Opinion: Why we still don’t know what foods are the source of the cyclospora outbreak

OpinionFirst Opinion Why we still don’t know what foods are the source of the cyclospora outbreak Government cuts have contributed to an information vacuum on foodborne illness Manage alerts for this article Email this article Share this article By Katrine WallaceJuly 15, 2026 Wallace is an epidemiologist at the University of Illinois-Chicago School of Public Health. As of Tuesday, a parasite called cyclospora has sickened nearly 7,000 people in 34 states so far this summer. On Monday, Michigan health officials announced their first potential source: lettuce and salad greens. Their advice was sound: Buy whole heads, discard the outer leaves, wash what’s left. But it landed after vinegar rinses and peeling rituals had circulated online for weeks. Michigan’s announcement was careful: Lettuce keeps surfacing in interviews, other foods cannot be ruled out, and no grower or supplier has been named. That is what a “potential source” is — a hypothesis strong enough to keep pulling on. Weeks into one of the largest cyclosporiasis surges on record, we still cannot answer the question people need answered: Which food is making us sick? The Centers for Disease Control and Prevention cannot yet say whether this is one outbreak with a common source or several unconnected clusters.Advertisement That uncertainty is not a footnote. It is the problem. Cyclospora is hard to trace under the best conditions, and these are not the best conditions: The systems designed to compensate for the difficulty — surveillance, staffing, traceability — has been cut, thinned, or deferred. I am an epidemiologist and spend much of my time pushing back on health misinformation online. I saw the folk remedies early — vinegar soaks, warnings to swear off berries — building for weeks, ahead of any official word. This week the feeds finally had something official to cling to: “Lettuce is a potential source” became “it’s the bagged salad.” Google searches for “cyclospora” spiked to their highest level of the year — a crude measure of how many people were searching for guidance that does not yet exist. An information vacuum does not stay empty. People are not irrational when they fill it; they are trying to protect their families with the only tools within reach.Advertisement But when certainty is the one thing nobody can honestly offer, certainty gets invented. And invented certainty carries a cost. If lettuce is not the source — or not the only one — people who give up salad may feel protected while still buying the food that is actually contaminated. False reassurance is not a neutral outcome. It is a risk that looks like safety. This week the vacuum also swallowed a corporation. Signs went up in some Taco Bell restaurants announcing that lettuce and other fresh toppings were unavailable because of a nationwide recall. There is no such recall. The FDA has issued no notice naming Taco Bell or any supplier, and the company has not explained the signs. Federal and state investigators are now looking at the chain — but some people who got sick never ate there. So, we have a second hypothesis, not an answer. When the system that is supposed to name a contaminated food cannot do it, the naming happens anyway: on Instagram, and now on a fast-food menu board. Here is what gets lost when the hedges fall off: None of this can be fixed at the kitchen sink. Cyclospora cannot be reliably washed off. Vinegar soaks do not work. Michigan’s own guidance says as much: “Pre-washed” is no guarantee, and rewashing is unlikely to dislodge a parasite that arrives in irrigation water and clings to the leaf. The only sure way to kill cyclospora is to cook it, at 158 degrees Fahrenheit or hotter. Everything else (rinsing, soaking) is risk reduction. It is worth doing, but no one at a health department ever pretended it was a shield. Meanwhile, we still can’t confidentally name the food at the heart of the outbreak. Not for lack of trying. Labs can fingerprint salmonella by whole-genome sequencing and tie to a common source within days, so if public health infrastructure is lacking, the lab can pick up the slack. Cyclospora is a different story. The molecular genotyping tools for cyclospora are far less mature: They can tell investigators which cases cluster together, but not which field or shipment to pull. Many routine stool panels do not even test for cyclospora, so countless infections are never diagnosed. That leaves mostly epidemiological outbreak investigation: investigators asking sick people what they ate. Because the parasite can take two weeks to sicken you, and cases weeks more to be reported, they are asking about meals from up to six weeks ago. Try to remember what was in your salad in early June.Advertisement None of that would be easy even in a fully staffed system. But there are fewer people doing the work, with fewer tools. Hundreds of federal food-safety staff have been cut, and last July the CDC narrowed FoodNet — the sentinel network that provides early warnings of foodborne illness — from eight pathogens to two, dropping cyclospora altogether, citing inadequate funding. We almost had another tool here: the Food Traceability Rule, which would require the records that make a rapid traceback possible. But implementation was pushed from 2026 to 2028 — urged by industry, ratified by Congress. The lot-level records that would let investigators walk this summer’s lettuce back to a field do not yet exist. None of these policy decisions singularly explains why we cannot yet name the food. But you cannot thin a system, defer the tools that would speed it up, and then wonder why the public is left guessing. Michigan shows what that costs. The state kept its own cyclospora surveillance running, interviewed thousands of patients, and logged 3,309 cases (as of July 14) against the roughly 50 it sees in a normal year. That effort produced this week’s potential source lead — and it is as far as that effort reaches. A state can generate a hypothesis. It cannot match its cases against Ohio’s and New York’s or trace a shipment back to the farm. Michigan may not be the epicenter. It may only be the state still rigorously counting — leading the tallies not because it is worst hit, but because it kept interviewing patients after other states had moved on. A state with a smaller count might have just as many cases and simply stopped looking for them. The bottleneck is not effort. It is connection, and connection is national work. That is fixable, and none of it is expensive by government standards. Restore cyclospora to FoodNet. Stop deferring the Food Traceability Rule — 2028 is a choice, not a deadline, and Congress already ratified an earlier one. And fund the genotyping tools and staff that would let a state’s hypothesis become a national answer.Advertisement Millions of Americans are still searching online what foods to avoid, and watching social media prevention rituals, trying to solve a problem that was never theirs alone to solve. The question is not whether they should peel their lettuce. It is whether we will rebuild the systems that can answer the simplest — and most important — one: Which foods are making us sick? Katrine Wallace, Ph.D., is an epidemiologist at the University of Illinois-Chicago School of Public Health. Letter to the editor Have an opinion on this essay? Submit a letter to the editor. CDC, infectious disease, Policy Submit a correction requestReprints Katrine Wallace Newsletter The smartest thinkers in life sciences on what's happening — and what's to come Recommended First Opinion July 14, 2026 It looks like your doctor and talks like your doctor. But it’s not your doctor First Opinion July 13, 2026 The shortage of forensic pathologists is hurting justice, public health, and families Advertisement First Opinion July 13, 2026 Is there a doctor on board? Yes, and airlines depend on it First Opinion July 10, 2026 The primary care crisis paradox First Opinion July 9, 2026 Your grandparents are using cannabis. Doctors can help them do it safely Subscriber Picks