October, FDA announced an E. coli outbreak linked to ready-to-eat chicken wraps. Fifty people were sick. The agency said the wraps had been “proactively removed” and named no brand, no maker and no state. The public learned the brand from a lawsuit. When FDA closed the outbreak, it said more detailed product information would not have given consumers an action they could take.

Tell that to the next family.

The petition asks FDA to write a rule saying the identity of a firm implicated in an outbreak is not confidential commercial information. That means where epidemiology implicates a food, and traceback or laboratory evidence identifies the firm as the source, the agency names it. It covers the whole chain, grower to retailer, and it applies whether or not anyone uses the word “recall.” The firm gets a day to correct factual errors.

It also asks the agency to start now, using authority it already has. It backs up the petition Stop Foodborne Illness filed in February. Check FDA’s own outbreak table any week and count how many investigations name no food at all.

3. Tell people where the recalled food went

Since 2008, USDA has published the list of stores that received recalled meat and poultry. FDA generally does not. California’s health department already publishes recall distribution lists down to the street address of each store.

The petition asks FDA to:

  • publish the retail consignee list in every Class I food recall on the day the recall is announced, reaching restaurants, schools and hospitals
  • classify every recall within ten days and publish how long it took
  • say in writing that removing an outbreak food from the market is a recall, whatever the company calls it

The chicken wraps were “proactively removed,” never classified, and never listed.

4. Let FDA take a sample where the contamination started

In the 2018 Yuma romaine outbreak, the outbreak strain turned up in irrigation canal water running past a large cattle feedlot. FDA had no clear authority to walk onto the feedlot and take a sample. I see the same pattern in my own files over and over. The investigation follows the evidence to the field, the field sits next to the animals, and the investigation stops at a fence.

Congress has been asked to fix this in four Congresses without a single hearing. This petition does not wait. Section 361 of the Public Health Service Act lets FDA write rules to stop the spread of communicable disease, and the agency used it in 1975 to regulate pet turtles.

The petition asks for a narrow rule. If the Commissioner finds in writing that an animal feeding operation is implicated in a multistate outbreak, the operation must allow entry and sampling for a limited time. That is all. No seizure, no quarantine and no production records.

A law professor whose book on foodborne illness is the standard account marked it up. Others pointed out that it reaches poultry and egg operations too, not just cattle. They are right, and I am working that in.

5. Salmonella in raw poultry

In 1994, after Jack in the Box, USDA declared E. coli O157:H7 an adulterant in ground beef. It took my beef practice away, and it is the best thing that has happened to food safety in my lifetime. We have never done the same for Salmonella in chicken.

USDA did take one step. In 2024 it declared breaded stuffed chicken with Salmonella adulterated at 1 colony-forming unit per gram. Then it delayed the testing indefinitely. Consumer Reports has asked for the databehind that decision. USDA also withdrew its broader poultry framework in April 2025.

The petition to FSIS asks it to declare raw ground chicken and turkey and raw chicken parts with Salmonella at 1 CFU/g or more adulterated. In the alternative, it asks the agency to do so for the three serotypes FSIS itself says cause the most illness. Either way, it asks FSIS to start the breaded stuffed chicken testing it already announced, or publish a date. And if FSIS says no to all of that, it should at least require serotype reporting by plant.

It is deliberately narrower than the 2020 petition, because the record FSIS has built since then is a poultry record.

6. Put a warning label on raw sprouts

Since 1998, FDA has told children, older adults, pregnant women and people with weakened immune systems not to eat raw sprouts. It has said so in press releases, consumer advice and outbreak notices, including two sprout outbreaks this fall. It has never said so on the package, where people actually decide.

The bacteria can be inside the seed, where washing cannot reach. The petition asks for a short, prescribed warning on packaged sprouts and at the point of sale, modeled on the juice warning FDA wrote in 1998 in 21 CFR 101.17. Sprouts treated to make them safe would be exempt.

I first asked FDA for this in 2003. A microbiologist who reads every outbreak notice, a restaurant culinary director and Stop Foodborne Illness all improved it.

7. A warning label on raw milk cheese, and the 60-day rule

Raw milk cannot cross state lines in a bottle. It can cross as cheese, on the strength of a 60-day aging rule written in the 1940s. In 2013, FDA and Health Canada published a risk assessment finding that raw milk cheese carries far more listeriosis risk per serving than pasteurized cheese, and FDA promised to reevaluate the 60-day rule. It never did.

Since then, four outbreaks in thirty-one months have come from cheese that followed the rule. Forty-three people got sick, seventeen were hospitalized, and six children had kidney failure.

The petition asks for:

  • a prescribed warning on raw milk cheese, including cheese cut at the counter, imported and sold online
  • the 60-day reevaluation, finished within a year
  • validation, under the preventive controls rule, by cheesemakers who rely on aging

My firm represents children from two of those four outbreaks, and the petition says so. Weigh it accordingly.

What happens next

I will spend the rest of November working your edits in (my evening job). Then I will file them, most if not all by the first week of December, and post the docket numbers on this blog as each is assigned. Once a petition is filed, anyone can comment directly to the agency at regulations.gov, and I hope you will.

None of these asks is radical. Every one of them is something the agency has already said, somewhere in its own record, that it ought to do. There is no kill step for Cyclospora, and there is no kill step for an agency that will not act. Petitions are how we make them answer.

Sources

  1. CDC, Infant Botulism Outbreak Linked to Powdered Infant Formula, November 2025
  2. CDC, Investigation Update on Infant Botulism Outbreak, June 2026
  3. 21 CFR 106.55, Controls to prevent adulteration from microorganisms
  4. H.R. 7867, Infant Formula Safety Modernization Act of 2026 (Congress.gov)
  5. S. 272, Protect Infant Formula from Contamination Act (Congress.gov)
  6. HHS, First Major Update to Infant Formula Standards in Nearly 30 Years (Oct. 29, 2026)
  7. Stop Foodborne Illness, citizen petition on company names as confidential commercial information (Feb. 2026)
  8. FDA, Investigations of Foodborne Illness Outbreaks
  9. FSIS, Risk Assessment of E. coli O157:H7 in Ground Beef
  10. FSIS, Salmonella in Not-Ready-To-Eat Breaded Stuffed Chicken Products, final determination (Federal Register, May 1, 2024)
  11. FSIS, Delayed Verification Sampling of NRTE Breaded Stuffed Chicken Products (Dec. 2025)
  12. Food Safety Magazine, Consumer Reports to USDA: Release Validation Data on Salmonella Tests (June 2026)
  13. FSIS, Salmonella Framework for Raw Poultry Products, withdrawal (Federal Register, Apr. 25, 2025)
  14. 21 CFR 101.17, Food labeling warning, notice, and safe handling statements
  15. Bill Marler, There is no kill step for Cyclospora (Food Safety News, July 2026)

I have been doing this since January 1993. A hamburger at Jack in the Box put me in the food safety business, and I have said for more than thirty years that I would like to be put out of it.

I am not out of it. I am busier than ever.

That is the problem, and it is why I spent this fall writing something I never expected to write: a working agenda for American food safety. I call it Project 2027. It is not a manifesto. It is a list of things that can be done, by people who already exist, with bills, dockets and deadlines attached.

What we lost in two years

Here is what happened while most of the country was looking the other way.

In April 2025, USDA withdrew its Salmonella framework for raw poultry after three years of work.

In July 2025, CDC cut FoodNet, the system that measures whether anything we do is working, from eight pathogens to two.

In September 2025, the Justice Department eliminated the Consumer Protection Branch, the people who prosecuted food crimes.

In November 2025, the law that ended the shutdown carried a rider that bars FDA from enforcing its own traceability rule before July 2028. The House bill for this year repeats the bar.

In December 2025, FSIS delayed indefinitely testing under its own 2024 determination that breaded stuffed chicken with Salmonella is adulterated. The standard is law. Nobody is checking for it. Consumer Reports asked for the data behind that decision in June. I am not aware of an answer.

Almost none of that took an act of Congress. It was done by notice, by budget line and by rider. Which means most of it can be undone the same way.

And then the bills came due

This summer the country ran the largest Cyclospora outbreak in its history. More than 18,000 laboratory-confirmed cases. That is only the people who got a stool test that found the parasite. There is no kill step for Cyclospora, and we had stopped measuring the trend a year earlier.

Fifty-two babies were hospitalized with botulism after drinking powdered infant formula: 48 in the ByHeart outbreak and four more in the Nara outbreak. Whole genome sequencing tied them to one supply of milk powder. Federal rules still require formula to be tested for exactly two organisms, and Clostridium botulinum is not one of them.

In October, fifty people got sick from E. coli in ready-to-eat chicken wraps. FDA said the product had been “proactively removed.” It named no brand, no maker and no state. The public learned the brand from a lawsuit. When the agency closed the file, it said more detailed information “would not have provided consumers with an action they could take.” Check FDA’s own outbreak table on any given week and count how many investigations name no food at all.

USDA’s economists put the cost of foodborne illness at $74.7 billion a year. Salmonella alone is about $17 billion of it. Deaths are 56 percent of the cost, and long-term health problems are another 31 percent. None of it buys anybody a new kidney.

This is not only about bugs

I have spent my career on bacteria and viruses. My paper shows it. But I want to say something plainly that the first draft did not say well enough.

Chronic disease is not somebody else’s fight.

Ninety percent of the nation’s health care spending goes to people with chronic and mental health conditions. Obesity alone costs the health system nearly $173 billion a year. The foodborne number is real, and it is a rounding error next to that.

The two halves of this community, the people who fight pathogens and the people who fight diet-related disease, have treated each other as separate worlds for too long. We want the same things more often than we admit. Honest labels. Real review of what goes into food before it is sold. School meals that are both safe and nourishing. States that keep the right to protect their own people.

That is why Project 2027 now treats chronic disease at the same weight as the pathogen items, in the same signature block and the same hearing requests. And it gives the chronic side its own first item.

David Kessler petitioned FDA in August 2025 (docket FDA-2025-P-3071). The statutory clock ran out long ago. FDA sent a procedural interim response in February. The Secretary then said on national television that the agency would act. It has not.

The ask is one letter to FDA and to OMB: release the response. Nourish Science has committed to sign it. EWG is willing. I will draft it before the end of the year. It costs the government nothing, and it is the one place in the whole agenda where both halves of this community are asking the same office for the same thing.

What Project 2027 is

The first draft went out in September to about thirty people: consumer groups, former agency heads, state regulators, industry food safety people, law professors and scientists. It came back with dozens of comments and more than 180 separate points.

Some of the best came from people I fight in court. A produce industry veteran sent twenty-six comments and offered to do the research herself. A former director of FDA’s food center corrected me on what the agency had already done, and she was right. A state regulator with fifty years in the work reminded me that most food safety in this country happens at the state and local level, not in Washington.

They also told me thirty-plus items is too many. I heard them. The next draft will be shorter.

What survives are things with a vehicle and somebody already pushing them:

  • Keep the traceability rider out of the next spending bill.
  • Test infant formula, and the milk powder it is made from, for the organism that paralyzed these babies.
  • Restore surveillance, laboratories first.
  • Name the companies in outbreaks, recall or no recall.
  • Publish recall consignee lists the day a recall is announced.
  • Let FDA take samples at the animal operation next to the field when an outbreak points there.
  • Bring back produce testing, this time in statute.
  • Release the response to the Kessler petition.

Bring back the MDP

That produce testing item deserves its own paragraph.

From 2001 to 2012, USDA’s Microbiological Data Program tested 10,000 to 15,000 produce samples a year. That was roughly 80 percent of all federal produce testing, for about $4.5 million a year.

In 2012 it found Listeria in cantaloupe from Burch Farms. The recall reached nearly 189,000 melons, and nobody was reported sick. That was eleven months after Listeria in Jensen Farms cantaloupe killed 33 people. I represented those families.

Then it was shut down after years of produce industry lobbying. The scientists who analyzed its data wrote in 2016 that the monitoring had to continue somewhere. It has not.

