Last month, the New England Journal of Medicine published a study made for television drama. A team of twenty-two researchers traced an E. coli outbreak to flour using the kind of cool epidemiological sleuthing you'd expect from an episode of House.
The T.V. version of the story would begin with fifty-six people in their homes schools, and workplaces, doubling over with abdominal pain, vomiting, and fleeing to the bathroom at inopportune moments. The next scene would show these people at doctors' offices, urgent cares, and emergency rooms across twenty-four states. All of them were confirmed cases of E. coli infection.
Before we go any further with our story, remember that our intestines are teeming with E. coli, as are the intestines of many other animals. Some types of E. coli that live in animals are dangerous to humans, including Shiga toxin-producing E. coli (STEC). A STEC infection generally presents with vomiting and diarrhea, along with fevers and abdominal pain. According to the Centers for Disease Control, most of the news cases you hear about foodborne E. coli stem from STEC. Complications from an E. coli infection can be life-threatening. Some patients experience hemolytic-uremic syndrome, a kind of kidney failure that can lead to death.
The investigation, a joint effort between the Centers for Disease Control and state health departments, looked for patients with a specific type of E. coli, STEC 0121. That's actually an oversimplification, but if you think "STEC 0121" is unmemorable, know that what the researchers were actually looking for was a gel electrophoresis pattern of EXKX01.0001/EXKA26.0001, EXKX01.0001/ EXKA26.0313, EXKX01.0389/EXKA26.0001, or EXKX01.0395/EXKA26.0001.
The patients determined to have STEC 0121 infections were given a questionnaire to determine what exposures all of the people had in common. In the next chapter of our medical mystery drama, this would be filmed as a montage read directly from the CDC's exhaustive 14-page script:
"Now I have some questions about beef products..."
"Now I have a few questions about eggs, dairy, and cheese products..."
"Now I have questions about herbs and sprouts..."
We have to pause our story here and ask: how did the CDC find those patients to question? The Foodborne Disease Active Surveilance Network (FoodNet) tracks nine foodborne pathogens, including Salmonella and STEC. It's likely that the cases were reported to FoodNet for follow-up. If you've had a kid with the chicken pox and later gotten a call from a state department health worker, you've experienced a similar kind of program, which is aimed at tracking and limiting the spread of illnesses.
Back to the questionnaire. It is a fascinating research tool, demonstrating how creative researchers can be in determining the source of foodborne pathogens. For example, in Section 5: Sources of Food at Home, patients are asked to explain where they bought food (grocery store, warehouse store, farmers' market). They are also asked for permission to use any membership card numbers so that researchers can retrieve a complete list of exactly which items they had purchased.
Despite that thorough questioning, the researchers came up blank. The questionnaires revealed many patients had consumed leafy greens, but there was no apparent common source.
So the researchers got on the phone and conducted open-ended interviews with ten of the patients, including questions about every food eaten in the week before the illness started and where it was eaten. In the T.V. version, those phone calls would start overlapping as patients revealed they had baked or eaten homemade baked goods.
Of the ten interviewees, five remembered baking. Of those five, four remembered tasting raw dough. Of those four, three remembered the brand of flour they had used. Of those three, two still had their flour in its original packaging. The FDA retrieved samples of the flour from those two households. What they found was made for T.V.: the bags came from the same facility, and were packaged just one day apart.
Meanwhile, other researchers searching for additional STEC cases identified three kids who developed STEC infection symptoms after playing with raw dough at restaurants. The children who had gotten sick had all eaten at restaurants who received flour from the same supplier as the two bags of flour retrieved from patients' homes.
The final scene would feature the massive recall of flour. (The study itself does not name the flour producer, but a quick Google search will let you know if was General Mills.) All of the now healthy patients would be sent home. One of them--let's imagine a mom of two young kids--promises her doctor never to let them eat raw dough again.
The show's epilogue would add just one more mystery. Although flour tested in patients' homes and from the facility confirmed the presence of STEC in the flour, the source of that contamination was never discovered. Testing at the flour facility suggested that it was not the facility itself, but perhaps its suppliers. The researchers' best guess is the wheat, which could have been contaminated by manure-dropping animals or animals closely associated with E. coli like white-tailed deer.
This is an investigation worthy of its own television show called Contamination Nation. The researchers confirmed that flour, not generally thought of as a likely source of infection, can be contaminated. That's a valuable lesson in the importance of checking one's premises. Just because something is thought to be improbable doesn't mean that it's impossible.
The study also indicates that resulting infections from STEC-contaminated flour can be serious. One-quarter of the patients included in this study were hospitalized, and one experienced kidney failure. One reaction to this news might be to never eat raw dough again. Another less charitable reaction may be to smugly judge all those Pinterest moms and their seasonally-scented, improbably beautiful homemade dough recipes for unwittingly poisoning their children.
But those are the wrong lessons.
The E. coli "outbreak" sounds enormous, spanning 24 states and a nationwide recall of a huge brand. But there were only fifty-six cases. Assuming that there are more than fifty-six raw dough eaters in America, that means a lot of people ate dough made with contaminated flour and did not get sick, or did get sick, but not sick enough to seek medical attention.
The number of cases seems even smaller when you consider just how much flour was recalled--about 45 million pounds. How many bags of flour were included in the recall? How many people were estimated to have eaten raw flour from those bags? How does that number compare to the number of people who got sick? What was the relative risk of people who consumed that flour? All of these questions could help determine the relative risk of consuming STEC-contaminated flour.
None of these questions are a criticism of the researchers' findings. Those questions are outside the scope of the study, which is making the simple but important point that flour can be a carrier for E. coli. That information will be of use to food safety experts and state health departments, who can respond by changing industry-wide food safety practices, like the heat-treatment of flour instituted after pre-packaged cookie dough was contaminated with E. coli in 2009.
Those questions are important to anyone reading the next "don't eat that!" news story. Reporting on scientific studies often scares us with rules ("Don't eat raw flour") and an exaggerated view of their consequences ("because you or your loved ones could get sick and die"). That scare-into-submission tactic is problematic because it teaches all of us to be poor risk assessors. Slate's L.V. Anderson criticizes the "abstinence only" approach instituted by the FDA after the 2016 flour recall. Instead of blindly following that directive, Anderson advocates a calculated risk strategy: "You should educate yourself about the risks of raw flour and make your own decision about whether they're acceptable, instead of letting the FDA's doom-and-gloom warnings automatically scare you away from one of life's most delicious treats."
56 sick people. 45 million pounds of flour. The indulgence of a snuck bite of raw chocolate chip cookie dough. Each of these pieces of information deserves weight in our risk calculations about household dangers.