In a normal year, Michigan finds about 50 cases of cyclosporiasis. In the last three and a half weeks it has logged more than 4,300, and 102 people have been hospitalized. Nobody has found the source.
The usual advice about rinsing your produce doesn’t work on this one. Cyclospora doesn’t wash off. Chlorine doesn’t kill it. Your freezer doesn’t kill it. Here’s what the parasite is, what does work, and where the real line sits between a risky salad and a safe one.
What Cyclospora is, and why it behaves strangely
Cyclosporiasis comes from Cyclospora cayetanensis, a single-celled coccidian parasite 8-10 micrometers across. Humans are its only known host. No cows, no chickens, no wildlife reservoir, which is unusual for an intestinal parasite.
An infected person sheds unsporulated oocysts, and those are not infectious. To become infectious, an oocyst has to spend 7-15 days in the environment at 73-81°F (23-27°C) and mature. Only then can it infect anyone.
The consequences:
- Person-to-person transmission is unlikely. This isn’t norovirus: a sick family member won’t pass it around the house through a doorknob.
- Fresh fecal contamination isn’t dangerous either. If a field worker breaks hygiene today, today’s harvest doesn’t become infectious from that.
- Every infection is an old story. The contamination happened weeks before that lettuce reached your plate: in irrigation water, in soil, at a packing shed. Somewhere in there, the oocysts quietly matured.
That’s the whole outbreak in one fact. Cyclosporiasis is a supply chain and field sanitation problem, and the fix sits with growers, packers and regulators. Not your sink.
The Michigan timeline
June 22, 2026 is the symptom onset date for the earliest cases.
July 1. MDHHS announces an investigation: more than 170 cases in nine days across seven southeast counties (Monroe, Lenawee, Washtenaw, Wayne, Livingston, Shiawassee, Jackson). The state’s baseline is roughly 50 cases per year.
July 13. 2,640 cases. A probable source is named for the first time: lettuce and salad greens. No specific grower, supplier or product.
July 15. 3,762 cases.
July 16. 4,312 cases, 102 hospitalizations.
The national picture isn’t better. The CDC issued a Health Alert Network notice on July 14:
- 1,645 lab-confirmed domestically acquired cases across 34 states since May 1, against 249 for the same window in 2025. More than a sixfold increase.
- Another 5,100-plus cases await analysis. Together, roughly 7,000 confirmed and probable.
- 141 hospitalizations (9%), no deaths.
- A cluster of 400-plus epidemiologically linked cases spans four states: Michigan, Ohio, West Virginia, Kentucky. Linked means a likely common source.
For scale: the US logged 2,700 cases in all of 2025. The record was about 4,700 in 2019. 2026 has already passed it, and Cyclospora season (May through August) isn’t over.
Taco Bell temporarily pulled some fresh ingredients at select restaurants. The FDA is running five parallel investigations. There’s still no source.
Why the source is so hard to find. Incubation runs up to two weeks. Add a few days before someone sees a doctor, plus weeks for lab confirmation. Roughly six weeks pass between infection and a case appearing in CDC data. The suspect salad was eaten or thrown out long ago. There’s nothing left to test.
Symptoms, and why this isn’t regular food poisoning
Symptoms show up 2 days to 2 weeks after infection, usually around a week:
- watery diarrhea, often explosive, that comes and goes in waves;
- cramping, bloating, gas;
- nausea, loss of appetite, noticeable weight loss;
- heavy fatigue, sometimes body aches and a low fever.
What separates it from most food poisoning is duration. Untreated, symptoms drag on for weeks and relapse. A month of grinding diarrhea is a realistic outcome, and people shed serious weight along the way. Treatment is antibiotics, usually trimethoprim-sulfamethoxazole (Bactrim), plus fluids.
A routine stool test misses Cyclospora. A standard ova-and-parasite panel often doesn’t see it. That’s why the CDC is telling clinicians to order a test specifically for Cyclospora, ideally PCR. On its own line on the requisition. If you’ve had watery diarrhea coming in waves for more than a few days, ask for that test by name, not for a general panel.
Why washing doesn’t save you
A Cyclospora oocyst is a capsule with a hard shell. It grips microscopic crevices on leaves and skins and doesn’t rinse off cleanly, no matter how long you hold it under the tap. At 8-10 micrometers it fits any crack you can’t see.