I have been asking to bring it back since 2024. Now there is a drafted bill, the Produce Pathogen Surveillance Act. It has permanent authorization, whole genome sequencing of every positive, and no more “blind” sampling. Firms behind confirmed positives get named. It was a Bush-era program, and it should not be a partisan one now.

Seven petitions you can comment on now

I have also drafted seven citizen petitions from the agenda, six to FDA and one to FSIS. None has been filed. All are still open for comment.

They cover:

Each one is paired with the legislation that would do the same thing if an agency will not.

The election changed the math

On January 3, a new Congress takes office with new majorities in both chambers. Every committee chair this agenda addresses will change. Every bill in it dies and has to be reintroduced.

That sounds like a setback. It is an opening.

New members come with new ideas and no settled habits, and plenty of them campaigned on what is in our food. The incoming House majority already has a planning memo with infant formula, heavy metals in baby food and GRAS reform at the top. It has almost nothing on pathogens. This community should hand them a page that does.

And there is still time this year. Government funding runs out December 11. The infant formula bill, S. 272, passed the Senate without a single objection in April. It has sat at the Speaker’s desk since May. The House could pass it before Christmas, or amend it with the testing language of H.R. 7867 and send it back.

Credit where it is due: this administration just proposed the first update to formula nutrient standards in nearly thirty years, and I said so. Food safety has never been a party question. In 1994, USDA declared E. coli O157:H7 an adulterant in ground beef after Jack in the Box. Every Congress and every administration since has left it alone. It took my beef practice away, and it is the best thing that has happened to food safety in my lifetime.

What I am asking

If you work in food safety, public health, nutrition, agriculture or the food industry, read the draft. Tell me which items your organization would sign. Item by item is fine; nobody has to sign the whole thing. Tell me what is wrong or missing. Comment on the petitions.

I would also like to host a working meeting in Washington in December, after the funding deadline. It will be one day, with a way to join remotely. I want both halves of this community in the same room, settling the shorter version that goes to the new Congress in January.

None of this is radical. Most of it is finishing what somebody already started. Food safety is not expensive. Outbreaks are expensive.

Put me out of the lettuce business, please. Put me out of the formula business and the chicken business too. That is what Project 2027 is for.

Sources

  1. FSIS, Salmonella Framework for Raw Poultry Products, withdrawal (Federal Register, Apr. 25, 2025)
  2. Congressional Research Service, Agriculture and Related Agencies: FY2027 Appropriations (R49343)
  3. FSIS, Delayed Verification Sampling of NRTE Breaded Stuffed Chicken Products (Dec. 2025)
  4. FSIS, Salmonella in Not-Ready-To-Eat Breaded Stuffed Chicken Products, final determination (Federal Register, May 1, 2024)
  5. Food Safety Magazine, Consumer Reports to USDA: Release Validation Data on Salmonella Tests (June 2026)
  6. Bill Marler, There is no kill step for Cyclospora (Food Safety News, July 2026)
  7. CDC, Infant Botulism Outbreak Linked to Powdered Infant Formula, November 2025
  8. CDC, Investigation Update on Infant Botulism Outbreak, June 2026
  9. 21 CFR 106.55, Controls to prevent adulteration from microorganisms
  10. FDA, Investigations of Foodborne Illness Outbreaks
  11. USDA Economic Research Service, Cost Estimates of Foodborne Illnesses
  12. Hoffmann et al., Economic Burden of Foodborne Illnesses Acquired in the United States (2025)
  13. CDC, Fast Facts: Health and Economic Costs of Chronic Conditions
  14. Regulations.gov, Docket FDA-2025-P-3071 (Kessler citizen petition)
  15. USDA Agricultural Marketing Service, Microbiological Data Program
  16. Food Safety News, MDP Shuts Down; USDA Testing of Produce for Pathogens Halted (Jan. 2013)
  17. Reddy, Wang, Adams and Feng, Prevalence and Characteristics of Salmonella Serotypes Isolated from Fresh Produce Marketed in the United States, J. Food Prot. (2016)
  18. Bill Marler, Time to bring back the MDP to help Make Produce Safe Again (Food Safety News, Nov. 2024)
  19. Draft citizen petition, Disclosure of Firms Implicated in Foodborne Illness Outbreaks (Marler Blog)
  20. Stop Foodborne Illness, citizen petition on company names as confidential commercial information (Feb. 2026)
  21. K&L Gates, 2026 Midterm Election Guide to the 120th Congress (Nov. 4, 2026)
  22. S. 272, Protect Infant Formula from Contamination Act (Congress.gov)
  23. H.R. 7867, Infant Formula Safety Modernization Act of 2026 (Congress.gov)
  24. HHS, First Major Update to Infant Formula Standards in Nearly 30 Years (Oct. 29, 2026)
  25. Bill Marler, Credit Where It Is Due on Formula Nutrients (Marler Blog, Oct. 29, 2026)
  26. FSIS, Risk Assessment of E. coli O157:H7 in Ground Beef
  27. Bill Marler, Food safety is not expensive. Outbreaks are expensive. (Food Safety News, Aug. 2026)
  28. Bill Marler, Put me out of the lettuce business, please (Food Safety News, Oct. 2026)

For twenty years I have been saying some version of the same thing about leafy greens and cattle. Water runs downhill. Cows poop. Lettuce grows at the bottom of the hill.

Now FDA’s own scientists have put real numbers on it, and they deserve credit for doing it.

A new paper in the Journal of Food Protection, posted online November 6, 2026, by Mehran Niazi and colleagues in FDA’s Human Foods Program, builds a model that follows rainwater off cattle pastures and an animal feeding operation in the Salinas Valley and traces which leafy-green farms that runoff reaches on its way to the sea.

This is exactly the kind of work I have hoped to see for years. It is careful, transparent and honest about its limits. And it is the closest thing I have seen to a map of the plumbing that keeps sending E. coli O157:H7 into our salad bowls.

We Keep Finding Cows at the Scene

If you have read this blog for any length of time, you know the pattern. Here is a short refresher.

In 2006, bagged spinach sickened about 200 people nationwide and killed three. I represented many of those families. Investigators found the outbreak strain in cattle feces, feral swine, surface water and sediment on a San Benito County ranch, roughly a mile from an implicated spinach field. The final state and federal report could not say definitively how the spinach was contaminated, but it flagged irrigation wells sitting near surface waterways exposed to cattle and wildlife feces.

In spring 2018, romaine from the Yuma growing region sickened 210 people in 36 states, hospitalized 96, caused 27 cases of kidney failure and killed five. FDA found the outbreak strain in irrigation canal waterrunning alongside a massive concentrated animal feeding operation.

In fall 2018, romaine from California’s Central Coast sickened 62 people in the United States. FDA found the outbreak strain in sediment from an on-farm water reservoir in Santa Barbara County.

In fall 2019, three separate outbreaks were traced to Salinas Valley romaine. The largest sickened 167 people in 27 states. In its investigation report, FDA concluded that adjacent and nearby cattle grazing appeared to play a critical role, and listed water runoff, wind, animals and vehicles as possible routes from the cattle to the lettuce.

In fall 2020, it happened again: 40 sick in 19 states, 20 hospitalized, four with hemolytic uremic syndrome. This time FDA found the outbreak strain in a cattle feces sample on a roadside about 1.3 miles upslope from a produce farm with multiple fields linked to the outbreak.

And an FDA and CDC review of farm investigations tied to leafy-green STEC outbreaks from 2009 through 2021 found the same themes again and again. One of that paper’s authors, Kurt Nolte, is also an author on this new study.

So nobody serious is arguing anymore about whether cattle matter. The harder questions have been how, when and which farms. This paper takes those questions on directly, and that is a real step forward.

What the FDA Team Actually Did

Strip away the jargon and the method is pretty intuitive.

The team drew a study area out of 14 small watersheds in the Salinas Valley, about 1,239 square kilometers, or roughly 480 square miles. They mapped the leafy-green fields, the grazing pastures (each tagged with an estimated cattle capacity) and one animal feeding operation.

Then they pulled 48 rainstorms from 2018 through 2021 out of NOAA data. For each storm, they used the standard USDA Curve Number method to estimate how much rain would run off a pasture versus a compacted feedlot surface instead of soaking into the ground.

Next came the plumbing. Using elevation data, a computer figured out which direction water flows from every patch of ground and built a drainage network. Each pasture, the feeding operation and each farm were “snapped” onto that network, so the model could tell whether a given pasture actually sits upstream of a given farm, and how far the water travels between them.

Finally, each pasture’s influence was scaled by its cattle capacity and shrunk with distance. At their baseline setting, a pasture’s weight drops by about 63 percent over 10 kilometers of channel. Multiply by the storm runoff and you get a relative “runoff-exposure” score for every farm in every storm.

What They Found

The results are genuinely useful. The routing turned up 1,818 pasture-to-farm connections, involving 58 pastures and 277 downstream farms. The single animal feeding operation was hydrologically connected to 45 farms.

These are not short trips. The median distance along the channel from a pasture to a connected farm was about 16.8 kilometers, a little over 10 miles. The longest was 82.6 kilometers, more than 50 miles.

Most farms, 86.3 percent, had no modeled upstream pasture contribution at all under the baseline assumptions. That is good news for those growers, and it tells regulators where not to waste their time.

Here is the finding I like best: being next to cows is not the same as being downstream of them.

One pasture with an estimated capacity of 1,170 cattle physically overlapped 13 farm fields. Only two of those 13 were actually downstream of it. Yet that same pasture was hydrologically connected to 56 farms. A second pasture, about 722 cattle, overlapped eight farms, but only one was downstream, and it was connected to 36 farms in all.

Think about what that means for a buffer-distance rule. A setback measured in feet from the fence line protects the farm next door that might not be at risk, and does nothing for the farm ten miles down the creek that is.

For one farm sitting just 0.6 kilometers downstream of that 1,170-head pasture, the model kept about 94 percent of the pasture’s weight. During the biggest storm in the dataset, that one pasture accounted for a large share of the farm’s highest score.

And speaking of the biggest storm: every farm with a nonzero score hit its peak during the January 27 to 29, 2021 storm, which dropped an area-wide average of 147.3 millimeters of rain, almost 6 inches. Big storms do not just add runoff; under the Curve Number math, they add it disproportionately.

The team also stress-tested its assumptions. Changing the runoff parameters did not change the farm rankings at all. Changing how fast influence fades with distance left the overall rankings almost untouched, but it did reshuffle the very top. Compared with a simple count of upstream pastures, the top 25 farms overlapped by zero at the steepest decay setting, eight at baseline and 24 at the gentlest. In plain English: the model is solid at sorting farms into “worry” and “don’t worry,” but you should not bet everything on which farm is number one.

Honest About Its Limits, and That Is a Strength

The authors are admirably clear about what this model is not. It does not measure pathogen presence, concentration, contamination probability or food-safety risk. It does not model groundwater, floods, overbank flow, buffers, wetlands, die-off, sediment or irrigation practices. It says nothing about wind, dust, wildlife or feral pigs, all of which have shown up in past investigations. And the animal feeding operation is treated as a simple yes-or-no flag, because nobody had data on how it handles manure and stormwater.

That candor is a strength, not a weakness. The authors built a screening tool, said plainly that it is a screening tool, and published the data and sensitivity tests so others can build on it. A tool that tells investigators where to look is worth a great deal, and it is a big step beyond reports that could only point to “adjacent land use.”

The paper also points naturally to the next questions.

First, timing. These outbreaks tend to hit in the fall, while the biggest runoff in the Salinas Valley comes in winter. Connecting the two is the logical next piece of the puzzle. Maybe it is pathogens settling into reservoir and canal sediment, as FDA found in 2018. Maybe it is cattle droppings on roadsides and dust blowing across fields at harvest. This study gives a strong foundation for answering that, and I am hopeful FDA’s California Longitudinal Study, which the authors say this work is meant to inform, gets us there.

