In Food Safety Regulations, There's a Surprising Scarcity of Research
While health and safety regulations in the United States are well-funded, studies that assess their efficacy aren’t.
In preparation for this month’s column, I set alerts to notify me of news about food safety. At least one or two popped up every day:ing: A multistate outbreak of salmonella tied to ground beef resulted in nine hospitalizations and one death. Ping: A flour recall linked to an outbreak of E. coli that sickened 21 people in 9 states expands. Ping: A mysterious multistate listeria outbreak claimed two lives.
But on November 22, the warnings really hit home. The U.S. Centers for Disease Control and Prevention advised people not to eat romaine lettuce from the Salinas Valley in Northern California (or of an unknown province) because it was linked to an E. coli outbreak involving the same virulent strain that caused several other outbreaks in the previous two years. Since January 2018, the CDC has reported that 474 Americans have fallen ill, 219 have been hospitalized, and six have died in multistate outbreaks of E. coli linked to leafy greens — primarily romaine.
I went to my refrigerator and peered into the crisper at a half-eaten head of romaine; I had no idea where it was grown. My husband and I felt fine. My two dogs, watching with great interest as they love nothing more than a crunchy romaine stalk, appeared healthy enough. Still, better safe than sorry, I thought — so I pitched the lettuce.
Even if you don’t get emails about shards of plastic in your Cheese Nips or botulism-causing bacteria in your smoked salmon, you may still have an uneasy sense that something’s up with our food supply. In 2019, the CDC announced 17 notable outbreaks of foodborne illness spanning multiple states — the second-highest number since the agency started publicizing such investigations in 2006. The record of 24 major outbreaks was set the previous year.
And while recalls for produce and processed foods have remained largely unchanged in the last few years, they’ve been rising for meat and poultry, according to a recent report from the U.S. Public Interest Research Group Education Fund, a federation of non-profit organizations that advocate for consumer issues. Between 2013 and 2019, the report states, overall meat, and poultry recall increased by 65 percent, with Class 1 — the category most likely to pose a risk to human health — surging by 85 percent.
Is our food becoming less safe? It’s hard to say for certain. Overall, the number of foodborne illnesses reported to the CDC has remained fairly constant over the last decade. And one reason for the increase in recalls and outbreaks, according to Daniel Payne, a CDC epidemiologist and lead of the agency’s Foodborne Diseases Active Surveillance Network (FoodNet) is that scientists have better tools for diagnosing and connecting infections. “Technological advancements are allowing us to detect things that we normally would have missed before.”
Still, after adjusting for better detection, a recent CDC report found that the incidence of some infections, including virulent Shiga-toxin-producing E. coli such as those involved in the lettuce recall, and two common infections linked to poultry — campylobacter and salmonella — increased in 2018 compared to the previous three years.
These sorts of countervailing data points suggest to Timothy Lytton, a law professor at Georgia State University and the author of the new book “Outbreak: Foodborne Illness and the Struggle for Food Safety,” that there isn’t really a reliable basis to say just how effectively our food safety net — a complex matrix of science and policy — is really performing. The implication is that we may be spending hundreds of millions of dollars on ineffective measures or guarding against pathogens that are relatively benign, while lethal strains still find ways to slip through.
“It’s actually something nobody likes to talk about,” Lytton told me. While we “spend an enormous amount on health and safety regulation in the United States,” he said, “we spend tiny amounts of money figuring out whether any of it actually improves health and safety. We just don’t know.”
While it’s true that we are missing needed evidence to link interventions to illnesses, said Michael Doyle, a retired regents professor of food microbiology at the Center for Food Safety of the University of Georgia, we could do more to chip away at persistent problems if regulators were better attuned to the evidence at hand. “Science is evolving, obviously,” he said, “but we’ve got some pretty good data now to enable them to make better decisions.”
OF COURSE, Americans spent the earliest decades of industrial food production with no government oversight at all. That all changed — and no doubt for the better — with the passage in 1906 of the Meat Inspection Act and the Pure Food and Drug Act, which put regulation of food safety squarely within the federal government’s purview. That oversight extends to this day — though it’s an imperfect regime.