What the data says:
- Cyclospora resists chlorine. Routine chemical disinfection used in water treatment, irrigation systems and food processing doesn’t kill it. That’s CDC’s phrasing, not mine.
- Gaseous chlorine dioxide at 4.1 mg/L doesn’t stop sporulation, while the same treatment inactivates Cryptosporidium.
- “Pre-washed” doesn’t mean safe. MDHHS says plainly that rewashing bagged lettuce is unlikely to remove Cyclospora.
- Vinegar, baking soda, commercial produce washes: no evidence.
Wash anyway. Running water mechanically knocks the count down. It guarantees nothing, but paired with peeling and cooking it adds protection.
What actually kills it
From Ortega and Sanchez in Clinical Microbiology Reviews:
| Treatment | Result |
|---|---|
| 158°F / 70°C | stops sporulation, works |
| 212°F / 100°C | works |
| -94°F / -70°C | works |
| 205°F / 96°C for 45 seconds | doesn’t fully prevent it, a short shock isn’t enough |
| 122°F / 50°C for 1 hour | fails |
| -4°F / -20°C for 2 days | fails, your home freezer isn’t enough |
| 99°F / 37°C for up to 4 days | fails |
| 39-73°F / 4-23°C (refrigerator) | no effect on sporulation at all |
The practical takeaway is inconvenient. The only reliable way to kill Cyclospora at home is heat to 158°F (70°C) or above. Not freezing. Not soaking. Heat.
Which is why every item in the July 13 MDHHS guidance ends the same way: safest when cooked.
About the produce washers
Sales of countertop “produce cleaners” like the Pure Pod jumped once the outbreak hit the news. They don’t work on this parasite, and the reasoning is short: Cyclospora survives chlorine at municipal treatment concentrations, so a capsule making trace oxidants in a salad bowl for 10 minutes was never in the fight. No manufacturer has tested against C. cayetanensis, and no agency recommends any device.
We took that question apart separately, including the one ozone study people cite and why it doesn’t say what they think it says: Will the Pure Pod work against Cyclospora?
Seven rules that actually lower your risk
Adapted from the MDHHS guidance of July 13, the most specific advice any agency has published on this outbreak.
- Buy whole heads, not bagged mixes. Salad kits and “pre-washed” blends show up in Cyclospora outbreaks over and over. Cut leaf edges give oocysts extra places to grab.
- Strip off the outer 2-3 layers of leaves and throw them away. Wash the inner leaves under running water.
- Wash everything under running water, including what you plan to peel. Scrub firm items like melons and cucumbers with a brush.
- Peel where you can. The parasite sits on the surface, so the skin takes it with it.
- Cook when you can. 158°F (70°C) and up. This is the only guarantee. Spinach, chard, broccoli, peas: anything you can cook is safer cooked.
- Wash your hands before and after handling produce. Wash and sanitize boards, knives, counters. Refrigerate cut and peeled produce right away.
- If you get sick, ask for a Cyclospora test by name. Not a general panel.
Higher-risk groups are chemotherapy patients, transplant recipients, infants and small children, and the elderly. For them dehydration from prolonged diarrhea is more dangerous than the parasite itself, so the rules tighten.
What about swimming? MDHHS addressed this directly: recreational water is not a recognized risk factor for cyclosporiasis, and no link to this outbreak has been found. Don’t swallow lake water as a general principle, but that’s a different story.
The riskiest produce categories
| Produce | Why it’s risky | What to do |
|---|---|---|
| Bagged salad mixes, mesclun | Cut edges plus a long supply chain. Prime suspects in this outbreak | Avoid during the outbreak, buy whole heads |
| Fresh herbs: basil, cilantro, parsley | Eaten raw, complex texture, often imported | Wash thoroughly, add to dishes after cooking |
| Raspberries, blackberries | Surface hairs physically hold oocysts | Wash under running water; cooked for high-risk groups |
| Snow peas and sugar snap peas | Historically tied to large outbreaks | Wash and rub the surface, safest cooked |
| Broccoli, cabbage, green onions | Featured in earlier outbreaks | Wash, cook where it fits the dish |
| Blueberries, smooth-skinned fruit | Oocysts attach poorly to smooth surfaces | Lower risk, wash anyway |
| Canned and frozen cooked vegetables | Heat-processed | Very low risk |
Water: where the line actually sits
Cyclospora arrives on lettuce with water. Not through the air, not from soil on its own, but from irrigation. So the real question behind “how safe is this salad” is “what was it watered with.”