Second, sharing. The paper’s map of farm scores leaves out exact livestock locations so as not to identify individual operations. That was a reasonable choice for a published study. Going forward, the real payoff will come when growers at the bottom of the hill can see who is at the top of it, ideally through cooperation between neighbors. Getting the right information to the right people is the same goal behind my outbreak firm disclosure petition to FDA.

Building on This Work

First, put the results to work. FDA can share them with the farms that score high, confidentially if need be, and the industry’s food safety programs should build hydrologic connectivity into preseason and preplant risk assessments. FDA already told Central Coast growers after the 2020 outbreak to treat the recurring outbreak strain as a reasonably foreseeable hazard. This model tells them where that hazard is most likely to come from.

Second, sample after big storms. The authors suggest targeting upstream water and sediment during and after runoff-producing rain. That is inexpensive compared with a recall, and far better than another outbreak.

Third, pair the map with access. A model can point to a pasture or a feedlot; investigators still need to be able to sample there. That is why I am drafting a citizen petition asking FDA for the ability to sample at animal feeding operations during outbreak investigations. This study makes the case for it better than I could have. Combine that authority with this kind of map and investigators would know exactly where to start.

Fourth, expand it. Yuma, Imperial and Santa Maria are obvious candidates. The authors rightly note that Salinas results cannot simply be copied to other regions, but the framework is built to be adapted, and that is one of its best features. The 2020 Leafy Greens STEC Action Plan already calls for looking beyond farm boundaries. This is what that looks like.

Fifth, bring livestock owners to the table. FDA asked them in its fall 2020 outbreak report to consider monitoring their herds for human-pathogenic E. coli and sharing results with neighboring growers. A map showing which pastures drain toward which farms makes that conversation a lot more concrete.

Water runs downhill. We have known that since the spinach outbreak. Now, thanks to Mehran Niazi, Francisco Garces-Vega, Kurt Nolte, Michael Mahovic and Jane Van Doren, we have a map that shows which way it runs. That is good science and good public health. Let’s use it.

Sources

Niazi M, Garces-Vega F, Nolte K, Mahovic M, Van Doren JM. Event-based surface-water connectivity modeling of leafy-green farm exposure to livestock-associated runoff. Journal of Food Protection (2026), in press. https://doi.org/10.1016/j.jfp.2026.100959

Jay MT, et al. Escherichia coli O157:H7 in feral swine near spinach fields and cattle, central California coast. Emerging Infectious Diseases 13:1908–1911 (2007). https://pmc.ncbi.nlm.nih.gov/articles/PMC2876768

NBC News/Reuters. Source of 2006 toxic U.S. spinach uncertain, study says (March 2007). https://www.nbcnews.com/id/wbna17757136

CDC. Multistate outbreak of E. coli O157:H7 infections linked to romaine lettuce (2018). https://www.cdc.gov/ecoli/2018/o157h7-04-18/index.html

FDA. FDA investigated multistate outbreak of E. coli O157:H7 infections linked to romaine lettuce from Yuma growing region (2018). https://www.fda.gov/food/outbreaks-foodborne-illness/fda-investigated-multistate-outbreak-e-coli-O157:H7-infections-linked-romaine-lettuce-yuma-growing

FDA. Investigation summary: Factors potentially contributing to the contamination of romaine lettuce implicated in the Fall 2018 multistate outbreak of E. coli O157:H7 (2019). https://www.fda.gov/food/outbreaks-foodborne-illness/investigation-summary-factors-potentially-contributing-contamination-romaine-lettuce-implicated-fall

CDC. Outbreak of E. coli infections linked to romaine lettuce (2019). https://www.cdc.gov/ecoli/2019/o157h7-11-19/index.html

FDA. Factors potentially contributing to the contamination of romaine lettuce implicated in the three outbreaks of E. coli O157:H7 during the fall of 2019 (2020). https://www.fda.gov/food/outbreaks-foodborne-illness/factors-potentially-contributing-contamination-romaine-lettuce-implicated-three-outbreaks-e-coli

FDA. Factors potentially contributing to the contamination of leafy greens implicated in the fall 2020 outbreak of E. coli O157:H7 (2021). https://www.fda.gov/food/outbreaks-foodborne-illness/factors-potentially-contributing-contamination-leafy-greens-implicated-fall-2020-outbreak-e-coli

Viazis S, et al. An overview of farm investigation findings associated with outbreaks of Shiga toxin-producing Escherichia coli infections linked to leafy greens: 2009–2021. Journal of Food Protection 88:100542 (2025). https://doi.org/10.1016/j.jfp.2025.100542

FDA. 2020 Leafy Greens STEC Action Plan. https://www.fda.gov/media/135855/download

FDA. California Longitudinal Study (2020–present). https://www.fda.gov/food/environmental-studies/california-longitudinal-study-2020-present

Saturday, November 7, 2026

Last Saturday’s edition was about things getting put in writing. This week (week 18) was about files getting closed. San Luis Obispo County closed its Brookshire Farms investigation Friday afternoon, five and a half hours after the farm reopened with free pumpkins, at fifteen sick, three children with kidney failure, one unique strain and no source. FDA closed six outbreak files in a single Wednesday update, 540 illnesses in all, and named romaine and parsley on the way out but not whose. CDC and FDA closed the Evergreen broccoli sprout outbreak at thirty-four sick and said the seed came from New Zealand through a domestic supplier they still will not name. Virginia went the other way and extended its oyster closure past the end date on its own order. And on Tuesday the voters closed one Congress and opened another, the part of the week that decides whether any of these files get reopened. That is where this week starts.

The county closed the Brookshire Farms file at 4:27 Friday afternoon, five and a half hours after the farm reopened at eleven with thousands of free pumpkins. San Luis Obispo County Public Health’s closing release counts five laboratory-confirmed E. coli O157 infections, all in people who visited the farm on October 17 or 18, plus ten symptomatic visitors between October 16 and 26, fifteen in all, up from thirteen on October 30. Whole genome sequencing matched three of the five to each other, a fourth result is pending and a fifth could not be sequenced; the matched strain did not closely resemble anything in the national databases, which the county reads as a common and unique source. Water, food and environmental samples came back negative for O157, other E. coli turned up in some of them, four exposures remain notable but unconfirmed, edible corn cups, pumpkins, hay and non-potable water, and no further testing is planned. The release also says no farm animals or petting zoo were open to the public during the outbreak, which takes the animals off the table and leaves the things children handle and eat. The words HUS, hospital and child do not appear in it; the three children with hemolytic uremic syndrome are in the October 30 release and nowhere since. The farm had announced the reopening on Instagram Thursday night on the strength of negative tests. I wrote Friday morning that a negative swab does not clear a farm and that the county should say where it stood before the gates opened, and Friday night that it closed the file five hours after they did without the pending sequence, a sampling summary or a word about the children. Will County has posted nothing on Bengtson’s Pumpkin Farm since October 23, with five confirmed cases and the animal, bedding and water samples at the state lab since the twenty-seventh, and Dallas County has still published nothing on the nine State Fair of Texas petting zoo cases beyond what its director told a television station on October 28. I put the national number on the record Friday: CDC estimates roughly a million illnesses a year from animal contact, and we are still building petting zoos without sinks. The firm represents a Brookshire Farms family.

FDA closed six outbreak files in one Wednesday update, 540 illnesses, and named two foods on the way out without saying whose. The CORE table stamped November 4 marks ref #1402 closed at 43 Salmonella I 4,[5],12:i:- cases with romaine lettuce named as the likely source twelve weeks after the row was posted, every environmental sample negative and the product past its shelf life; ref #1392 closed at 175 Cyclospora cases, adjusted down from 241, with parsley named; a new ref #1396 split off from it was posted and closed the same day at 66 cases with no food; and the 119-person Salmonella Newport file, the 99-person Oranienburg file and a 38-person Cyclospora file closed with nothing named. Of the 540, 218 got a food on the way out and 322 never will. No grower, brand or state appears for the romaine or the parsley; FDA’s explanation to Food Safety News is that the products were past their shelf life. I wrote Thursday about what twelve and fifteen weeks buys, and on Sunday about why a negative sample is not a dead end when sequencing, epidemiology and traceback are in hand. That leaves ten active rows. Five carry no food and 161 illnesses between them, down from 693 in ten rows last Saturday because the closures moved the cases to the other table, not because anyone found a food; the Salmonella Javiana file posted October 28 is at 65, up from 59, with traceback now checked; the Listeria file posted in September is at 51; and the chicken wrap file is now marked confirmed. CDC’s dashboard reads 28 active multistate investigations, sixteen Salmonella, seven STEC, three Listeria, two Campylobacter, and CDC posted no new notice this week.

CDC and FDA closed the Evergreen Fresh Sprouts outbreak Wednesday at thirty-four sick in six states, and the closing update names the country the seed came from but not the company that sold it. The final count is 34 ill, Washington 26, Montana 4, Idaho, Minnesota, Oregon and Utah one each, three hospitalized, onsets July 7 to September 11, ages two to 87. FDA’s advisory says its own sample from the domestic seed supplier matched the outbreak strain by whole genome sequencing, traceback showed the lot was imported from New Zealand, and the supplier recalled it; the advisory still calls it the seed supplier, while FDA’s enforcement report, as I noted two Saturdays ago, lists an Albany, Oregon seed company recalling two broccoli seed lots to consignees in eleven states and the United Kingdom. Three retail seed recalls off that lot, EAK, Eden Brothers and Gardeners Basics, are the only names a home sprouter will find. CDC’s final update drops Evergreen’s name from its text. The Justice Department told FDA on October 26 that its disclosure rule is its own and can be rewritten; the sprout file closed nine days later written exactly as it would have been before. The firm represents a Washington resident sickened in this outbreak.

Virginia extended the Little Wicomico River closure past the end date on its own order, and the order itself still reads November 7. VDH’s outbreak page, updated Friday, says that effective November 8 the harvest-area closure extends beyond the thirty-day emergency closure because of the positives on October 13 and 21, and that reopening will take three consecutive clean rounds. Round two, collected October 21, found Salmonella in three of the four company leases, and its whole genome sequencing is still pending sixteen days later; a third round came out of the cages November 2 and water and oyster samples from the public grounds went to the University of North Carolina on November 4, neither with results posted. The October 26 public-ground samples were negative. Emergency Condemnation #010A, the legal instrument, was last updated October 8 and still says October 9 through November 7, with no amended order on the page the day before it lapses. CDC is unchanged at 28 sick in 13 states, five hospitalized, as of October 30, and Virginia counts three residents. The company’s notice still does not say Salmonella.

A Yakima County dairy worker died of H5N1 on Sunday, the first bird flu death in the United States this year, and Washington’s herd count reached forty in three weeks. The Yakima Health District announced the death November 1: a resident with underlying conditions, exposed at a dairy with infected cattle. The county has three confirmed human cases, none since, and its herd table runs from two confirmations on October 8 to thirty-nine by October 30; WSDA confirmed one Benton County herd on October 26, making forty. CDC read the virus’s genetic sequence and found nothing suggesting easier spread, and says the risk to the public is low. USDA confirmed Indiana’s first two dairy herds November 2 through the national milk testing program, the twenty-first state since March 2024. Yakima’s page says pasteurized milk is safe and raw milk is not; Washington still licenses raw milk for retail sale, and nothing public says whether licensed raw milk from the two counties is being tested for the virus. Food Poison Journal has what Washington families need to know.

