I have been tracking this outbreak daily since it began. These are the questions I am actually being asked — by reporters, by clients, and by people who just want to know whether to eat a salad. Every figure below is current as of the evening of August 5, 2026, and every one of them will move. Where a number is uncertain, I have said so.

What is Cyclospora, and what does it do to you?

Cyclospora cayetanensis is a microscopic parasite. Swallowing even a small amount of it in contaminated food or water causes cyclosporiasis, an intestinal illness whose defining feature is frequent, watery diarrhea “Explosive Diarrhea”. Other common symptoms are loss of appetite, weight loss, stomach cramps, bloating, gas, nausea and fatigue; some people get a low fever, body aches or vomiting.

Two things make it different from most food poisoning. First, symptoms usually start about a week after exposure — the range is two to fourteen days — so by the time you are sick, the meal that did it is a distant memory. Second, untreated, it does not simply run its course over a bad weekend. It can last weeks, and it characteristically comes and goes, so people think they are better and then relapse.

How many people are sick right now?

That depends entirely on who is counting, and the honest answer is three different numbers.

The federal outbreak count is 6,358. On August 5, CDC and FDA updated their advisory to cover fifteen states, with 278 hospitalizations and two deaths. That figure counts only people whose illness has been laboratory-confirmed and tied to this specific investigation.

CDC’s national count is larger. Its surveillance page reports 10,468 laboratory-confirmed domestic cases since May 1, plus more than 12,255 additional cases not yet confirmed, across 47 states, with 517 hospitalizations.

And the fifty state health departments, added up, report 25,213. That is the number I keep, because it is what each state is telling its own residents. It is a floor, not a ceiling — thirteen states publish no count at all.

Why are the numbers so different from each other?

Because they are answering different questions, and FDA says so on its own page. The federal counts include only laboratory-confirmed cases. States also report probable cases — people who are clearly sick with this and were clearly exposed, but whose lab work is not finished or was never done. Neither number is wrong. The federal one is narrow on purpose, and the state one is closer to the number of people actually ill.

There is also a lag. CDC updates weekly; many states update daily. On the day CDC posts, the two move close together, and the distance opens again every day after.

What food is this?

Iceberg lettuce grown in central Mexico. On July 17, Taylor Farms de Mexico voluntarily recalled all of it from the US market. The recalled product went to foodservice and to Walmart under the Marketside brand, across 29 states in total.

Until this week the federal investigation counted only people who reported eating at Taco Bell. On August 5, FDA disclosed that CDC had widened the definition to include anyone who reported eating at Taco Bell oreating the recalled Taylor Farms lettuce. That is why the federal number more than tripled overnight. It is worth understanding what that means: this was always a lettuce problem. The restaurant chain was where investigators could see it.

Is the recalled lettuce still in my refrigerator?

Almost certainly not. Shredded iceberg has roughly a two-week shelf life, and the recalled product was distributed between June 29 and July 16. Anything from that window is long gone — eaten, thrown out or spoiled.

Which is also the uncomfortable part. A recall reaches forward, never backward. By the time this one was announced on July 17, the lettuce that made people sick had already been served “People Ate the Evidence.”

Should I stop eating salad?

No, and health officials across the affected states have been consistent about this: keep eating fresh produce. The recalled lots are out of the supply chain, and the nutritional cost of avoiding vegetables is real while the current risk from any particular head of lettuce is very low.

If you want to reduce risk anyway, buy whole heads rather than pre-cut or bagged product, discard the outer leaves, and wash the rest under running water. Understand what that does and does not accomplish — see the next question.

Will washing my produce kill it?

No. This is the single most important thing for the public to understand, and it is FDA’s own position, not mine. On its produce fact sheet, FDA states that chlorine and the other antimicrobial chemicals commonly used on fresh produce are not effective against this parasite. CDC said the same thing in its July health alert: sanitizing produce might not fully eliminate Cyclospora.

Washing helps at the margins. In published studies, a minute under cold running water removed most parasites from some produce and well under half from others — the oocysts stick, and they hide in crevices. No method removed all of them.

Put plainly: for ready-to-eat produce there is no validated kill step for this organism. Cooking to a high temperature works, which is no help for salad. That is why prevention has to happen upstream, in the water and the fields, and why I keep writing about irrigation water rather than about your kitchen sink.

Can I catch it from someone who has it?

No. When the parasite leaves the body it is not yet infectious — it needs one to two weeks in the outside environment, at the right temperature, to mature. That is why cyclosporiasis does not spread person to person the way norovirus does, and why a sick family member is not a danger to you. It also means the contamination that caused this outbreak happened weeks before the first person got sick, somewhere upstream of the field.

I have had diarrhea for two weeks. What should I do?

Call your doctor and be specific. Say how long it has lasted, that it is watery, that it has come and gone, and ask directly whether you should be tested for Cyclospora. That last part matters more than it should have to.

