Opinion: The wicked problem that sickens 48 million Americans

Editor’s Note: Martin Wiedmann is the Gellert Family Professor of Food Safety at Cornell University. He received a veterinary degree and a doctorate in veterinary medicine from the Ludwig-Maximilians University in Munich, and a PhD in food science from Cornell. The author of more than 450 peer-reviewed papers, he focuses on farm-to-table microbial food quality and food safety. The views expressed here are his own. Read more?opinion?on CNN.

It may seem as though foodborne disease outbreaks in the US never stop sometimes and there’s no food left untouched by outbreaks. Multistate foodborne disease outbreaks investigated by the US Centers for Disease Control and Prevention (CDC) in 2023 alone involve a staggering diversity of foods, including peaches, nectarines, and plums, ice cream and leafy greens (all linked to outbreaks caused by listeria monocytogenes) as well as cantaloupes, onions, ground beef and flour (all linked to salmonella outbreaks).

Earlier this year, a salmonella outbreak with 26 reported illnesses was linked to commercially produced cookie dough, which contains uncooked eggs or raw flour that may have bacteria like salmonella and E. coli (this outbreak can serve as a reminder to be careful about baking instructions before the holidays!). On Friday,?Quaker Oats announced a recall of granola bars?and other products because of Salmonella; so far there are no confirmed reports of illness related to the products covered by this recall.

However, large national foodborne disease outbreaks that make news headlines are the tip of the iceberg. Additional foodborne disease cases can occur in more localized outbreaks, such as a handful of customers affected by illness linked to a specific restaurant. The CDC has published studies estimating that annually in the US a total of 48 million people get sick due to foodborne illnesses.

So, how is it that a country like the US can’t seem to eliminate these issues?

Essentially, food safety can be considered a wicked problem, which is defined broadly as one with complex roots and many stakeholders that is not so much solved as managed.

Some key reasons that US food safety is so vexing are the diversity of the food supply and the fact that contamination leading to foodborne illness can occur anywhere, despite the best precautions, when we’re dealing with a variety of factors, from the nature of raw materials to the challenges of reducing food safety risks across agricultural environments, processing plants, retailers, restaurants and home kitchens.

Another reason is the sheer magnitude of the food in question. With about 330 million Americans eating on average three meals a day, more than 350 billion servings of food per year are potentially in play. This translates into the need to produce a vast amount of food. For instance, the US raises about 10 billion chickens a year and Americans consume about 6.2 billion pounds of tomatoes a year (roughly 19 lbs per person a year). Importantly, these numbers also clearly indicate why a focus on food testing is not the best approach to a safer food supply.

Not only is it impossible to test that much food, but food testing for microbes is also what’s called “destructive testing,” so once you test the food, it’s gone and you can’t eat it anymore. Furthermore, testing one tomato taken out of a box with 200 does not mean the remaining 199 tomatoes are negative for salmonella even if the one tested is negative.

The US is not alone in dealing with this issue – worldwide, an estimated 600 million cases of foodborne diseases occur annually, resulting in 420,000 deaths. It all adds up to the reality that, even in the best of worlds, food will never be 100% safe.

But the same applies to many daily or routine activities such as driving, and none of them are – or ever will be – risk-free. For example, there were more than 40,000 vehicular deaths in the US in 2021, compared to an estimated 3,000 annual deaths due to foodborne illnesses in the US. Yet, eating is an essential activity, and for many Americans so is driving.

Of course none of this means we should accept the status quo, but it does mean that we need to take the same risk-based approach to food safety that we take with other daily activities.

This risk-based approach applies to individuals as well as the federal government. As consumers, we have to choose the risks we are willing to take.
For example, will I eat a slightly undercooked burger my host serves me at a backyard barbecue (a decision I personally have had to make many times)? Sometimes I do but more often, I will not. Will I drive on a curvy mountain road in a snowstorm at night? Sometimes, but more often I will not (even though I will happily drive that road during the day in the summer – risk is contextual, after all).

Better access to reliable food safety information and improved education should help individuals make better decisions, or at least know when they are taking an elevated risk. On a regulatory basis, the government may take a risk-based approach by not requiring recalls if nominal levels of pathogens are found in some low-risk foods, while having more stringent rules for high-risk foods. For example, Canada allows higher levels of listeria monocytogenes in some foods, including those where this disease-causing microbe cannot grow, while setting more stringent limits for foods that are higher risk. Similarly, government agencies could test high risk foods more frequently than lower risk foods.

To be clear, risk-based approaches to food safety do get tricky at the macro level. For example, should there be more limits on sales and distribution of high-risk products (such as raw milk, which cannot be sold across state lines, but can be legally sold in many states)? Should we allow smaller companies to comply with fewer food safety regulations? Where do we draw the line?

One challenging part about all of this is that science can only partially help.
Science can help us estimate the risk associated with different foods and production processes as well as the risk associated with different individuals at different stages of their lives. A 21-year-old, for example, has a much lower risk of acquiring foodborne illness than an 80-year-old person who is immunocompromised. But science cannot tell us what level of risk is “appropriate” – that’s a personal or societal decision.

Guidelines for identifying societally acceptable risk levels exist in the form of the US Healthy People 2030 framework, which sets specific public health goals, such as reducing human salmonella infections by 25% (the goal for 2030, as compared to the 2020 baseline). But in past iterations of this framework, there were no apparent consequences if goals were not met. We need clear responsibilities and action plans to meet the public health goals outlined in the Healthy People 2030 framework. And we need elected officials as well as government agencies to communicate these goals, develop and implement strategies to achieve these goals, and assure that there are consequences if the goals are not met.

This means we also need consumers and regulators to acknowledge that zero risk isn’t possible; well intentioned efforts that imply it is possible may cause more harm than good. A one size fits-all approach that isn’t risk-based poses problems for many reasons. It may decrease affordability and availability of certain healthy foods. It also may direct resources towards efforts that have limited impact on improving food safety (e.g. more testing) versus efforts that are more effective at reducing food safety issues (e.g. prevention) – a concept known as “opportunity cost.” Requiring recalls of any food that tests positive for a pathogen, regardless of how large or small a risk they pose, could be counterproductive if you consider a chocolate bar contaminated with salmonella is substantially more likely to cause foodborne illness than an equally contaminated raw chicken that will ultimately be cooked, thereby substantially reducing the risk.

Different regulatory requirements for different circumstances are, after all, commonplace – the speed limit in front of a school differs from the speed limit on a highway. Similarly, it makes sense to regulate infant formula more stringently than food consumed by the general population.

So, no, we shouldn’t consider the status quo as acceptable and yes we need to do better with regard to improving food safety. But improving food safety requires nuanced and collaborative risk-based efforts that involve everyone who contributes to growing, producing, preparing, distributing or serving food. As frustrating as it may seem, what is needed is steady progress that utilizes a strong scientific foundation, involves all parties, aims for continuous improvement and takes a realistic approach that does not imply zero risk is possible. After all, we don’t drive 20 miles an hour on all roads. Importantly, simplistic approaches such as more testing – even if they sound good – may often not do the trick.