In the modern era, the event that thrust the issue of food safety most squarely into the American consciousness was a deadly outbreak caused by undercooked Jack in the Box hamburgers contaminated with E. coli in 1993. Hundreds were made ill, four children died, and dozens of people suffered from kidney failure. In the wake of the Jack in the Box outbreak, the government required industry to assume more responsibility for preventing bacterial contamination. Among other requirements, the U.S. Department of Agriculture has established a zero-tolerance policy for virulent forms of E. coli in ground beef and evaluates meat and poultry producers on whether they meet performance standards, or maximum limits, for pathogens such as salmonella.
And yet, while it makes sense that less-germy meat will result in fewer people getting sick, there’s surprisingly little research to back that up. Some of the strongest evidence of benefit comes from USDA and CDC studies linking a reduction in the strain of E. coli known as 0157:H7 in ground beef to fewer infections. Doyle notes that other factors also played a role, especially better adherence to guidelines for cooking meat to a high enough temperature to kill bacteria.
But it’s a different story for chicken and turkey, where efforts to curtail campylobacter and salmonella haven’t always had the expected pay off. USDA standards for salmonella, for example, have worked to slash contamination rates in raw poultry, which is responsible for about one out of five salmonella infections, according to a recent CDC report. The “big news” from the latest analysis of the CDC’s FoodNet data, said Payne, is that those concerted efforts to curb salmonella haven’t resulted in fewer illnesses. “Those numbers haven’t gone down in probably 10 years.”
While it makes sense that less-germy meat will result in fewer people getting sick, there’s surprisingly little research to back that up.
According to Doyle, one explanation for why reduced salmonella contamination hasn’t translated to fewer infections is that most of the bacteria we take out are relatively harmless. About 60 percent of salmonella on broiler chicken carcasses is the Kentucky serotype, which causes only about 100 of the estimated 1.35 million cases of salmonella in the U.S. each year. “If you focus on Kentucky as part of the big salmonella pool, you’re not going to have much bang for the buck,” Doyle said. Instead, regulators should target strains with genetic factors that make them virulent and more tolerant of disinfectants, he said. “They need to be more attuned to science.”
The science isn’t quite there yet said Mindy Brashears, a food safety researcher who joined the USDA as the Deputy Undersecretary for Food Safety about a year ago. In a recent report prepared for the USDA, a panel of experts concluded that there isn’t enough evidence to reliably distinguish bacteria based on virulence or other factors. “Salmonella is adaptable and changes continually, thus it is important to address it comprehensively,” she said, explaining that benign bugs can become dangerous after exchanging genetic information with virulent counterparts. “All salmonella can cause illness under the proper conditions.”
The question of whether we are eliminating enough of the right bugs in meat and poultry may be answered in the next few years as new standards take effect and (hopefully) more producers come into compliance. “The current salmonella poultry standards have been in place only a couple of years,” said Brashears. “And the standards for campylobacter and salmonella in other products, such as beef or pork, have just been proposed or will be proposed in the months to come.”
Brashears emphasized all of USDA standards are tied to the federal government’s “Healthy People” goals for reducing foodborne illness, but so far, we’re well short of the mark. In 2018 (the most recent year for CDC data), the incidence of salmonella infections was more than 60 percent higher than the Healthy People 2020 target.
FOR PRODUCE and processed foods, which are regulated by the FDA, the moment of reckoning was a 2006 outbreak of E. coli linked to bagged Dole baby spinach that sickened more than 200 people across 26 states and Canada and led to three deaths. That and other high-profile outbreaks galvanized Congress to pass the Food Safety Modernization Act, or FSMA — a massive overhaul of the food safety system that shifted the FDA’s responsibility from reacting to foodborne illness to working to prevent it.
Nearly a decade later, consumer advocacy groups complain that the FDA has been too slow in implementing FSMA. In an email, an FDA spokesman told me that the agency has made the best of limited resources by focusing first on the seven main rules outlined in the legislation, then turning to other regulations in the statute. That’s begun to pay off, he said, as recalls reached a five-year low in 2019. “This suggests that the preventative measures that are required by FSMA are working and that industry is detecting and responding to problems before products reach consumers.”
And yet, Americans keep getting sick from salads. One of the most contentious delays concerns standards for agricultural water, which the agency pushed out until 2022 to 2024, depending on the size of the farm. Tainted irrigation water is the most likely culprit for 2018’s romaine outbreak, according to an FDA report, and has been implicated in ongoing safety issues with leafy greens.
