In September 2026 the FDA was running two sprout outbreak investigations at once. One covered alfalfa sprouts from Minneapolis: 55 people sick across 15 states. The other covered broccoli sprouts from Idaho: 22 people sick across four states. Different companies, different pathogens, different seed lots.
What the two cases share is something else. The Idaho grower was showing up in outbreak records for the third time since 2011. The Minneapolis grower had received a warning letter sixteen months earlier describing water dripping through a leaking roof onto uncovered sprouts.
Read enough FDA paperwork and the pattern stops looking like coincidence. The same addresses come back for three reasons: the seed, the building, and what the agency is actually able to do about a firm that has been here before.
Idaho: 2011, 2014, 2026
Evergreen Fresh Sprouts operates in Moyie Springs, a small town in the Idaho panhandle. Its sprouts appear in CDC outbreak records three separate times.
2011. Salmonella Enteritidis, 25 people sick in five states: ten each in Montana and Washington, three in Idaho, one each in New Jersey and North Dakota. Onsets ran from April 12 to June 15. The alfalfa and spicy sprouts, sold under the Evergreen Produce label, were recalled on July 1, 2011.
2014. E. coli O121, 19 people sick in six states, eleven of them in Washington. Forty-four percent were hospitalized. The source was raw clover sprouts from the same firm. There was no recall: CDC noted that Evergreen had stopped using the implicated seed lot and that sprouts grown from it were likely gone anyway, given a shelf life of about 14 days. The product reached people and expired before the investigation reached a conclusion.
2026. Salmonella Bovismorbificans, 22 people sick in Idaho, Montana, Utah and Washington, two hospitalized, onsets from July 7 to August 26. Of 19 people interviewed, all 19 reported eating sprouts and 17 specifically reported broccoli sprouts. Montana state officials collected the positive sample, and whole genome sequencing matched it to the outbreak strain. The firm recalled 215 cases of broccoli sprouts on September 4, 2026.
Fifteen years separate the first episode from the third, and the operation ran throughout.
Minneapolis: the letter that described the outbreak in advance
The second case is shorter and, for that reason, easier to read. On March 28, 2025, the FDA sent Everything Sprouts a warning letter based on an inspection that ran from October 30 to November 14, 2024.
The letter lays it out. On October 31 the investigator watched the roof leak through ceiling tiles in the middle of the production room, with water dripping onto totes of cooled, exposed mung bean sprouts as they were moved for packaging. On November 4 a second leak in the warehouse dripped near a rack of clean smocks waiting for employees who handle produce directly. Add condensate dripping from the cooler units. The FDA concluded the sprouts were adulterated: prepared, packed or held under insanitary conditions.
The most telling sentence in the letter is not about 2024 at all. The FDA writes that its investigators observed the same dripping condenser units during inspections of this operation in 2023 and in 2018, when the business was under different ownership but running in the same facility. The defect outlived two changes of owner, because what stayed put was the building, the equipment and the way the work is done.
The company answered with a plan: a written procedure for leaks, daily checks, a squeegee for condensate, drip pans later on. The building is rented, the landlord was to fix the roof as a long-term project, and no repair date appears in the letter. Twice the FDA used the same phrase: “We will evaluate the adequacy of your corrective actions at our next inspection.” The next inspection on the public record began on August 19, 2026, by which point investigators were counting patients.
That outbreak produced 55 cases in 15 states, four hospitalizations, and an unusual combination in a single investigation: three separate Shiga toxin-producing E. coli strains (O26:H11, O103:H25, O168:H8) plus Salmonella Agona. We covered it in detail in our piece on the alfalfa sprout outbreak that hit 15 states.
Four operations out of ninety-four
The clearest evidence that this is a structural feature of the industry rather than two unlucky firms comes from the FDA’s own sampling assignment. From February 2014 through September 2015 the agency collected 825 samples across 37 states at three points in production: seeds, finished sprouts, and spent irrigation water.
The results were lopsided. Violative samples turned up at eight of the 94 sprouting operations visited. And of the 14 positive samples found in the entire study, ten came from just four establishments.
The same report contains a quiet line about the size of the industry. The FDA had planned to collect 1,600 samples but stopped at 825, because it had already sampled many of the sprouting operations known to the agency and its state partners more than once. There are few enough sprout growers in the country that one federal program worked through most of them in about eighteen months.
The study also mapped where contamination sits. Salmonella showed up on 2.35 percent of seed samples, 0.54 percent of spent irrigation water samples, and 0.21 percent of finished sprouts. The gap between seed and finished product is statistically significant.
One seed lot makes several companies guilty at once
That is the first reason for repeats. A sprouter does not grow seed, it buys seed. A lot keeps for a long time and travels widely, so contamination in a single lot becomes several companies’ problem simultaneously. Who in that chain is required to test the lot before it sells, and why the American answer differs from the European one, is covered in whose job it is to test sprout seed.