The recall ledger was light on pathogens and heavy on things that are not supposed to be in food at all. Brad’s Raw Chips recalled about 25,000 bags of Crunchy Kale and Veggie Chips Kale nationwide on November 3 for thallium, a heavy metal, after the California Department of Public Health traced one person’s elevated thallium level to the chips; the lots shipped October through December 2025 and the best-by dates run to February 2027. FDA added a twenty-first product, Wise Wife ground cinnamon at 2.118 parts per million of lead, to its elevated-lead cinnamon list. Vitality Works’ recall of one lot of 365 Whole Foods Market ashwagandha capsules for Salmonella, on sale eleven months in thirty states, reached FDA’s page October 30. Phyllis Entis at eFoodAlert logged no United States alerts from October 31 through November 2 and on November 4 caught the one pathogen recall nobody announced: Arrow Reliance, which makes Darwin’s Natural Pet Products, pulled two lots of Natural Selections chicken cat food for Salmonella by emailing consignees, with no public notice, nine days after FDA’s October 26 advisory on the same brand. She also carried FDA’s warning letter to La Cabana Food Distributing of Oceanside, California, posted November 3, with Listeria monocytogenes on a scale, cart wheels and the floor of a salsa plant and no environmental testing program. FSIS posted one public health alert, for Betterballen frozen snacks with undeclared sesame and anchovy, and no pathogen recall; its site would not load this week and that item comes from Food Safety News. West Virginia has said nothing since its October 29 advisory on 41 Salmonella Typhimurium cases with no source.

Perrigo told investors Wednesday that infant formula was consuming cash, and the congresswoman who wrote the botulism testing bill said parents deserve answers before the ByHeart founders buy the business. On its November 4 earnings call Perrigo priced the sale of its formula business to MeridianMade at up to $175 million, $40 million at closing and $135 million in milestones through 2030, moved formula into discontinued operations, and did not mention ByHeart. Rep. Rosa DeLauro’s October 30 statement says families still do not know how, when or where their children’s formula became contaminated, and that she worries about consolidation. FDA’s formula testing page, updated October 29, now covers 312 samples for four heavy metals, thirty PFAS, 318 pesticides and twenty-one phthalates, and still has no row for the organism that put 52 babies in the hospital. I wrote the committees a third time on Wednesday, the morning after the election, with five legislative weeks left and an offer to fly the families to Washington, and the Clostridium botulinum testing petition goes to FDA this week with Stop Foodborne Illness and the parents of twenty-three infants as co-petitioners. The supply chain behind Nara and the long version of ByHeart are both on the blog.

The voters handed both chambers to the Democrats on Tuesday, and the people who chair the committees that fund and oversee FDA and USDA change in January. CBS has Democrats at a minimum of 54 Senate seats with Alaska and Mississippi uncalled and on pace for roughly 255 House seats. Frank Pallone has asked for the Energy and Commerce gavel, Rosa DeLauro is in line for Appropriations, Bernie Sanders is expected back at HELP, and both Agriculture committees need new Democratic leaders because Angie Craig and Amy Klobuchar are leaving. Congress returns November 9 with the continuing resolution, and its Section 780 bar on traceability enforcement before July 20, 2028, running to December 11. I published fifteen things the lame duck can do and a Project 2027 working paper this week. FDA held its traceability public meeting Friday: twenty-nine registered commenters, from FMI to Oceana to a retired field epidemiologist, with neither the acting commissioner nor the acting head of the Human Foods Program on the agenda; comments on the lot-code flexibilities are due December 6, and no account of what was said had been published by Saturday morning. FSIS said in its October 30 constituent update that it will swab food-contact and non-food-contact surfaces in ready-to-eat plants nationwide starting January 2027. Judge Chhabria refused Wednesday to lift his stay on USDA’s relocation notices. Consumer Reports’ November 6 deadline for FDA to answer its eight questions on the Cyclospora outbreak passed without a public reply that I could find. And USDA’s romaine shipping study got my answer at Food Safety News: truck routes did not put manure on lettuce, the field did.

That’s the week — a county that closed its file five hours after the farm reopened, with fifteen sick, three children in kidney failure and no source; FDA closing six files in an afternoon and naming romaine and parsley but not whose; a sprout outbreak closed with the seed traced to New Zealand and the supplier still unnamed; an oyster closure extended past its own order; a dairy worker dead of bird flu in Yakima County; and an election that decides who gets to ask about any of it. Next week brings the pending Brookshire sequence, Illinois’s animal and water results, Virginia’s third round, Congress back in session and, if it posts, the botulism petition on FDA’s docket. 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.

At 4:27 this afternoon, five and a half hours after Brookshire Farms reopened its pumpkin patch on Los Osos Valley Road with thousands of free pumpkins, the County of San Luis Obispo Department of Public Health published the release I asked for this morning. The county is wrapping up its investigation of the Shiga toxin-producing E. coli (STEC) outbreak linked to the farm. It says there is no ongoing public health threat, and that no source was identified. The count stands at five laboratory-confirmed E. coli O157 infections and ten suspect cases, fifteen in all, up from the thirteen the county reported on October 30.

What the county found.

  • All five confirmed cases visited the farm on October 17 or 18.
  • Genetic testing matched three of the five to each other; a fourth result is pending and a fifth could not be run.
  • The matched strains do not closely resemble anything in the national databases. In the county’s words, the genetics “strongly suggest exposure to a common and unique source.”
  • Ten more people who visited between October 16 and 26 and became ill are suspect cases; the county will say no more about them.
  • The timing of the visits fits the incubation period.
  • Water, food and other environmental samples did not find O157. Other bacteria, including other strains of E. coli, were found in some samples; those results did not identify the source.
  • Sampling was limited by the number of possible sources, the absence of items left to test, the delay between exposure and sampling, and “other factors.” No further testing is planned.
  • Among the exposures shared by the infected, four “remain notable”: edible corn cups, which the release’s Spanish text describes as edible containers made of corn; pumpkins; hay; and non-potable water sources.
  • No farm animals or petting zoo were available to the public or operating on the premises.
DateConfirmed O157Other casesHUSSource
Oct. 275—not statedfarm suspected
Oct. 3058 under investigation3“remains under investigation”
Nov. 65 (3 matched by genetic testing, 1 pending, 1 not testable)10 suspectnot mentionednot identified; investigation closing

Sources: San Luis Obispo County Public Health releases of October 27, October 30 and November 6, 2026. Check: five confirmed plus ten suspect equals fifteen.

No animals. That changes the picture, including mine.

In my posts of October 27 and 28 I grouped Brookshire with Bengtson’s Pumpkin Farm in Illinois and the petting zoo at the State Fair of Texas as E. coli investigations tied to animals at fall attractions, because that is what O157 at a farm usually is, and because a U-pick directory lists farm animals at Brookshire. This morning I asked whether the animal area would reopen with the pumpkin patch. The county’s answer is that none was open to the public during this outbreak. I take the county at its word; a reasonable assumption on October 27 is off the table today.

It matters because the prevention playbook for farm outbreaks is built around animal contact: sinks at the exit of the pens, no food inside, toddlers kept out. Take the animals away and the county’s list is produce, hay, food served in an edible cup, and water nobody was supposed to drink. Three of the four are things visitors put in their mouths or handled before eating; the fourth is what children climb on, and Brookshire advertised hayrides, a hay climb, a hay maze and duck races. The county does not rank the four, and it is careful to say that none was confirmed.

One line deserves a second look: other strains of E. coli were found in some samples. Generic E. coli is the standard indicator of fecal contamination in water and on produce. Finding it does not tell you what carried O157, which is exactly what the county says. It does tell you the farm had detectable fecal contamination somewhere when sampled, and that the sampling could not take the next step.

What the county did not say.

The children. On October 30 the county counted three HUS cases among the five confirmed, and two days earlier its spokesman told ABC News that the three children among the five were seriously ill in pediatric intensive care. Today’s release does not contain the words hemolytic uremic syndrome, hospital or child; it refers to “the families affected.” Three weeks after exposure, the public does not know whether those children are home; privacy allows a sentence that says so.

The fourth test. The county is closing the file with one genetic result still pending. If it matches, four of the four testable cases match. That result should be published when it arrives, not filed.

Whose tests. The farm told its Instagram followers on October 30 that it was waiting on the county’s test results, and on November 5 that “recent tests and environmental samples” had come back negative for the outbreak strain. Today’s release describes the county’s testing and its limits, including the delay between exposure and sampling. It does not say whether the farm’s negative is the county’s negative, when the samples were collected, or how many there were.

The reopening. The release does not say the county reviewed the reopening or agreed to it, only that the risk of additional exposure “has passed.” The gates opened at 11 a.m.; the release went up at 4:27 p.m. The order should have been reversed, which is a complaint about sequence, not about the work.

The ten. No additional information will be provided on the ten suspect cases: not whether any were tested, hospitalized, or children.

“Pumpkins have always been safe” is the farm’s conclusion, not the county’s.

The farm said yesterday that its pumpkins “have always been safe.” The county today lists pumpkins among the notable shared exposures, in the same sentence that says no source was identified. Both can be true in the narrow sense that no pumpkin tested positive. But the county’s October 30 advice to scrub pumpkins and wash hands after handling them has not been withdrawn, and a clean result on whatever was sampled in November says little about what was handled on October 17 and 18. The county explains why in its own words: contamination is “often temporary and difficult to detect.” If you took home a free pumpkin today, wash it, the surfaces it touched and your hands, as the county advised last week.

No source is not the same as no answer.

Here is what today’s release establishes: five people with laboratory-confirmed O157, one farm in common, visits on two days, a strain that matches itself and nothing else in the national databases, ten more sick visitors from the same eleven days, and onset timing that fits. The county says the clinical and epidemiological information “continues to point to a short-lived shared exposure.” In more than thirty years of this work, that is how most outbreaks are solved, by epidemiology, not by finding the organism in a swab three weeks later. The vehicle is unknown. The venue is not.

That is also how it works in a courtroom. A negative environmental sample is evidence of what was on the farm the day it was collected, not on October 17, and it does not subtract a single case from the count. We represent one family in this outbreak, as Food Poison Journal noted on October 28.

What should happen next.

The county should publish the fourth genetic result when it comes in, post a one-page summary of what was sampled, when and what grew, say whether the November 5 results are its own, and tell the public, at whatever level privacy allows, whether the three children with HUS are home. None of that requires a name.

Every pumpkin patch without animals should read the county’s four-item list as a checklist: potable water in anything a child’s hands go into, sinks at the hay attractions and the food line, not only at animal pens, and food sold under permit and inspection. The last bullet of the county’s safety advice is addressed not to visitors but to anyone “selling or otherwise providing food to the public.” The release does not say why that line is there.

Anyone who visited Brookshire between October 16 and 26 and had diarrhea, especially bloody diarrhea, should still see a doctor, mention the farm and ask for Shiga toxin testing. HUS is a medical emergency: decreased urination, unusual tiredness and pallor in a child who has had diarrhea are reasons to go in, not to wait.

Fifteen people, three children in kidney failure, one unique strain, and a file that closes with a list of four things it might have been. The county did careful work and wrote most of it down. Closed is not the same as answered.

I have been doing this for more than thirty years, and every fall feels the same. The leaves turn, the pumpkin patches open, the state fairs roll into town, and a few weeks later the phone rings. A child is in the ICU. Her kidneys are failing. The family went to a petting zoo.

This fall it is happening again, in three states at once. And CDC just put a number on the problem that ought to stop everyone in their tracks.

The Number: About One Million

A new study in CDC’s journal Emerging Infectious Diseases, Estimated Illness Linked to Animal Contact for 7 Pathogens, United States, 2019, updates the government’s estimate of how many Americans get sick from touching animals or the places they live. The answer: roughly one million illnesses a year, an estimated 9,592 hospitalizations and 236 deaths.

The researchers looked at Campylobacter, Cryptosporidium, Shiga toxin-producing E. coli (STEC), Listeria, nontyphoidal Salmonella and Yersinia, plus Toxoplasma for hospitalizations and deaths. They adjusted for the fact that most sick people never see a doctor and most who do never get a stool test. Those six bacterial and parasitic pathogens alone caused an estimated 995,022 illnesses linked to animal contact around 2019.

Campylobacter led in illnesses and hospitalizations, with an estimated 3,601 hospitalizations a year, followed by Salmonella at 2,415. Campylobacter also passed Salmonella in deaths. The biggest killer was Toxoplasma, the parasite spread through cat feces, at an estimated 91 deaths a year.