As with meat and poultry, people disagree on how to best reflect the science through regulation. “Politics twist the policy,” said Trevor Suslow, vice president of food safety at the Produce Marketing Association, an industry trade organization. Standards for agricultural water, he added, “have been really not based on the best available science for the better part of 15 years or more.” For example, the voluntary standards that will likely form the basis for FDA rules focus on generic E. coli, which we’ve long known is not a good indicator of whether water harbors the conclusively virulent 0157:H7 strain implicated in romaine outbreaks or salmonella.
Suslow, who formerly directed the University of California-Davis Postharvest Technology Center acknowledges that regulators are “in a tough spot.” Routine testing for a range of pathogens would be prohibitively expensive — and it’s not entirely clear how to act on results. We lack a scientifically-validated measure of what concentration of bacteria in the water it takes to contaminate produce and, in turn, how that corresponds to consumer risk.
Right now, key research is focused on how bacteria from animal feces make their way to your salad. And irrigation water is just one piece of the puzzle, according to Michele Jay-Russell, program manager for the Western Center for Food Safety at the University of California-Davis. E. coli can also waft in the wind and travel into fields with birds, rodents, and other creatures. New guidelines calling for a wider buffer between fields and feedlots may help, but won’t entirely solve the problem, she said. “We have to coexist with animal agriculture and wildlife,” said Jay-Russell. “You can make barriers, but it’s not theirs. It’s obviously not foolproof.” Ultimately, she predicts, that the answer may lie in a more effective sanitizing step after harvest.
We lack a scientifically-validated measure of what concentration of bacteria in the water it takes to contaminate produce and, in turn, how that corresponds to consumer risk.
The point of contention is whether regulations based on uncertain and incomplete science will better protect the public. Lytton of Georgia State University is doubtful. “It’s not like the science is magically going to appear in 2022,” he said. “These are enormously difficult questions for which we don’t have the tools to answer right now.”
That may be so, but consumer advocates say we have evidence now to know that testing water for E. coli will save both lives and money. “We’ve advocated for them to move forward with that standard,” said Sarah Sorscher, deputy director of regulatory affairs at the Center for Science in the Public Interest, a consumer advocacy group pushing the FDA to get standards in place, “because even though it’s not perfect, it will still provide some sense to the farm of water quality.”
She added that standards can be refined as the science evolves.
I’VE BECOME somewhat inured to the pings in my inbox about the myriad ways food is making people sick. Still, as the romaine outbreak investigation dragged on into January, I switched to making salads with other greens. (The hounds and husband are unsure of kale.) I also treat raw meat and poultry as if they are contaminated: I wash my hands after handling them, disinfect surfaces they touch, and use a thermometer to check that I’ve cooked a dish to a high enough temperature to kill bacteria.
Despite the steady stream of warnings, I’ve actually grown more impressed by how safe our food is. This business of weaving a safety net capable of filtering out disease-causing germs along the often long, convoluted journey food travels from farms to our kitchens is enormously complicated. What we need is more and better science so that we don’t waste time and resources guarding against things that won’t harm us, or letting things that will harm us slip through unnoticed. The dozen or so experts I talked had widely different perspectives on food safety but agree on one point: We are not investing nearly enough in that science.
Despite the steady stream of warnings, I’ve actually grown more impressed by how safe our food is.
One bright spot is that Congress allocated an additional $15 million in funding for the fiscal year 2020 to the FDA for implementing FSMA and to improve detection of foodborne illness. That’s a start, but we need to invest far more in public health surveillance at both the state and national levels said Lytton. “Once you can start to identify the outbreaks and figure out what their source is, you have a much better opportunity to focus your energies on figuring out how to fix that particular source of contamination.”
Jay-Russell said drastic cuts in funding over her career have reduced both the quality and quantity of food safety research. While small grants are still available for small studies lasting a year or two, she said, money has dried up for longer, more complex studies needed to answer nagging uncertainties. “I often am talking to young scientists coming out of their Ph.D. that isn’t going to pursue any research in academia because it’s just too hard to get funding,” she said.
The other thing that everyone agrees on is that, with increasingly better detection, the trend of a rising outbreak and recall notifications is bound to continue. At a recent meeting of public health officials, Doyle said he learned that state health departments are finding 30 percent more outbreaks than a year ago.
“It’s gonna only get worse,” he said, “before it gets better.”