The German E. coli outbreak is demonstrating the difficulty of identifying a “smoking gun” in outbreaks of foodborne illness.

Some are asking, “Where is the evidence?” By “evidence,” I believe they mean identifying the outbreak strain of bacteria in a food source that is genetically identical to clinical isolates by Pulse-Field Gel Electrophoresis (PFGE).

There is another kind of evidence that is just as important … it’s the evidence provided by shoe-leather epidemiology. Shoe-leather epidemiologists (the name is used to describe applied epidemiology performed mostly in the field, outside of an office, or “wearing a hole in one’s shoe soles”) work in local and state health departments and at the Centers for Disease Control and Prevention (CDC) and conduct outbreak investigations to try to find that smoking gun.

They design case-control studies, develop questionnaires, interview case patients and controls, and analyze the data to determine statistically significant associations with food or other exposures. These identified food items are then traced back to restaurants, food distributors, and manufacturers where food and environmental sampling can be done.

Samples of the food from the same batch eaten are rarely still available by the time the connection to the outbreak is made. You can’t sample an entire farm or factory.  Epidemiology may be the only evidence you have.

Some are criticizing the German health authorities for blaming cucumbers from Spain too soon and for not focusing on sprouts earlier. Sprouts are served in many sandwiches and dishes at restaurants and many people don’t even recall eating them.

In 2009, a multi-state outbreak of Salmonella Saintpaul infection was tied to alfalfa sprouts from a local grower in Omaha, Nebraska by shoe leather epidemiology (Outbreak of Salmonella Serotype Saintpaul Infections Associated with Eating Alfalfa Sprouts — United States, 2009).  The outbreak strain was never grown from sprouts or seeds.  Epidemiology was the sole evidence that prompted a recall of sprouts.      

We interviewed 32 case patients as part of the case control study conducted by health departments in Nebraska and Iowa.  Only 8 answered “yes” to the question, “Did you eat any sprouts?” But upon further inquiry, 27 had actually eaten them. Most of these sprouts were included in sandwiches or salads eaten at restaurants.

Also, according to the Control of Communicable Diseases Manual, the incubation period for Salmonella is from 6 to 72 hours.  In 13 case patients where the date of exposure to sprouts could be determined, 7 had eaten them more than 72 hours prior to becoming ill, up to 8 days.  It would have been very easy to miss the sprout association.

In another national outbreak of Salmonella Saintpaul infection in 2008 (Outbreak of Salmonella Serotype Saintpaul Infections Associated with Multiple Raw Produce Items — United States, 2008), CDC initially identified fresh tomatoes as the probable source of food contamination.

But as the outbreak continued on, more evidence surfaced that pointed to jalapeño peppers, which are commonly eaten with tomatoes in items like fresh salsa.  I can imagine many case patients who were interviewed answering “no” to “Did you eat jalapeño peppers?” even if they had eaten salsa at a restaurant.

Identifying the culprit in a foodborne outbreak is not an easy task.  Much focus has been placed on finding a genetically matched outbreak-causing organism in a food source, or “the smoking gun,” but not on the epidemiology that points us to that source. This has prompted arguments for increased spending on food safety agencies to increase regulation and monitoring of food production.  While that is an important area, we should not forget about the shoe-leather epidemiologists in our public health agencies who produce the evidence that points us to those food sources in the first place.


Dr. Anne O’Keefe is the Senior Epidemiologist at the Douglas County Health Department in Omaha, Nebraska.