Now the number that matters at a petting zoo. STEC, the E. coli that causes hemolytic uremic syndrome, accounted for an estimated 134,738 animal-contact illnesses, 960 hospitalizations and 19 deaths a year. Most of that was non-O157 E. coli: an estimated 117,477 illnesses, more than ten times the 10,097 CDC estimated in 2012. E. coli O157 went from 5,960 illnesses to 17,261.

The study’s tables show how big a slice of the problem animals are. Based on a structured survey of experts, CDC attributes about 12 percent of domestically acquired E. coli O157 illnesses and 22 percent of non-O157 STEC illnesses to animal contact. And O157 is not a stomach bug you sleep off: in FoodNet surveillance, 43 percent of lab-confirmed O157 patients were hospitalized.

The 2012 estimate, published by Hale and colleagues, put all animal-contact illness at about 450,000 cases, 5,000 hospitalizations and roughly 70 deaths a year. Hospitalizations nearly doubled. Deaths more than tripled, although Toxoplasma was not counted last time. Take it out and deaths still roughly doubled. The authors are careful to say the methods differ, so the two studies are not a straight comparison.

They are also honest about why the numbers jumped. Since 2013, labs have moved to PCR-based gastrointestinal panels that catch infections old culture methods missed, especially non-O157 STEC and Campylobacter. The authors think much of the increase reflects better detection. I do not find that comforting. It means these kids and adults were getting sick all along, and we were not counting them.

For some perspective, CDC’s 2019 foodborne illness burden estimates put STEC from food at about 357,000 illnesses, 3,150 hospitalizations and 66 deaths a year. So for every three Americans who get STEC from food, more than one more gets it from an animal or its environment. Animal contact is not a footnote to food safety. It is food safety’s next-door neighbor, and it is sharing the same fairgrounds.

Three Outbreaks, One Month

The timing of the CDC paper could not be more on point.

In Illinois, health officials are investigating five confirmed E. coli cases tied to Bengtson’s Pumpkin Farm in Homer Glen. As I reported last week, the patients range from four to thirty-eight years old, and four were hospitalized.

In California, Brookshire Farms in San Luis Obispo has closed after five lab-confirmed E. coli O157 infections and eight more people with matching symptoms who visited between October 16 and 26. Three of the confirmed cases have hemolytic uremic syndrome, the kidney failure complication that keeps me up at night.

And in Texas, Dallas County is looking at nine suspected cases in children and families who had been to the State Fair of Texas. According to local reporting, the common thread is a visit to the livestock and petting zoo areas at Fair Park. Doctors noticed it first. Kids kept showing up sick, and parents kept mentioning the fair.

I wrote about all three when they broke. Here is what I said then and what I will say again: none of this is new, and none of it is a mystery.

We Have Seen This Movie Before

Just one year ago, the Arizona State Fair. The Maricopa County outbreak report found 18 people sick, 11 hospitalized and six with HUS, most of them children. Thirteen of the 18 had been at the fair on October 26, 2025, the last day. Here is the detail that matters: of the sick people who visited the animal areas, only five of eleven said there were handwashing stations. Eight said there was hand sanitizer.

Hand sanitizer is not handwashing. A study in the Journal of the American Veterinary Medical Associationfound that alcohol-based sanitizer was not protective against STEC in a fair outbreak. Manure on little hands does not care about your Purell.

Go back further. The 2004 North Carolina State Fair: about 108 people sick, 15 with HUS, traced to two commercial petting zoos. Investigators found the outbreak strain in shavings from a baby stroller and on visitors’ shoes.

2005: three separate outbreaks at fairs, festivals and petting zoos in North Carolina, Florida and Arizona, written up together by CDC.

2019: the San Diego County Fair, where a two-year-old boy died and other children were sickened after visiting the animal areas.

2022: a farm summer camp in Tennessee, written up in MMWR, where the camp operators said they knew the national guidance and had handwashing stations. Kids still got sick.

2023: the Appalachian Fairgrounds in Tennessee, where more than 64 elementary school children on field trips were infected with E. coli O157:H7. Five children, ages 15 months to 7 years, went into kidney failure and needed dialysis.

I could keep going. CDC’s own Animal Contact Outbreak Surveillance System counted 557 animal-contact outbreaks from 2009 to 2021, with 14,377 illnesses, 2,656 hospitalizations and 22 deaths. Fairs and festivals accounted for 36 of the single-location outbreaks, petting zoos 28, and farms and dairies 89. And those are just the outbreaks somebody found and reported.

Outbreaks are the tip of it. A Minnesota study I cited in my earlier post found that 28 percent of STEC patients not tied to any outbreak had visited a fair, petting zoo, agritourism farm or similar venue. The million-a-year number from CDC is mostly made of those quiet, unconnected cases.

Why the Animals, Why the Kids

Cattle, goats and sheep carry STEC in their guts and look perfectly healthy doing it. They shed it in manure. It ends up on their coats, the bedding, the fence rails, the dirt and the stroller wheels. A new analysis of CDC outbreak data found that when outbreaks were tied to sheep or goats, nearly half were caused by STEC.

Then you add small children. They touch everything. They put their hands in their mouths. They sit on the ground. The JAVMA review noted that in one outbreak, illness in kids under six was tied to touching manure, falling or sitting on the ground, and thumb-sucking or pacifiers. Another fair study found the risk of illness went up 51 percent for every extra day a child attended.

And then, too often, we put a corn dog stand at the exit. In the 2004 North Carolina outbreak, investigators flagged a food vendor positioned right next to the petting zoo exit. You could not design a better fecal-oral pipeline if you tried.

The Fix Has Been Written Down for Years

This is the part that makes me crazy. We know how to prevent this. CDC and the National Association of State Public Health Veterinarians publish the Animal Contact Compendium, and CDC has a plain-language page on staying healthy at animal exhibits. The recommendations are not complicated:

Put real handwashing stations, with soap and running water, at every exit. Not sanitizer alone. Real sinks.

Keep food and drink out of animal areas, and keep animal areas away from food. Separate entrances and exits.

Keep strollers, sippy cups, pacifiers and toys out of the pens.

Supervise. Put a person at the exit making sure kids wash.

Keep sick animals, especially those with diarrhea, away from the public. And clean the place.

The problem is that almost none of this is required. There is no federal law addressing disease risk at animal exhibits. It is a patchwork of state rules, county permits and voluntary guidance.

After Arizona, a bill was introduced to require handwashing stations, warning signs and a monitor at animal exhibits. It died in the legislature. To their credit, the Maricopa County Fair dropped its petting zoo, and the Arizona State Fair dropped direct animal contact on its own. That is something. But it should not take a cluster of kids on dialysis for each fair board to figure it out one at a time.

Groups like Parker’s Promise, started after a child developed HUS from an agricultural summer camp, are pushing for exactly the kind of mandatory rules that should already exist. They are right.

What I Want to See

Here is my ask, and it is not radical. Every state should require a permit for any public animal exhibit. No permit without soap, running water and a monitor at every exit. No food sales within a set distance of an animal area. No sanitizer-only setups. Mandatory signage that tells parents, in plain words, that children under five are at the highest risk of HUS. And when an outbreak happens, health departments should name the venue right away, not weeks later.

To parents: you do not have to skip the pumpkin patch. Just wash your kids’ hands with soap and water every single time they leave the animals, keep the snacks away from the pens, and if your child gets bloody diarrhea in the week after a visit, call the doctor and tell them about the animals. Do not give antidiarrheals, and do not push for antibiotics. Both can make HUS more likely.

CDC just told us a million Americans a year get sick from animal contact. Ten thousand end up in the hospital. Some of them die. A lot of them are kids who went to the fair.

We have the playbook. We have had it for twenty years. What we do not have is the will to make it the law.

Sources

Medford R, et al. Estimated Illness Linked to Animal Contact for 7 Pathogens, United States, 2019. Emerging Infectious Diseases 32(12), Dec. 2026 (early release, Oct. 27, 2026). https://wwwnc.cdc.gov/eid/article/32/12/26-0787_article

Hale CR, et al. Estimates of Enteric Illness Attributable to Contact With Animals and Their Environments in the United States. Clinical Infectious Diseases, 2012. https://academic.oup.com/cid/article-abstract/54/suppl_5/S472/432063

CDC. Enteric Disease Outbreaks Associated with Animal Contact — ACOSS, United States, 2009–2021. MMWR Surveillance Summaries 74(3). https://www.cdc.gov/mmwr/volumes/74/ss/ss7403a1.htm

CDC. Estimates: Burden of Foodborne Illness in the United States (circa 2019). https://www.cdc.gov/food-safety/php/data-research/foodborne-illness-burden/index.html

Comparison of zoonotic enteric disease outbreaks associated with ruminant and non-ruminant animal contact — United States, 2013–2021. Frontiers in Public Health, 2026. https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2026.1845080/full

KWBU. Investigations underway for potential E. coli outbreak from State Fair petting zoo. Oct. 30, 2026. https://www.kwbu.org/news-from-across-texas/2026-10-30/investigations-underway-for-potential-e-coli-outbreak-from-state-fair-petting-zoo

KSST Radio. Health Officials Investigate Potential E. Coli Outbreak Linked to State Fair of Texas Petting Zoo. Oct. 29, 2026. https://www.ksstradio.com/2026/10/health-officials-investigate-potential-e-coli-outbreak-linked-to-state-fair-of-texas-petting-zoo/

Food Poisoning News. Five E. Coli Cases Possibly Linked to Bengtson’s Pumpkin Farm in Illinois. Oct. 26, 2026. https://www.foodpoisoningnews.com/five-e-coli-cases-possibly-linked-to-bengtsons-pumpkin-farm-in-illinois/

Marler Blog. Weekend Wrap-Up: Twenty-Seven Children and Adults Sick After Two Pumpkin Patches and a State Fair Petting Zoo. https://marlerblog.com/case-news/weekend-wrap-up-twenty-seven-children-and-adults-sick-after-two-pumpkin-patches-and-a-state-fair-petting-zoo-three-of-them-with-kidney-failure-the-justice-department-told-fda-in-writing-that-its-ow

Marler Blog. Pumpkin Patches, Petting Zoos and State Fairs: Three New E. coli Investigations. https://www.marlerblog.com/case-news/pumpkin-patches-petting-zoos-and-state-fairs-three-new-e-coli-investigations-twenty-five-years-of-outbreaks-just-like-them-and-the-prevention-playbook-everyone-already-has/

Maricopa County Department of Public Health. 2025 STEC Outbreak Report (Arizona State Fair). https://media.ronsimonassociates.com/reports/maricopa-county-2025-stec-outbreak-report.pdf

Food Poisoning News. Arizona State Fair Petting Zoo Linked to 2025 E. coli Outbreak That Sickened 18, Hospitalized 11 Children. Oct. 14, 2026. https://www.foodpoisoningnews.com/arizona-state-fair-petting-zoo-linked-to-2025-e-coli-outbreak-that-sickened-18-hospitalized-11-children/

Arizona Legislature. SB 1082 (2026). https://legiscan.com/AZ/text/SB1082/2026

Arizona Republic. Maricopa County Fair: no petting zoos following E. coli outbreak. Apr. 4, 2026. https://www.azcentral.com/story/news/local/arizona-health/2026/04/04/maricopa-county-fair-no-petting-zoos-following-e-coli-outbreak-arizona-state-fair/89461033007/

CIDRAP. North Carolina E. coli outbreak traced to petting zoo (2004). https://www.cidrap.umn.edu/e-coli/north-carolina-e-coli-outbreak-traced-petting-zoo

CDC. Outbreaks of E. coli O157:H7 Associated with Petting Zoos — North Carolina, Florida, and Arizona, 2004 and 2005. MMWR 54(50). https://www.cdc.gov/mmwr/preview/mmwrhtml/mm5450a1.htm

San Diego Union-Tribune. E. coli death, infections spark claims against San Diego fair board. July 30, 2019. https://www.sandiegouniontribune.com/2019/07/30/e-coli-death-infections-spark-claims-against-san-diego-fair-board-2/

Tennessee Department of Health. STEC O157:H7 outbreak, Appalachian Fairgrounds (2023). https://healthdata.tn.gov/Infectious-Disease/Shiga-toxin-producing-Escherichia-coli-O157-H7-out/8crc-mqrk/about_data

Notes from the Field: An Outbreak of STEC O157:H7 Associated with a Farming Camp — Tennessee, 2022. MMWR. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10360649/

Direct and indirect zoonotic transmission of Shiga toxin–producing E. coli. JAVMA 232(6), 2008. https://avmajournals.avma.org/view/journals/javma/232/6/javma.232.6.848.xml

CDC and NASPHV. Compendium of Measures to Prevent Disease Associated with Animals in Public Settings. https://stacks.cdc.gov/view/cdc/133062

CDC. Staying Healthy at Animal Exhibits. https://www.cdc.gov/healthy-pets/about/staying-healthy-at-animal-exhibits.html

Parker’s Promise. https://parkers-promise.com/

Brookshire Farms, the Los Osos Valley Road pumpkin patch in San Luis Obispo that closed on October 27 after five visitors were confirmed infected with E. coli O157, reopens today, Friday, November 6, 2026, at 11 a.m. with thousands of free pumpkins, on the strength of an Instagram post saying its “recent tests and environmental samples” came back negative for the outbreak strain, as KSBY reported overnight. San Luis Obispo County Public Health, whose last release on October 30 counted five confirmed cases, eight under investigation and three children with hemolytic uremic syndrome, has published nothing since.

