The Centers for Disease Control and Prevention (CDC) has issued a series of disquieting food warnings over the past few days after an E. coli outbreak linked to romaine lettuce caused at least 65 people in the United States and Canada to fall ill (and left countless more scrambling to find a substitute for their Thanksgiving salads).
The CDC had good reason to be ominous: Only a month earlier, ground beef tainted with another strain of E. coli sickened at least 18 people and left one dead. But even that paled in comparison to the multi-state E. coli outbreak this past spring that infected over 200 and killed five. (Here romaine lettuce was again an associated ingredient.) And E. coli is hardly the only pathogen afflicting American eaters.
The CDC's index of Selected Multistate Foodborne Outbreak Investigations, a running record of the year's large and notable outbreak investigations, also shows 15 salmonella outbreaks and multiple outbreaks of listeria and cyclospora. The CDC's investigations list currently features 21 cases—the most, according to CNN, undertaken by the agency in the last decade.
Predictably, media reactions reached for superlatives. "This Is a Record Year for Foodborne Illnesses," read a Grub Street headline. "This [is a] record-setting year for foodborne illness," declared the Tampa Bay Times.
Those clippings are clicky and grim, but they're also at least somewhat misleading, because the current state of foodborne illnesses in the U.S. is actually, well, quite sunny.
The CDC's Selected Outbreak Investigations list is not an exhaustive list of all foodborne outbreaks in the U.S. It's not even a full list of all the foodborne illness outbreaks that were investigated by the CDC this year. It's just a list of outbreak investigations the CDC has deemed noteworthy. For example, the selected list shows 14 outbreaks in 2016, but, according to the CDC dashboard's full list of outbreaks, 2016 had 3,950 outbreaks across the U.S.
According to Benjamin Chapman, a professor of food safety at North Carolina State University, outbreaks are usually placed on the CDC's select list of investigations because they contain new information relevant to the field of public health; if there's something that surprises experts about an outbreak—like, say, the food of transmission—it goes on the list. "If you look at [the associated foods in the selected outbreaks for] 2007, we had pizza, and we had pot pies, and we had ground beef patties. These are all prepared type foods. These are kind of notable, because before 2007 we weren't investigating many frozen, prepared foods that made people sick," Chapman says. "In 2010, they selected their investigation into 'Frozen Mamey Fruit Pulp,' which was notable because we'd never looked at mamey as a source of foodborne illness before."
Additionally, experts in foodborne illness caution against attempting to compare the number of outbreaks across different time periods, because many outbreaks stretch across different months and years. "Pathogens don't care what month it is. Outbreaks don't care about the year," Chapman says. "We have outbreaks that span two years. So to try to say 'this is a January outbreak' or 'this is a December outbreak' doesn't make sense to me." Instead, experts mostly focus on the number of individual foodborne illnesses—tracking whether the raw number of people getting sick has increased.
"We have a lot of foodborne illnesses each year. The CDC estimates that one in six Americans are sickened each year, and 3,000 die. And only a fraction of those get reported to [state and federal] health department[s]," says Barbara Kowalcyk, an epidemiologist and professor of food science at Ohio State University. Over 80 percent of foodborne illness are considered "sporadic," meaning they haven't been associated with an outbreak of infections. "Doesn't mean they're not part of one," Kowalcyk says. "It just means that they're not linked to an identified outbreak."
When there's an increase in outbreaks without a rise in individual illnesses, it means that technology and epidemiologists have gotten better at identifying why people are getting sick—that more illnesses which might have been considered sporadic in past years are now successfully linked to a common source. And the CDC's latest available charting of foodborne illness rates shows that they are not on the rise in the U.S. Between 2008 and 2015, E. coli infections dropped by 30 percent, and most foodborne illnesses saw no change.
Accordingly, any increase in identified outbreaks is good, likely the result of two major technological changes in roughly the last two decades that have led to scores more outbreaks being successfully identified. In 1996, the CDC began using Pulsenet, a system that tracks foodborne illnesses across the country by comparing the DNA fingerprint of discovered pathogens to see if they are similar. "Our foodborne disease surveillance systems were established in 1996," Kowalcyk says. "They have identified a large number of outbreaks over the years, and have really led to significant public-health impacts."
Starting in 2008, the public-health community began using a new method of DNA fingerprinting called whole genome sequencing, which has led to a large spike in detected outbreaks. "I get leery because I don't think we can compare pre-2008 to today, because we measure things differently," Chapman says. "We're getting better at detecting the outbreaks, and there are better-trained public-health individuals now looking to solve foodborne illness outbreaks than we've ever had."