In the August 2026 outbreak the FDA identified a common seed supplier and at least one additional sprout grower that received the same lot. When the supplier finally agreed to a recall on September 15, the paperwork showed the lot had gone to growers and wholesale distributors in 16 states and Puerto Rico, plus Canada, Mexico, Tahiti and Venezuela. We walked through that mechanism step by step in how one contaminated seed lot becomes a 15-state outbreak, and separately covered what happens when a company refuses an FDA recall request.
The 2012 record reads the same way. Investigating an E. coli O26 outbreak that sickened 29 people in 11 states, the FDA traced a single lot of seed that had been grown and distributed by multiple sprouting firms. Six restaurants bought sprouts from different suppliers, and the seed underneath them was shared.
The Listeria that moves into the plant
The second reason lives in the walls rather than the seed. In the same sampling program, Listeria monocytogenes turned up on 1.28 percent of finished sprouts, with no meaningful difference between stages of production. For Salmonella the difference exists and points at seed. For Listeria there is no difference, and that points at the production environment.
The FDA states the consequence plainly: once L. monocytogenes becomes established in a sprouting operation, it can be a source of repeated contamination of sprouts, food contact surfaces and the production environment. “Repeated” is the operative word. Cleaning up one batch does not settle it, because the organism is living in a drain, a pipe, or the seam of a cooler panel.
That is why the Everything Sprouts letter spends paragraphs on black buildup inside an irrigation hose, plastic sheeting torn in several places around the growing cabins, and repairs attempted with tape that had sprouts stuck to it. This is not housekeeping fussiness. Every one of those gaps is a place where bacteria survive a wash.
What the FDA can do with a repeat offender
The third reason is the enforcement sequence itself. Inspection, then a Form 483 listing observations, then a warning letter, then an evaluation of corrective actions at the next inspection. Months or years pass between the links, and the firm keeps operating the whole time.
The sharpest example involves a buyer rather than a grower. On February 21, 2020, the FDA told Jimmy John’s that the evidence showed a pattern: the chain had been receiving and offering for sale adulterated fresh produce, specifically clover sprouts and cucumbers. The letter lists five outbreaks: E. coli O103 in Iowa in late 2019, where all 22 patients had eaten at Jimmy John’s; Salmonella Montevideo in 2018; E. coli O121 in 2014, the same outbreak traced to the Idaho grower; E. coli O157:H7 from cucumbers in Colorado in 2013; and E. coli O26 in 2012. Seven years, at least 17 states.
The passage about promises is the one worth remembering. In May 2012 the chain requested a meeting with the FDA and said it would serve only clover sprouts and buy them only from specified suppliers. After that commitment the company was implicated in three more sprout outbreaks, and traceback records showed its restaurants were buying sprouts from other sources as well. At the time the letter was written, roughly 2,800 franchised locations in 43 states were in scope.
A harder tool exists and is used sparingly. In April 2015 the Justice Department obtained a consent decree of permanent injunction against Wholesome Soy Products of Chicago and two of its officers. The FDA had inspected the plant in August and September 2014 and again that October, and the October findings were repeat observations from the earlier inspection: inadequate cleaning, ineffective pest control, poorly maintained equipment. The decree bars the company from producing food at that facility until it completes a list of remedial measures, including a Listeria monitoring program.
There is a rule, and the outbreaks continue
Sprouts are the only category of fresh produce that FSMA singled out with its own requirements. Since 2015, Subpart M of 21 CFR Part 112 has required growers to work inside a fully enclosed building; to clean and sanitize food contact surfaces before they touch seed or sprouts; to treat seed with a scientifically valid method or document that the supplier did; to test spent irrigation water from every production batch for Salmonella and E. coli O157:H7; and to run a written environmental monitoring plan for Listeria with sampling no less than monthly.
Those are real obligations, and the FDA still counts 52 sprout-related outbreaks in the United States between 1996 and 2020, with at least 2,700 illnesses, 200 hospitalizations and three deaths. The reason is that no step in the process reliably kills the pathogen. Seed treatment reduces the load, the warm wet germination conditions bring it back, and batch testing catches contamination only when it is spread evenly enough and present in high enough numbers to land in the sample.
What this means if you buy sprouts
Three practical conclusions.
- The grower’s name matters more than the brand. In August 2026 the same sprouts sold under two labels, and the shopper at the register saw a brand rather than an address. Recalls and investigations are published by grower, which makes the FDA’s table of open outbreak investigations more useful than the front of the package.
- Sprouting at home does not remove the seed risk. A contaminated lot behaves the same way in a factory and in a jar on your counter, and the warm, wet conditions are identical. The home jar is safer in exactly one respect: it never passes through a plant where Listeria may have settled into a drain.
- Sprouts and microgreens are different products in risk terms. People use the words interchangeably, though the growing method and the part of the plant you eat are not the same. We laid out the difference in sprouts vs microgreens.
One more thing follows from those ten positive samples out of fourteen at four establishments. The industry is small, and a handful of sites account for much of the trouble, which means caution is better aimed at a specific grower than at the whole category. For handling greens at home, the practical guidance sits in our food safety section. The same pattern shows up outside sprouts: one firm recalled its greens powders twice in six months, and the second lot range starts where the first one ended.