Fields are irrigated with untreated water. A river, a canal, a pond, a holding reservoir. That water meets no drinking standard, it’s just wet. If there’s a discharge upstream, or the crews working the field have no decent toilets, oocysts end up in the water. Sprinklers spread them across the leaves, they lodge in the microscopic crevices, they mature over 1-2 weeks in 73-81°F (23-27°C) summer heat, and the lettuce ships to a supermarket already infectious. That’s not a theory. It’s the mechanism every major Cyclospora investigation keeps landing on: poor field sanitation plus irrigation water carrying fecal contamination.
Greenhouses are more complicated than you’d like. A greenhouse guarantees nothing by itself. The walls aren’t the point, the water source is. Some greenhouse operations draw from the same open sources fields do. When Little Leaf Farms publicly distanced itself from this outbreak, its lead argument was water: they irrigate with treated water. Not “we have a greenhouse.” They know where the weak point is.
City farms are a different case. Most vertical farms and serious hydroponic systems run on reverse osmosis water. That brings in a mechanism no chemical treatment has.
RO doesn’t try to kill the parasite. It just doesn’t let it through.
RO membrane pores are around 0.0001 micrometers. A Cyclospora oocyst is 8-10 micrometers, roughly a hundred thousand times bigger. Shell toughness, oxidant dose, contact time: none of it applies. The oocyst physically cannot pass the membrane, like a brick through a window screen. Bacteria, viruses and Cryptosporidium don’t get through either.
That’s the difference between the two approaches. Chlorine fails against Cyclospora because chlorine tries to kill it, and the shell is built for that fight. RO works because it doesn’t kill, it excludes. Different mechanism, different result. It’s also why the countertop gadgets fail: they’re trying to kill, only far more weakly than chlorine does.
So lettuce from a city farm grown on RO water indoors really does sit in a lower risk category for cyclosporiasis. There’s no entry point for the parasite: clean water, no field, no runoff, no irrigation canal 2,000 miles away.
One caveat, or this becomes an ad instead of a fact. RO removes water as a route. It doesn’t remove the other routes. A pathogen can enter a city farm on seed, on a worker’s hands, on a pest, during packing. That’s exactly how a 2021 Salmonella Typhimurium outbreak traced back to packaged greens from a closed hydroponic operation. Water had nothing to do with it. Process did. Rodney Rohde of Texas State University framed it for Newsweek: home-grown and farmers’ market produce can be lower risk if it’s grown with safe water, good sanitation and sound agricultural practice, but it isn’t inherently safer.
The accurate version: RO water removes the main and practically only path Cyclospora has. The risk isn’t zero. It moves into the category of risks you can manage, because they live indoors and inside a process. An irrigation canal in another state doesn’t.
If you grow your own
For Cyclospora to reach your microgreen tray you’d need human fecal contamination plus 1-2 weeks of oocysts maturing at 73-81°F (23-27°C). In an apartment with tap or filtered water, clean trays and washed hands, that chain doesn’t exist.
Your real risks growing microgreens are different: mold on the substrate and seed-borne bacteria. Those are solvable, and they’re yours to control. Unlike a head of lettuce that traveled 2,000 miles, where you’ll never learn what the water on that field was like.
Bottom line
- Cyclosporiasis is a parasitic intestinal infection. The source is fresh produce contaminated with human feces weeks earlier. It doesn’t spread person to person.
- Michigan: 4,312 cases and 102 hospitalizations as of July 16, against a baseline of 50 a year. No source identified; lettuce and salad greens are the working suspects.
- Nationally: roughly 7,000 cases across 34 states, more than six times last year’s pace. 2026 is already a record.
- Washing guarantees nothing. Chlorine fails. Freezing fails. Vinegar fails.
- Heat to 158°F (70°C) works. So does stripping outer leaves and peeling.
- Produce washers don’t work. No data, no mechanism, no mention in any official guidance. What they deliver is false confidence.
- Water decides everything. Fields run on untreated irrigation water, and that’s the main route in. RO on a city farm screens the oocyst out physically, so that route is closed.
- The best long game is shortening the supply chain on the greens you eat raw. Ideally to a few feet.
If you’ve had watery diarrhea coming in waves for weeks, that isn’t “something I ate.” See a doctor and ask for a Cyclospora test by name.
