In July 2026, U.S. health agencies linked one of the largest cyclosporiasis outbreaks on record to iceberg lettuce, and within hours the familiar advice was circulating again: wash your greens. The advice is not wrong. It is simply aimed at the wrong stage of the problem.
Nor is the outbreak an accident of one bad season. CDC outbreak attribution work covering 1998 through 2008 assigned 46 percent of foodborne illnesses to produce, with leafy vegetables alone accounting for 22 percent — more than any other single commodity, and the second most frequent cause of foodborne hospitalizations. Lettuce is also the second-most consumed vegetable in the country after potatoes, which means the exposure base is enormous before any pathogen enters the picture.
The vulnerability is structural. Lettuce grows outdoors in open fields, sits inches from soil, and reaches the plate without ever passing through a kill step — the heating, pasteurization or freezing that eliminates pathogens in most other high-risk foods. Its crevices and folds shelter organisms from rinsing. And bagged mixes pool greens from several farms at once, so a single contaminated lot can seed an entire production run and travel across state lines inside thousands of identical packages.
All of which relocates the useful question. If the decisive contamination happens weeks earlier and hundreds of miles upstream, what matters is not how to clean lettuce but which remaining consumer decisions measurably change the odds. Three candidates exist: washing it, choosing what form to buy, and controlling its temperature. The published evidence ranks them in nearly the reverse of how most kitchens prioritize them.
The contamination arrives through the sprinklers, and on the wind
Two documented routes now explain most of what happens in the field. Researchers at Cornell published work in Scientific Reports in May 2025 attributing 52 percent of romaine-linked E. coli O157 illnesses to overhead spray irrigation using untreated surface water — water applied directly onto the leaf. Their modeling found the risk could be cut by as much as 96 percent through water treatment or a switch to drip or furrow systems that never touch the edible tissue.
The second route is stranger. Between May 2019 and April 2024, the FDA ran an environmental study across roughly 54 miles of the Yuma, Arizona growing region, sampling irrigation water, soil, sediment, airborne dust, livestock manure and plant tissue. A concentrated animal feeding operation holding more than 80,000 cattle sits near some of the ground examined. Shiga toxin-producing E. coli recovered from air samples genetically matched strains found in water, sediment and plant tissue — evidence that dust can carry pathogens from adjacent land onto crops.
The agency has been careful not to convert proximity into proven causation for any individual outbreak, and that restraint is appropriate. Still, both routes are now documented rather than assumed, and together they explain why leafy greens keep generating outbreaks that no amount of downstream diligence has managed to stop.
Rinsing is the weakest lever you hold
The current outbreak makes the limits of washing unusually clear. Cyclospora cayetanensis survives inside a tough oocyst wall that resists chlorine and household produce washes, and CDC states that no method is known to completely eliminate the parasite from contaminated fresh produce. Washing under running water remains recommended, because it removes soil and may reduce other contaminants, but it cannot be relied on to remove this one.
Some infectious disease specialists suggest that vigorous washing with physical scrubbing is the best available step against Cyclospora, and that is reasonable advice. It is worth noticing, though, that best available is not the same as reliable, and that the agency language is deliberately hedged. Scrubbing a head of romaine is a mechanical action against an organism that lodges in microscopic folds and clings to damaged tissue. It helps at the margin. It is not a barrier.
A cut leaf behaves like a different food
The form you buy matters more, and there is a clean experiment showing why. Work published in Applied and Environmental Microbiology tracked E. coli O157:H7 on lettuce left intact, bruised, cut into large pieces, or shredded. Within four hours, populations rose 1.95-fold on intact leaves, 3.99-fold on bruised, 4.54-fold on cut and 11.05-fold on shredded.
The mechanism is straightforward. Damaged plant cells leak their contents, and the sap released at a wound is a nutrient supply the pathogen exploits, while the injured surface gives bacteria a preferred place to attach. A shredded leaf is not merely a leaf with more surface area. It is a leaf converted into a growth medium.
That experiment ran at 28 °C, well above refrigeration, so the figures describe cut greens sitting out rather than cut greens held cold. But the ranking they establish compounds with the pooling problem: a bagged mix is simultaneously the product most likely to contain material from a contaminated farm and the product whose physical state most favors multiplication. It is why current CDC guidance points to whole heads instead of prewashed bagged mixes, with the outer two or three layers of leaves discarded and the inner ones washed.
The few degrees that decide whether anything multiplies
The third lever is the one almost nobody uses, and the evidence behind it is the strongest of the three. USDA researchers stored packaged fresh-cut romaine and iceberg salad at different temperatures and tracked the pathogen. At 5 °C, E. coli O157:H7 survived but its growth was limited. At 12 °C, populations rose by more than 2 log CFU per gram within three days and kept climbing through the rest of storage.
One detail in that study deserves more attention than it has received: the salad’s quality was still fully acceptable at the point where pathogen growth had reached a statistically significant level. The lettuce looked fine. The common heuristic that a tired-looking bag is the one to discard is not a safety test, because the dangerous change arrives before the visible one.
That makes refrigerator temperature a genuine intervention rather than a housekeeping detail. A nationally representative survey of 2,060 U.S. households, published in the Journal of Food Protection, found that only 11 percent owned a refrigerator thermometer and roughly three in ten kept their refrigerator above the recommended temperature. The instrument that would tell you whether your greens sit in the growth band or safely below it costs a few dollars, and nine out of ten kitchens do not have one.
Where the temperature argument stops
This is where the case has to be bounded honestly, because the pathogen behind the current headlines does not obey it. Cyclospora does not multiply on produce. Its oocysts are not infectious when shed and require one to two weeks of sporulation in the environment before they can infect anyone, which is also why person-to-person transmission is considered unlikely. The dose on a contaminated head is fixed before purchase, and no refrigerator setting reduces it.
Temperature control governs the bacteria, not the parasites. Against Cyclospora the only reliable consumer step is heat, with CDC putting inactivation at 70 °C and treating cooking as the safest option for greens that can be cooked during an active outbreak. Presenting cold storage as a general shield would be the kind of overreach that makes food safety advice easy to dismiss.
The mistake that turns an infection into kidney failure
One further asymmetry deserves space, because it inverts a common intuition. In a prospective cohort of 71 children under 10 with E. coli O157:H7 diarrhea, hemolytic uremic syndrome developed in 14 percent, and antibiotic treatment carried an adjusted relative risk of 17.3 for developing it, with a confidence interval running from 2.2 to 137. The proposed mechanism is that antibiotics increase the Shiga toxin available for absorption from the gut.
Cyclosporiasis, by contrast, is treated with antibiotics. Two illnesses that can each begin with days of watery diarrhea after a salad therefore call for opposite responses, and only a stool test separates them. That is a strong argument against self-medicating a foodborne illness, and it is the part of prevention no purchasing decision covers.
The uncomfortable conclusion is that the most visible safety ritual is the least effective one, while the two measures with measurable effect are unglamorous: buy leaves that have not been cut, and keep them cold enough that nothing on them can grow. Both are worth doing, and both are damage control. The number that should unsettle anyone is the 96 percent risk reduction the Cornell modeling attributes to changing how fields are watered — a single upstream decision that dwarfs everything available at the kitchen sink. Consumers are managing the residue of an agricultural design choice. Managing it well is worth the price of a thermometer. It is not a substitute for fixing the design.
Discover more from TekCrispy
Subscribe to get the latest posts sent to your email.



TekCrispy