The farm’s two Instagram posts, six days apart, say different things.

On October 30 the farm told its followers, in a post KSBY quoted that afternoon: “Safety is our main concern.” “We are waiting on the test results from the county health department.” “We have opted to remain closed for the remainder of the pumpkin season until test results have been determined.” No further comment until the results were in.

On November 5 the farm posted again. As KSBY summarized it, recent tests and environmental samples taken from the farm came back negative for the E. coli outbreak strain, the pumpkins “have always been safe,” the gates open Friday at 11 a.m., and the community that sent kind messages during the closure gets another chance at a pumpkin, free. The farm’s own website still reads “We are closed for the season” this morning. The October 30 post said the results belonged to the county. The November 5 post does not say who ran the tests, what was sampled, how many samples there were, when they were collected, or whether the county agrees with the conclusion.

The county’s last word is October 30: thirteen sick, three with HUS, farm closed as a precaution.

The county’s numbers have not moved in a week because the county has not spoken in a week. Its October 30 release said the farm “remains closed as a precaution,” thanked it for cooperating, and said the source of the infections on the farm “remains under investigation.” Here is the record as it stands.

DateSourceWhat it said
Oct. 27SLO County Public HealthFive lab-confirmed STEC cases; Brookshire Farms suspected as a potential source; exposure window Oct. 16 to 26; farm closed voluntarily
Oct. 30SLO County Public HealthFive confirmed E. coli O157; eight more under investigation; three with HUS; one more hospitalized and released; farm “remains closed as a precaution”; source “remains under investigation”
Oct. 30Brookshire Farms (Instagram)Waiting on county test results; closed for the rest of the season until results are determined
Nov. 5Brookshire Farms (Instagram)Tests and environmental samples negative for the outbreak strain; pumpkins always safe; reopening Nov. 6 at 11 a.m.; free pumpkins
Nov. 6SLO County Public HealthNo release since Oct. 30

Sources: San Luis Obispo County Public Health releases of October 27 and 30, 2026, and KSBY reports of October 30 and November 6, 2026. Check: five confirmed plus eight under investigation equals thirteen; three HUS cases among five confirmed is 60 percent.

Three HUS cases out of five confirmed infections is a ratio I wrote about last week, and it has not improved with time. In most O157 outbreaks a small minority of confirmed cases go on to HUS, and the ones who do are usually children. Here it is three of five, and the eight pending cases have not been resolved in public one way or the other.

A negative environmental sample does not clear a farm.

It would be good news if the samples are negative. It is not the same as an answer. The exposure window closed on October 26. Any sample taken after the farm shut on October 27 describes the farm on the day it was swabbed, not the farm on the days thirteen people walked through it. Cattle, goats and sheep that carry E. coliO157 shed it intermittently, and the animal that was shedding on October 18 may not be shedding on November 3. CDC’s own guidance for farm animals and petting zoos says it plainly: animals can still spread germs even if they look healthy and clean. Nothing about a negative swab changes what the epidemiology already found: five people with lab-confirmed O157, one farm in common, an eleven-day window.

There is also the question of what “negative for the outbreak strain” means in practice. A rinse of a few pumpkins, a swab of a hand-washing sink, a sample of bedding, a fecal sample from each animal in the contact area and a water sample from the trough are five different things. The county knows which of them were collected. The public does not.

What the November 5 statement does not say.

It does not say who tested, what was tested, how many samples, on what dates, or by which laboratory. It does not say whether the county reviewed the results or cleared the farm to open. It does not say whether the animal area reopens with the pumpkin patch, or whether anything about hand-washing stations, signage or animal access has changed since October. It does not mention the thirteen people, the three children in kidney failure, or the eight cases still being worked. It says the pumpkins were always safe, which was the county’s first line of advice too, and it leaves the animals out of the sentence.

The county should publish the results before the gates open, and say whether the animals are back.

None of this is a complaint about the people doing the work. The county epidemiologists found this cluster, named the farm within a day, widened the window, and counted honestly. The gap is at the end of the process, where the results of the investigation they ran are being announced by the business they were investigating, on Instagram, at night, with a giveaway attached. The fix is a short release from the Health Officer today that says four things: what was sampled and when, what the results were, whether Public Health agrees the farm can reopen, and whether the animal-contact area is part of the reopening. If the county’s position is that the source was never found, it should say that too. Families reading an Instagram post at 11 a.m. deserve to know whether the county stands behind it.

If you go today, wash your hands and watch the calendar.

Wash hands with soap and running water after touching animals, fences, bedding or produce, and before eating. Keep food and drinks out of any animal area. Keep children under five out of animal contact areas. Anyone who develops stomach cramps and diarrhea, especially bloody diarrhea, three to ten days after a visit should see a doctor, mention the farm, and ask that the stool be tested for Shiga toxin-producing E. coli. HUS is a medical emergency; decreased urination, unusual tiredness and pallor in a child who has had diarrhea are reasons to go in, not to wait.

We represent one family in this outbreak, which Food Poison Journal noted on October 28. The part of this worth stopping on is simpler than any lawsuit. Three children went into kidney failure after a trip to a pumpkin patch. The agency that investigated has the results. The farm has announced them. The county should.

A Rare Disease With a Strange Mechanism

Infant botulism is rare, and it does not work like the botulism most people have heard of. In foodborne botulism, an adult eats food, often badly home-canned vegetables, in which the bacterium Clostridium botulinum has already produced its toxin. Botulinum toxin is one of the most lethal substances known. It blocks the signals between nerves and muscles, and it paralyzes from the top down.

In infant botulism, the baby does not swallow the toxin. The baby swallows the spores of the bacterium, which are tough, dormant forms that survive drying, heat, and time. In an adult’s gut, the spores pass harmlessly through, crowded out by mature intestinal bacteria. In a young infant’s gut, which has not yet developed those defenses, the spores can germinate and multiply. The baby’s own intestine becomes the place where the toxin is made.

The signs come slowly and then quickly. Constipation is often first. Then come weak sucking, poor feeding, a soft or altered cry, drooping eyelids, a flat face, and a body that goes limp, the classic “floppy baby.” In the worst cases, the muscles that control breathing fail, and the baby needs a ventilator. Because the early signs look like so many ordinary infant problems, the disease is easy to miss.

There is a treatment, and it works. BabyBIG, an antitoxin made from human plasma, neutralizes toxin still circulating in the blood. In a California trial, it cut the average hospital stay from 5.7 weeks to 2.6. But it cannot undo the toxin already bound to nerves. Those nerve endings have to regrow, and that takes weeks. Babies need feeding tubes, intensive care monitoring, and often months of physical, occupational, and feeding therapy afterward. In the ByHeart outbreak, CDC’s preliminary data found that about two-thirds of the infants evaluated needed physical therapy or feeding support after they left the hospital.

Most cases of infant botulism are isolated. Spores live in soil and dust everywhere, and in most cases no source is ever found. The one well-known food source is honey, which is why pediatricians tell parents never to give it to a baby under a year old. For decades, surveys of commercial dried milk and infant formula almost never found C. botulinum.

That is why, when the outbreak investigation was over, the numbers were so striking. Public health officials counted 48 infants, 28 confirmed and 20 probable cases, in seventeen states, from Arizona to Rhode Island. Every one of the 48 babies was hospitalized. No baby died. The earliest illness onset was December 24, 2023. The last was November 29, 2025.

“The More You Control the Process”

ByHeart was founded in 2016 by a brother and sister, Ron Belldegrun and Mia Funt. Belldegrun, who came from venture capital and hedge funds, became chief executive. Funt, who came from marketing and business development, became president. They raised hundreds of millions of dollars from investors. In the spring of 2025, a financing round valued the company at more than $900 million. Before the recall, ByHeart sold roughly 200,000 cans a month at about $42 a can.

The founders built their brand on control. Instead of hiring an established contract manufacturer, ByHeart bought its own plants: a factory in Reading, Pennsylvania, converted from toddler food to infant formula for $21.6 million; a plant in Allerton, Iowa, bought in January 2023 from a subsidiary of Dairy Farmers of America; and a blending and canning plant in Portland, Oregon.

They talked about that control constantly. In June 2021, before ByHeart sold its first can, Belldegrun told Fast Company: “We believe that the more you control of the process, the more you control the quality of the formula.” At launch in March 2022, Funt told Forbes that ByHeart had “acquired our own manufacturing facility, sourced every ingredient, and applied rigorous quality standards, from farm to formula,” and that she was proud the company had kept its promise of “no shortcuts and no compromises.”

The founders made the same promises directly to parents on ByHeart’s Instagram account. In December 2021, the company posted a quote from Funt telling parents that “every parent deserves to feel confident when feeding baby.” In June 2022, standing in the Reading plant with Pennsylvania’s governor, Funt said the founders had set out “to own our supply chain — own and oversee our supply chain,” which “meant acquiring our manufacturing facility, hand-selecting every ingredient.” In a March 2025 video, she told parents that it took “nine months to go through our rigorous ingredient selection process” and that ByHeart ran “over a thousand tests on our ingredients and throughout every step of the process” for “contaminants and heavy metals and bacteria.” “What really matters,” she said, “is the testing protocols.”

In June 2025, Belldegrun made the same point on camera: “We own our supply chain, we own our manufacturing, and we directly handpick all of our ingredients.” He promised “incredibly rigorous testing — like, every step, from the ingredients, to while we’re running, to before we release product.” He also explained why the stakes were so high. For a baby who is not breastfed, he said, formula “is the only nutrition that a baby consumes in that first year of life, when all of their systems are developing.”

That idea, that formula is a baby’s sole source of nutrition, is the reason infant formula is held to a higher standard than any other food. It is also the foundation of the families’ cases.

The Ingredient

What set ByHeart apart was its milk. Most formulas are made from skim milk, with vegetable oils added for fat. ByHeart used organic, grass-fed whole milk. Belldegrun promoted the choice in a June 2025 op-ed and took credit for it in his video: “Prior to our work, you could only use skim milk here.”

The whole milk powder came through a brokered supply chain. According to the complaints, liquid whole milk from about twenty organic dairies was delivered to a Dairy Farmers of America plant in Fallon, Nevada, and spray-dried into powder. Organic West Milk, acting as a broker, sold the powder to ByHeart and shipped it to the Allerton plant, where it went into the base mix for the formula.

