Fifty-five people in fifteen states are sick with Salmonella Agona or one of three strains of Shiga toxin-producing E. coli after eating alfalfa sprouts grown by Everything Sprouts, LLC of Minneapolis and sold under the Everything Sprouts and Calco brands. Four have been hospitalized and no one has died. Illness onsets run May 31 through August 8, 2026. On Friday, hours after USA TODAY asked why there was still no recall, the company issued one. As of Saturday morning, August 22, that recall does not give a consumer a single thing she can act on.

Mary Walrath-Holdridge asked why there was no recall, and by Friday night there was one.

Credit where it belongs. Mary Walrath-Holdridge at USA TODAY spent Friday on the question that actually mattered in this outbreak, which was not the case count. It was why a company whose product had been named by two state health departments, by CDC and by FDA still had not recalled anything. Her piece went up at 5:38 p.m. Eastern on Friday, and it says plainly that as of August 21 there had been four hospitalizations and no recall.

She asked me why an outbreak carries four organisms at once. I told her it is unusual, but that when it happens it tends to happen in sprouts and in raw milk. The reason is structural and I gave her the short version: sprouts are a raw agricultural product grown outside, harvested and put into fifty-pound bags, and if the seed picks up contamination in the field, putting it in a warm wet environment to sprout is a perfect place for bacterial growth. Nothing between the field and the sandwich kills what arrived on the seed.

I also told her something that belongs next to every count in this outbreak. Two of the fifty-five are recorded as carrying both Salmonella and E. coli. That two is a floor, not a finding. When a stool sample is cultured, the lab is generally looking for a pathogen, not for every pathogen, and once something grows the work usually stops. Nobody goes back through the plate hunting for a second organism in a patient who already has an answer. The real number of people in this outbreak carrying more than one germ is almost certainly higher than two, and we will never know by how much.

She was asking the same thing I have been asking since Wednesday, and I have asked it about sprout growers before — in 2011, when an Idaho sprouter told a reporter she wanted concrete evidence before pulling product and most of it had probably been eaten anyway.

Late Friday, Everything Sprouts posted a two-page voluntary recall notice on its own website. Read it. It is the most revealing document to come out of this outbreak so far, and not for the reasons the company intended.

The notice says it covers certain lot numbers and then lists none.

The recall describes the affected product as Everything Sprouts alfalfa in a five-ounce plastic container with a lid, distributed from May 27 to the present, and then says the products subject to the recall include the following lot numbers. Nothing follows. The list is empty. There is no lot code, no best-by date, no product photograph and no UPC in the document.

The notice also says the product went to certain third-party distributors and to a number of retail grocery stores in Minnesota and Wisconsin. It does not name one distributor or one store. Sick people live in fifteen states, including Florida, Washington, Pennsylvania and the Carolinas. A shopper in any of them reads this notice and learns nothing about whether the sprouts in her refrigerator are the sprouts in question.

The Calco brand is missing, and it is half of what people were told to look for.

FDA and CDC both told consumers not to eat, sell or serve alfalfa sprouts under the Calco and Everything Sprouts brands. Minnesota and Wisconsin named both brands in their releases on August 20. The recall names one. Calco Sprouts, a Minneapolis company founded in 1991 and now operating under Everything Sprouts, does not appear anywhere in the document.

The organisms are short a few names too. FDA identifies three STEC serotypes in this outbreak — O26:H11, O103:H25 and O168:H8 — plus Salmonella Agona. The recall notice names the first two serotypes and refers to Salmonella generically. A recall that describes less of the hazard than the federal advisory it was supposedly written to satisfy is not a recall that was written carefully.

The company says no illnesses have been reported to it, one sentence before saying agencies reported illnesses to it.

Two consecutive paragraphs. The first says that to date no reports of illness involving the product have been reported to the company. The second says the company became aware of reports of illnesses from federal and state agencies. Both sentences call the firm Everyday Sprouts, which is not its name. Fifty-five people are sick.

The notice adds that during this outbreak the company has tested its products daily and received no positive results. Sprouts have a shelf life measured in days. The outbreak was announced publicly on August 20 and the last illness onset FDA has is August 8. Daily testing that began after the announcement is testing this week’s product, and this week’s product is not what made anyone sick. A negative lot today says nothing about a lot in July, which is precisely why the lot codes matter and precisely why they are not in the document.

FDA’s blank sample press release is still attached at the bottom.

FDA publishes a model recall notice so that a firm in the worst week of its corporate life does not have to invent one. Everything Sprouts used it, which is fine. What is not fine is that the template is still in the file. The document ends with FDA’s sample, unaltered — a fictional XYZ Inc. of Anywhere, MS, recalling five-ounce packages of Snackies, lot number 666666, expiration date 12/12/99. Nobody deleted it before publishing. As far as I can tell from the public record, nobody at FDA read the notice before it went out either.