Routine stool testing does not reliably find this parasite. It needs a specific stain or a molecular panel that includes a Cyclospora target, and not every panel does. On top of that, the parasite is shed intermittently and in small numbers, so one negative stool does not rule it out — CDC advises that more than one specimen, collected on different days, may be needed. If you were told your test was negative early in an illness that has not resolved, that is worth revisiting.

How is it treated?

The standard treatment is a common antibiotic combination, trimethoprim-sulfamethoxazole – Bactrim. It works well and it shortens the illness considerably — in the one controlled trial, treated patients stopped shedding the parasite in about five days against roughly twelve days untreated.

People who cannot take sulfa drugs have meaningfully fewer options, and that is a real gap in an outbreak this size. If that is you, raise it with your doctor early rather than waiting. Nothing here is medical advice; it is the reason to make the call.

Have people died?

Two, both in Michigan, both in people with underlying health conditions. FDA has since added a detail worth noticing based on information Michigan provided, both had illness onset dates before the July 17 recall.

278 people have been hospitalized in the federal outbreak count; CDC’s broader national surveillance puts hospitalizations at 517. For most healthy adults this is a miserable few weeks, not a life-threatening event. For people who are immunocompromised, elderly or already ill, it is more serious.

My state is not on the list. Am I safe?

The fifteen-state outbreak list and the 29-state distribution list are not maps of where people are sick. CDC is receiving reports from 47 states this season. Two states in the outbreak had no documented shipment of any recalled product, which tells you that distributors moved it further than the original list records.

A state gets on the outbreak list when its health department interviews patients, finds the exposure and reports it — which requires that somebody test, somebody ask, and somebody have the staff to do both. Absence from a list is not evidence of absence of illness.

How many people has this outbreak really made sick?

Nobody knows, but the federal government has a published model for exactly this, and it is worth walking through because it is the most striking number in the entire outbreak.

In 2011, Elaine Scallan and seven colleagues at CDC published the agency’s foodborne illness burden estimate in Emerging Infectious Diseases. Its Table 2 remains the only published United States burden estimate specific to Cyclospora. The underdiagnosis multiplier it assigns to this parasite is 83.1. In other words, CDC’s own model assumes that for every laboratory-confirmed case, about eighty-three people actually got sick.

Apply that where it belongs — to the 10,468 laboratory-confirmed cases CDC reports this season — and you get 869,891 people. Roughly eight hundred and seventy thousand.

Two cautions, and I want to be the one to give them rather than have someone else. First, the multiplier does not go on the 25,213-state figure. Those state totals already include the unconfirmed cases the multiplier exists to estimate, so multiplying them would count the same people twice.

Second, that number is arithmetic, not a measurement, and it should be read as a ceiling. The model’s own credible range for Cyclospora spans 137 to 37,673 domestically acquired foodborne illnesses in a normal year — a 275-fold spread, the least precise line in the table. The 83.1 was built from data gathered before molecular test panels carried a Cyclospora target at all, which happened in 2014, and an outbreak that has doctors actually ordering the test pushes the real multiplier down.

What the figure is genuinely good for is the shape of the problem. The same table puts Listeria at 2.1 — because severe illness gets diagnosed and a miserable, self-limiting one does not. The gap between 83.1 and 2.1 is not a statement about parasites. It is a statement about how many sick people never enter the count at all.

Are there long-term effects?

For most people, no. The illness resolves, with or without treatment, and there is no evidence of the kind of lasting organ damage that follows some other foodborne infections.

That said, weeks of relapsing diarrhea takes a real toll — weight loss, dehydration, missed work, and in some people a lingering irritable bowel pattern afterward. The long-term picture for an outbreak of this size has not been studied, because there has never been an outbreak of this size.

When will this be over?

CDC defines the Cyclospora season as May 1 through August 31, which leaves 26 days. But there is a reporting lag of several weeks between someone getting sick and their case reaching the national count, so the numbers will keep climbing well into September even if not one additional person is exposed after tonight.

One state health department said this week that some experts believe the outbreak is beginning to wane. I hope they are right. The case counts will not show it for a while either way.

Where do I get the current numbers?

CDC’s outbreak page and surveillance page and FDA’s advisory are the official sources. Your own state health department will usually be days ahead of all three, and I post a fifty-state tally most days.

One last thing, because it is the point of the whole exercise. The reason there are three different national numbers, and the reason the honest estimate of the true toll is eight hundred thousand rather than six thousand, is not that anyone is hiding anything. It is that we count foodborne illness by waiting for sick people to seek care, for a doctor to think of ordering an unusual test, for a laboratory to run it, and for a state to report it. Every one of those steps loses people. In an outbreak this large, it loses most of them.

Should I talk to a lawyer?

That is your call, and you do not need one to report an illness or to get medical care. If you were hospitalized, missed significant work, or are dealing with lasting complications, it is reasonable to get advice about your options. Any lawyer worth talking to will tell you the same things listed above first: get treated, get cultured, and report it.

Where can I follow the investigation?

CDC’s outbreak page and FDA’s outbreak investigation table are the two primary sources. Note that the two agencies have not always been in step — as of this writing FDA’s table still showed an older case count. Following www.marlerblog.com is good too.