Powdered infant formula is not sterile. Pasteurization and spray-drying kill Salmonella and Cronobacter, the bacteria the formula industry usually worries about. They do not reliably kill C. botulinum spores, which are built to survive heat. Spores are common in soil, manure, bedding, and silage, and they can travel from those places to a cow’s udder and into the milk tank. On organic dairies, where cows spend time on pasture, weather can make that more likely.

The complaints allege that ByHeart never required testing for C. botulinum in its whole milk powder, its base powder, or its finished formula before release. Finished lots were tested for Cronobacter, Salmonella, Listeria, Staphylococcus, E. coli, and Bacillus cereus, but not for botulism. ByHeart relied instead on a general indicator test for “sulfite-reducing clostridia,” a broad family of related bacteria. That test could not do the job. ByHeart later reported that every sample in which its lab detected C. botulinum by genetic testing, 33 in all across finished product, base powder, and ingredients, had indicator counts below the level that would have raised a flag.

None of this was unknowable. In 2003, the international food standards body known as Codex named C. botulinum among the pathogens of concern that may be present in powdered infant formula. A 2010 study by California’s infant botulism program found spores of related clostridia in seven of nine unopened, store-bought powdered formulas. In March 2023, the FDA wrote to every formula maker warning that ingredient suppliers are an extension of the manufacturing process and that the historical links between powdered formula and Clostridium botulinum “should be considered when designing and implementing controls.” At least one major competitor, Mead Johnson, the maker of Enfamil, has said it tests every lot of its dairy ingredients for infant botulism contamination.

The Clean Label Project “Purity Award” that ByHeart displayed on its cans, and that Funt cited when she told parents in a November 2024 video that ByHeart was made with “certified-clean ingredients,” never involved testing for pathogens like the one that causes infant botulism.

Warning Signs

ByHeart’s own history, as described in FDA records, showed trouble long before the outbreak.

In July and August 2022, months after launch, ByHeart’s contract laboratory repeatedly found Cronobacter sakazakii in the processing environment of the Reading plant. In October, the same bacterium turned up in a batch of finished formula made with Reading base powder. ByHeart’s own analysis blamed lab error. The laboratory rejected that conclusion, writing that “laboratory error was not found or supported.”

In December 2022, ByHeart recalled five batches. Belldegrun and Funt sat side by side in a video posted to Instagram. Belldegrun told parents that the company had built its own plant in Reading “to be accountable to quality,” that “we own our entire manufacturing supply chain, except for final packaging,” and that the recall was “not related to ByHeart’s own manufacturing facility in Reading, Pennsylvania, in any way.” The company blamed a third-party packager. It told parents whose babies had already finished the recalled formula that there was “no reason for concern.”

The FDA saw it differently. In August 2023, it sent ByHeart a Warning Letter, addressed to Belldegrun. The FDA found that ByHeart had not done the investigation needed to eliminate the hazard, had not compared the genetic fingerprints of the bacteria from its plant and its product, and had let potentially contaminated formula reach customers without notifying the agency.

The problems continued. When FDA investigators returned to Reading in December 2023 and January 2024, they found mold in a tank meant to hold clean water, a leaking roof, flies over a sensitive production area, and more than 2,500 dead insects near light traps. The plant had also broken its own rules on the temperatures needed to kill bacteria in the formula base. The FDA gave the inspection its most serious classification, “Official Action Indicated.” At Allerton in February 2025, investigators found twelve Cronobacter positives in environmental samples over thirteen months, rusted surfaces, peeling epoxy, and repeated rodent activity. They also found that ByHeart had more than once released an ingredient from a shipment that showed rodent excreta and torn bags. ByHeart’s senior food safety director told the FDA that rodents were not a food safety concern because the process had a “kill step.” The kill step, the complaints point out, does not kill botulism spores.

ByHeart was also hearing from parents. Between April 2024 and March 2025, according to FDA records, ByHeart received 3,247 complaints, 2,167 of them health-related. At least seven reported infants diagnosed with Salmonella. ByHeart closed each of the investigations after a records review.

After each of these findings, the families allege, Belldegrun and Funt decided to keep producing and selling formula without changing how ByHeart qualified or tested its dairy ingredients for spore-forming bacteria.

The Recall

The California program’s alert in October 2025 set off an investigation by the FDA, the CDC, and state health departments. On November 7, the FDA asked ByHeart to recall two lots of formula. ByHeart announced the recall the next day.

In its November 8 press release, ByHeart quoted Funt: “The safety and well-being of every infant who uses our formula is our absolute highest priority.” The release also said that “no ByHeart product has tested positive for any contaminants” and that “botulism is extremely uncommon in dairy products or infant formula.” The company called the recall a step taken out of “an abundance of caution.”

That weekend, Funt recorded a video for parents and posted it on ByHeart’s social media. “There is no reason to believe that infant formula can cause infant botulism,” she said, “and multiple regulatory bodies such as Codex have come to this conclusion.” She said that “the FDA has never found a direct connection between infant formula and infant botulism, and no related toxin has been found in ByHeart.” ByHeart, she said, was committed to “real transparency, real safety.” ByHeart later deleted the TikTok version of that video. Codex had concluded no such thing. More than twenty years earlier, it had named C. botulinum among the pathogens of concern in powdered infant formula.

On November 9, after the California Department of Public Health reported C. botulinum in an opened can fed to a sick baby, the founders published an update to families signed “Mia & Ron.” It said that testing an opened can “lacks scientific basis to establish causation,” and that “no U.S. or global infant formula company tests for Clostridium botulinum.” The families allege that statement was false, because Mead Johnson had been testing its dairy ingredients for years. “We, along with our own families, friends, and team members, use ByHeart every day,” the update said, “and continue to do so.”

On November 10, the FDA asked ByHeart to recall everything. On November 11, ByHeart announced a recall of all its products, while again stressing that “no unopened ByHeart product has tested positive.” The same day, ByHeart posted on Instagram recommending another organic whole-milk formula, Nara Organics, and offering its customers a 20 percent discount code to help them switch. That post was later deleted too.

By November 19, ByHeart’s own testing had found C. botulinum in five of 36 unopened cans from three lots. On December 3, six samples of ByHeart formula tested positive, and they came from two batches: 251261P2 and 251131P2. Those were the batch numbers printed on the two opened cans that A.B.’s parents had handed to California health officials in early November. Their third can, from batch 251131P2 and still sealed, remains in their possession.

What the Science Found

Through the winter, the investigators followed the genetic trail. Whole-genome sequencing, which reads an organism’s entire DNA, can show whether bacteria from a sick child, a can of formula, and a factory ingredient are the same strain. In the ByHeart investigation, the trail led back to the milk.

The FDA collected samples of organic whole milk powder at Dairy Farmers of America, the processor for Organic West Milk, which supplied ByHeart. Two isolates from one lot of that powder matched a cluster that included a sick infant, an unopened can of ByHeart formula, a sealed can tested by New York’s Wadsworth Center, and isolates from milk powder that ByHeart itself had tested. Other clusters linked sick babies to opened cans, to finished formula, and to the base powder used to make it. The FDA inspected all three ByHeart plants and found no facility problem that could explain the outbreak. Its root-cause investigation turned to incoming ingredients, especially the dairy supply chain.

The FDA has been careful. It has said that testing for C. botulinum is complex, that sequencing has identified seventeen different strains across patient, product, and ingredient samples, and that it has not formally declared a root cause. The complaints acknowledge that. But they point out that for decades, surveys almost never found C. botulinum in commercial dried milk or formula. This time it was found in the ingredient, in the base, in finished formula, and in babies in seventeen states.

On February 26, 2026, the CDC declared the outbreak over: 48 babies, all hospitalized, none killed.

Four months later, it happened again. In June 2026, the FDA and CDC opened a second infant botulism investigation, this one tied to Nara Organics whole-milk formula, the same brand ByHeart had pointed its customers toward. Four babies in California, Pennsylvania, and Washington were confirmed with type A botulism. C. botulinum was found in an open can of Nara formula fed to one of them. Nara and ByHeart shared the same dairy suppliers: Organic West Milk and Dairy Farmers of America. Genetic sequencing matched a Nara case to a ByHeart case, to ByHeart base powder, and to four unopened ByHeart cans.

In July 2026, the FDA wrote to the entire infant formula industry. It said a manufacturer’s responsibility for safety extends to every ingredient, every supplier, and every link in the supply chain. It said its investigation had found Clostridium botulinum contamination in a dairy-based ingredient linked to both formulas. The agency, along with international food safety bodies, began a formal review of spore-forming bacteria in powdered infant formula.

ByHeart changed its testing. In a letter signed “Ron & Mia,” the founders wrote that “we now know that the existing safeguards are insufficient,” and committed to testing every batch of finished formula and its supply chain for C. botulinum. In August 2026, ByHeart announced “new testing protocols specifically designed to detect C. botulinum in dairy ingredients and finished infant formula.” That testing, the families argue, was always possible, and nothing stopped the company from doing it before it sold its first can.

After

The families have sued ByHeart, the companies that supplied and processed its milk powder, and the stores that sold it to them. The FDA sent warning letters to four major retailers, including Target, after recalled ByHeart formula stayed on store shelves for weeks. In Arkansas, state inspectors found recalled ByHeart “anywhere pack” sticks on a Target shelf under a “Sale!” sign. Now the families are amending their complaints to add Belldegrun and Funt personally.

They allege that the founders made the decisions that mattered: choosing the ingredient, approving the suppliers, setting the testing program, continuing production after each warning, and deciding what to tell parents when the outbreak came. They also allege that the founders left ByHeart with only $10 million in products liability insurance, coverage that shrinks with every dollar spent defending the company, for a business that raised more than $330 million and sold formula to babies nationwide. In October 2026, a spokesperson for the founders’ new company, MeridianMade, said ByHeart had been discontinued and would not relaunch. MeridianMade, led by Belldegrun and Funt, has agreed to buy Perrigo’s infant formula business for up to $175 million.

Some facts are not in dispute. Forty-eight babies were hospitalized. The formula their parents chose was marketed as the cleanest and safest they could buy. The milk in it carried a bacterium that is rare in dried milk but has been known for decades to be a risk in infant formula. And during those years, nobody at ByHeart was testing for it.

“For parents who can’t or choose not to breastfeed,” Belldegrun said in June 2025, five months before the recall, “this is the only nutrition that a baby consumes in that first year of life, when all of their systems are developing.”

He was right. That is why what happened matters.

And, you have to love social media.

1. Funt on TikTok, the weekend the outbreak was announced (November 2025)

“There is no reason to believe that infant formula can cause infant botulism, and multiple regulatory bodies such as Codex have come to this conclusion… no related toxin has been found in ByHeart.”

It was posted as babies were being hospitalized. Within days ByHeart recalled everything, and C. botulinum was found in opened and unopened ByHeart cans. Codex said the opposite: it flagged C. botulinum in powdered formula in 2003. ByHeart then deleted the video. That’s three problems in one statement: it was false, it misrepresented a regulator, and it was taken down.

2. The founders’ “Update for Families,” signed “Mia & Ron” (November 9, 2025)

“No U.S. or global infant formula company tests for Clostridium botulinum.”
“We, along with our own families, friends, and team members, use ByHeart every day—and continue to do so.”

The first line was false. Mead Johnson (Enfamil) tests every lot of its dairy ingredients for infant botulism contamination. The second line told parents it was safe to keep feeding ByHeart. The FDA had it recalled the next day, and ByHeart announced that recall on November 11.

3. Funt on Instagram (March 21, 2025)

“It takes nine months to go through our rigorous ingredient selection process… We have over a thousand tests on our ingredients… for unwanted things like contaminants and heavy metals and bacteria.”
“What really matters is the testing protocols.”

She held up testing as the standard to judge ByHeart by. None of those thousand tests was for C. botulinum, and the ingredient selection process she was praising approved the milk powder that carried it.

4. Belldegrun on Instagram (June 5, 2025)

“We own our supply chain, we own our manufacturing, and we directly handpick all of our ingredients.”
“Incredibly rigorous testing — like, every step, from the ingredients… to before we release product.”