Which brings me to the other document that has not changed. FDA’s outbreak advisory, content current as of August 21, still says the agency has recommended that the company conduct a recall, and still says FDA is working to determine additional distribution information. The recall happened and the advisory does not reflect it. The one page a worried person is most likely to find still tells her a recall has been asked for, not issued. The March 2025 warning letter to this same facility ended the same way, with FDA saying it would evaluate the corrections at the next inspection. The next inspection was August 19, 2026, and it happened because people were in the hospital.

There is a third page that does not carry it either. FDA’s Recalls, Market Withdrawals and Safety Alerts list is where a shopper, a grocer or a reporter goes to find out what has been recalled. As of Saturday morning, content current as of August 21, Everything Sprouts is not on it. The ten most recent entries run back to August 17: a canned dog food recalled for metal fragments, a sourdough loaf with metal in it, two allergen notices, an IV solution, ice cream bars with glass, jalapeños. The dog food went up on August 21, the same day as the sprout recall, so this is not a case of the list running a day behind.

The page carries its own caveat that not all recalls are posted there. Fair enough as a general matter. Four organisms, fifty-five people, fifteen states, four hospitalizations and an ongoing federal advisory telling the country to throw the product away is not a general matter. A recall that appears on no government page, names no lot codes and identifies no customers exists in one place only — a PDF on the recalling company’s own website, which a person would have to already know to look for.

A recall is paperwork, and this paperwork does not do the work.

I will concede the shelf-life point before anyone makes it. Sprouts sold in July are gone. Nobody is going to find a contaminated clamshell in the back of the fridge from the week the outbreak peaked. That is not what a recall is for.

A real recall produces a record. Lot codes, so a distributor can match what it received. A customer list, so notification runs down the chain to every wholesaler, restaurant and grocer that took the product. An entry in FDA’s Recall Enterprise System and the weekly enforcement report, with a hazard classification attached to it. A press release that names the brands and the stores. Two years from now that record is what tells a court, a regulator and the next family what moved where, and when the company knew. A notice with no lot codes, no customer list, one brand out of two and a Gmail address as the point of contact produces none of it.

Fifteen states have sick people. Two states are named in the recall. That gap is the whole story, and it has not closed since Friday.

Questions people are asking this weekend

Is there a recall on Everything Sprouts and Calco alfalfa sprouts?

Yes, as of August 21, 2026, but a narrow one. Everything Sprouts, LLC issued a voluntary recall of Everything Sprouts alfalfa sprouts in five-ounce plastic containers with lids, distributed from May 27 onward to distributors and grocery stores in Minnesota and Wisconsin. The notice lists no lot numbers, does not mention the Calco brand, and as of August 22 does not appear on FDA’s recall list. FDA and CDC continue to tell consumers not to eat alfalfa sprouts under either brand.

What should I do if I have these sprouts in my refrigerator?

Throw them away or return them to the store, and do not eat them or feed them to anyone. Wash and sanitize any surface, container or drawer they touched, using hot soapy water or the dishwasher. Do not try to wash the sprouts themselves. Washing does not remove bacteria that grew inside a sprout, which is why every agency in this outbreak is saying discard rather than clean.

How many people are sick, and where?

Fifty-five, in fifteen states: Minnesota, Wisconsin, Florida, Indiana, Iowa, Kansas, Michigan, New Hampshire, New York, North Carolina, North Dakota, Pennsylvania, South Carolina, South Dakota and Washington. Four have been hospitalized. Forty-six had E. coli, seven had Salmonella, and two are recorded as having had both — a floor, because a lab generally stops culturing once one pathogen grows. Of thirty-four people interviewed, twenty-six — seventy-six percent — reported eating alfalfa sprouts. Those are states of residence, not states where the sprouts were eaten, and the real number of sick people is higher than the confirmed count in every outbreak of this kind.

What symptoms should I watch for, and when?

STEC symptoms begin from a few days up to nine days after exposure: severe stomach cramps, diarrhea that is often bloody, and usually little or no fever. Salmonella usually shows up within twelve to seventy-two hours, with diarrhea, fever and abdominal cramps. Call a doctor right away for bloody diarrhea, a fever above 102, diarrhea lasting more than three days, or signs of dehydration. In children, watch for decreased urination, unusual tiredness and loss of color in the cheeks and inside the lower eyelids — those can signal hemolytic uremic syndrome, a form of kidney failure that follows STEC infection.