The CEO says on camera that he personally picked the ingredients, which is direct proof of personal participation. He promises testing at “every step” five months before the recall, and the testing never looked for the hazard.

5. “Ron & Mia” letter after the outbreak (January 2026)

“We now know that the existing safeguards are insufficient.”

They admitted the safeguards failed and committed to C. botulinum testing they could have done from the start.

6. Belldegrun in the December 12, 2022 recall video

“We own our entire manufacturing supply chain, except for final packaging. This recall is not related to ByHeart’s own manufacturing facility in Reading, Pennsylvania, in any way.”

ByHeart’s own lab had found Cronobacter at Reading that summer. The lab rejected ByHeart’s “lab error” explanation, and the FDA’s Warning Letter to Belldegrun said ByHeart hadn’t done the investigation needed to eliminate the hazard. This shows a pattern: blame someone else and tell parents not to worry.

7. Belldegrun on control (Fast Company, June 2021)

“We believe that the more you control of the process, the more you control the quality of the formula.”

Funt said the same thing at launch: “the more you control the process, the more you can control the quality of the product” and “no shortcuts and no compromises.” These close off the “we didn’t know what our supplier was doing” defense.

8. Funt on Instagram (November 8, 2023)

“The ingredients are tested before they even ever enter the facility… there’s no rushing this and no shortcuts, ever.”
“Bacteria is something that naturally exists in the environment.”

She vouched for ingredient testing that didn’t cover the risk, and she acknowledged that environmental bacteria were a known risk.

9. Belldegrun during the 2022 formula shortage (Fox Business)

“This is a category where we cannot make mistakes.”

10. Belldegrun on Instagram (June 2025)

“For parents who can’t or choose not to breastfeed, this is the only nutrition that a baby consumes in that first year of life, when all of their systems are developing.”

Sources

I know, I sound like a broken record (you have to be old to understand that).

FDA Finally Names the Foods in Two Outbreaks: Romaine and Parsley. It Still Won’t Say Whose, Where, or Why It Took So Long

On August 12, 2026, FDA added a new line to its CORE outbreak investigation table: reference #1402, Salmonella I 4,[5],12:i:-. Forty-three people sick. Product linked to illnesses: “Not Yet Identified.”

On November 4, twelve weeks later, FDA moved #1402 to the Closed Investigations table and filled in that blank. The food was romaine lettuce. The same day, FDA did the same thing to a Cyclospora outbreak, #1392, that had been on the table since July 22. That food was parsley. Between the two, 218 people were confirmed sick, and FDA named not one farm, processor, distributor, or state.

Here is the whole of what FDA chose to tell the public about the romaine:

CDC, FDA, and state and local partners conducted epidemiologic and traceback investigations and identified romaine lettuce as the likely source of illnesses in this outbreak. FDA collected and analyzed environmental samples as part of the investigation, but all samples were reported as negative for Salmonella. Additionally, once investigators had enough information to confirm romaine lettuce as the source of this outbreak, the products were past their shelf life and no longer available for sale. This outbreak has ended, and there is no risk to consumers.

That’s it. No grower. No brand. No processor. No growing region. No states. No hospitalizations. No ages. No onset dates. No explanation of when investigators actually knew it was romaine. As Food Safety News put it, FDA closed the file with a roughly hundred-word notice that names nobody.

I have been doing this for more than thirty years. I am tired of this movie.

What the table showed, and when

Look at the checkmarks on #1402. Traceback initiated. On-site inspection initiated. Sample collection and analysis initiated. All three boxes were checked while the “product” column still read “Not Yet Identified.”

You do not send investigators on-site, or start a traceback, without knowing what food you are chasing. FDA had a working theory, almost certainly romaine, for weeks before it told anyone. When I wrote about the table in September, #1402 was already listed as ongoing with an inspection done and no food named. When I looked again in October, it was one of nine outbreaks that were over but still open, 43 sick, no food identified.

So, the public got the word “romaine” only when it could no longer do anything with it.

“No risk to consumers” is not the point

FDA’s closing line is always the same: the product is past its shelf life, so there is no risk to consumers. That is true as far as it goes. Nobody is going to buy that particular head of lettuce in November.

But that sentence treats the public as nothing more than a shopping cart. People who got sick are not shopping carts. They are 43 confirmed patients, and if you apply CDC’s multiplier for unreported Salmonella illness (roughly 29 sick for every case confirmed, a figure Food Safety News noted on November 5), the real number could be well over a thousand.

Those people have a right to know what made them sick and who produced it. So do their doctors. So do the restaurants, schools, and hospitals that bought that romaine and served it. So does every other grower in the region.

And the companies involved already know. Traceback means FDA walked the invoices back from the point of sale to the field. A negative sample taken weeks after harvest does not change what the epidemiology and the invoices showed. The only people kept in the dark are the ones who got sick and the public that pays for the investigation.

This is not a one-off

#1402 is the second romaine outbreak this year that FDA named only after the fact and without naming a grower. FDA posted an E. coli O157:H7 outbreak, #1382, on July 1. CDC declared it over on July 29. FDA did not report romaine as the vehicle until September, and neither FDA nor CDC has ever named the grower. Fourteen people were confirmed sick.

Meanwhile, the closed table now carries a long list of 2026 outbreaks where no food was ever identified: #1396 (Cyclospora, 66 sick), #1403 (Salmonella Newport, 119 sick), #1387 (Salmonella Oranienburg, 99 sick), #1385 (Cyclospora, 38 sick), and #1384 (Cyclospora, 18 sick). That is 340 confirmed illnesses with no food, no firm, and no answer.

Parsley got the same treatment

Cyclospora outbreak #1392 went up on the table on July 22 with no food named. By September it was 231 sick, then 241. Traceback was initiated. Sampling was initiated. No on-site inspection. For fifteen weeks the product column said “Not Yet Identified.” Here is how FDA closed it:

Based on CDC’s epidemiological investigation, ill people reported eating parsley before becoming sick. CDC, FDA, and state and local partners conducted epidemiologic and traceback investigations and analyzed product samples. All samples collected as part of this investigation were reported as negative for Cyclospora. Additionally, once investigators had enough information to confirm parsley as the likely source of this outbreak, parsley that would have made people sick was expired and no longer available for sale. This outbreak has ended, and there is no risk to consumers.

Same template. Same shelf-life excuse. Same “no risk to consumers.” No grower. No country of origin. No importer. No distributor. No states. No hospitalizations.

And there is a twist. FDA cut the count from 241 to 175 and moved 66 people into a brand-new outbreak, #1396, because of “differences identified during the traceback investigation.” Think about that. The traceback was detailed enough to split one outbreak into two, which means FDA knows which supply chains those illnesses ran through. Yet for #1396 FDA says it could not confirm a vehicle at all, and for #1392 it will only say “parsley.” Sixty-six people went from having an answer to having none, and nobody explained why.

Parsley, like romaine, is usually sold loose or bunched, with no brand on it, and much of it is served in restaurants as a garnish or in salsa, tabbouleh, and chimichurri. Most people who got sick have no way to know where it came from. Only FDA does.

I like the people at FDA. For why, then?

I have said this before, and I will say it again. I like the people at FDA. I have known many of them for decades. The epidemiologists, investigators, and lab staff who worked #1402 and #1392 did their jobs. They found the foods. That is hard, unglamorous work, and it rarely gets thanked.

But these decisions, to sit on “romaine” for twelve weeks and “parsley” for fifteen, and then close both files without a grower, a processor, or a state, look like FDA is hiding information from the public. For why?

It is not to protect consumers. FDA says there is no risk. It is not to protect the investigation. It is closed. It is not because the law ties FDA’s hands. FDA has named farms, processors, and brands after the product was gone at least eight times since 2006, and the Justice Department just told HHS that FDA can write a rule to do it every time. The only ones who benefit from the silence are the companies whose romaine and parsley made 218 people sick and the buyers who would rather not know.

This is the same agency that told the food industry last year that “radical transparency in food safety and nutrition actions is key to reducing foodborne illness.” I agree. Naming the source of an outbreak is not radical. It is the minimum.

When a government agency knows something about a public health event and will not say it, people fill the gap with the worst assumption, that someone is being protected. Maybe that is unfair to the people I know at FDA. But FDA is the only one that can fix it, and it can fix it by saying the name. Transparency is not an attack on FDA. It is how FDA earns the trust it needs the next time it asks the public to throw something away.

What FDA should release now

Both investigations are closed. Whatever reason FDA had for silence while they were open is gone. For each, FDA should post:

  • The grower(s), processor(s), importer(s), and brand(s) of the romaine and the parsley, and where each was grown.
  • The number of states, the onset dates, and the number of people hospitalized.
  • When the traceback converged, and when FDA first suspected each food.
  • Where the romaine inspection happened, and what was sampled in each investigation.
  • Whether any recall or market withdrawal was considered, and why not.
  • For parsley, what traceback “differences” split off the 66 people now in #1396.

If FDA does not volunteer that information, it will come out the slow way, through FOIA and through litigation. It should not take a lawyer to find out what made 218 people sick. It should take a rule.

#1402 and #1392 are Exhibits A and B for my petition

I am about to file two citizen petitions with FDA: one on Outbreak Firm Disclosure and a companion on recall consignee lists. I have added #1402 and #1392 to both. The first one opens with a romaine outbreak: #1382, the fourteen-person E. coli file FDA closed in September with no grower, no processor, no brand, and no state. #1402 and #1392 now sit right beside it, and every request in the petition fits them.

  • Name the food once traceback begins. Item 5 of the petition asks that when FDA starts traceback on a specific food, the table say what that food is, and that it carry the counts FDA and CDC already have: cases, states, onset dates, hospitalizations, and deaths. #1402 had traceback, inspection, and sampling checked for weeks while the product column said, “Not Yet Identified,” and it closed without a single state or hospitalization.
  • Reach the whole supply chain. Item 3 asks that the rule cover the grower, cooler, processor, distributor, and importer, not just a brand. Romaine and parsley usually have no brand at all.
  • Treat “romaine” or “parsley” alone as a problem, not an answer. The petition asks FDA to recognize that when the public record names only a commodity, consumers attribute the outbreak to every grower of that commodity, including those that did nothing wrong. Telling people “romaine” in November, and nothing more, punishes every honest romaine grower in the country and shields the one whose lettuce made people sick.
  • Name firms after the investigation closes. “Past shelf life” is FDA’s stock reason for silence. The petition asks for a rule that the identity of a firm implicated by epidemiology and traceback is not confidential commercial information, whether or not the product is still on shelves. That is the same evidence FDA used to close #1402.

And FDA cannot say its hands are tied. On October 26, the Justice Department’s Office of Legal Counsel told HHS that section 705(b) of the Food, Drug, and Cosmetic Act lets FDA amend its own regulations to authorize naming firms in a supply chain. The obstacle is FDA’s regulations, and FDA can change them. That is all the petitions ask.

The chicken wraps in E. coli outbreak #1421 show the same problem with a brand attached. On November 4 FDA wrote that it, USDA-FSIS, and CDC have “confirmed” the multi-ingredient wrap with chicken as the likely source. The table still does not say Trader Joe’s, even though FSIS confirmed it to USA TODAY in October. Fifty people are sick, and my firm represents two of them. I laid out the broader fix in Food Safety News.

Elsewhere on the November 4 table
  • Salmonella Javiana #1424 grew from 59 to 65 sick, and FDA has now started traceback. Again, no food named.
  • Salmonella Bovismorbificans #1411 is over. That one, tied to Evergreen Fresh Sprouts broccoli sprouts and recalled broccoli seed, sickened 34 people in six states. FDA named the sprouter and the brands, which shows it can be done.
Bottom line

Twelve weeks and fifteen weeks. Two hundred eighteen confirmed illnesses, likely many more. Traceback, inspections, and sampling. And at the end, two words: romaine and parsley.

The public paid for these investigations. The people who got sick lived it. They deserve the rest of the sentence.

Sources