Craig Hedberg, PhD, has dedicated his career to food safety. A professor of environmental health sciences at the University of Minnesota’s School of Public Health, he studies the surveillance of foodborne illness and the environmental factors associated with it, as well as the epidemiologic methods employed in outbreak investigations and disease control. Most recently, Hedberg contributed to a landmark study documenting the value of the PulseNet surveillance system. He earned his PhD in epidemiology at the University of Minnesota in 1993 and spent over a decade supervising the Foodborne, Vector-borne and Zoonotic Diseases Unit at the Minnesota Department of Health. Among his many awards is the 1999 FDA Group Recognition Award for outstanding efforts in the investigation of a multistate outbreak of Salmonella Agona in toasted oat cereal.

by Nancy Maddox, MPH, writer

Q: As one of the researchers who conducted the recent PulseNet economic study, were you surprised at the magnitude of your findings; that is, that PulseNet delivers a value worth 70 times the network’s costs?

A: I can’t say I’m entirely surprised, because most of what we do in public health is designed to prevent disease. The surveillance system for foodborne illnesses is part of that prevention system, and the way it functions is twofold: (1) more rapidly detecting outbreaks so we can intervene as they’re happening, which is a secondary prevention role, and (2) identifying the factors contributing to the occurrence of foodborne outbreaks in the first place. Identifying those factors enables us to work with the food industry to implement practices that reduce the overall risk of illness occurring, so that becomes a primary prevention method. If you look at our estimates for the savings generated by direct intervention and food recall versus savings associated with food industry changes, you will see that the latter generates richer rewards. Even though the focus of PulseNet is on secondary prevention, it creates tools for primary prevention, which is where the big gains come from.

Q: Where do you see PulseNet 10 years from now? 20 years from now?

A: What we’re going to see over the next ten years is that PFGE is going to be retired. We’re going to see something like PulseNet continue to serve as a tool to aggregate and disseminate information, but it will be a much richer stream of information based on whole genome sequencing (WGS).

Over the next 20 years as information technology improves, we will see the epidemiological piece—the patient exposure assessment— more readily and thoroughly integrated into the platform. Right now, PulseNet mostly deals with information related to the bacterial isolate, with limited information about the case linked to it. Critical information, such as where the patient went and what the patient ate, is generated by the epidemiologist who interviews the patient. I think we’ll see better opportunities to merge that detailed exposure assess- ment with the laboratory-generated information, which will increase the power of the system considerably.

Q: You spent a year as a visiting scientist at the Institut de Veille Sanitaire, essentially the French CDC. What differences, if any, did you find in the institute’s approach to foodborne disease surveillance and response compared with the US?

A: I think generally their approach is quite similar to the US system. One issue that limits their ability a little bit is that in France individual cases [of foodborne illness] are reported to a regional government agency without any patient identifying information. French culture places a priority on privacy, so the government doesn’t get that information until a reference laboratory confirms that the patient is part of an outbreak. In the US, the identity of the patient is generally part of the reportable disease process from the beginning.

Q: Any notable differences in governmental food safety oversight in France compared with the US?

A: One of the things that struck me as a little bit different is the relationship between regulatory agencies and food companies in France. In the US, regulatory agencies have specific rules and guidance on when they can take action against a company. In France, there seemed to be a lower threshold of engagement; if an investigation started to point to a particular company as the source of an outbreak, health officialswere much quicker to engage the company to figure out what might be wrong, rather than to build a regulatory case against it.

Q: Given your extensive food safety work, what do consider the biggest environmental and behavioral risk factors for food contamination?

A: One issue is simply the scale of the environment that food is produced in. Fresh produce is grown in open fields, and Nature is big and complex. It may come into contact with water, wild animals, domestic animals or other waste streams. Interaction among bacteria and plants and animals is always going to be a challenge. When produce is harvested, it is essentially ready to eat. Food is always going to have some potential risk for contamination from bacteria and parasites that can make us ill. Another issue is human nature. For example, getting people to follow recommended hand-washing practices is incredibly hard, but a critical issue. When things get busy in a commercial kitchen, people may not follow all the rules; they’re trying to get their main task done as quickly as possible. And there’s no immediate reinforcement for doing the right things and infrequent penalties for doing the wrong thing.

Q: With the increasing globalization of the food supply, do you foresee more opportunities for tainted foods to enter the US market from abroad? Or are the new FSMA regulations sufficient to address this?

A: I think the more we see food moving around the globe, the more opportunities there are for contaminated food to move around the globe. We can never create a regulatory structure that will ensure there is no contaminated food moving through the system. We can’t do in the United States, and we certainly can’t do it globally. We’re dealing with living [pathogenic] organisms that respond and adapt to changes in their environment. Our regulatory framework is designed to deal with the things we understand, but we don’t understand everything and the world around us is always changing. That’s why we need aggressive public health surveillance.

Q: If you could make three changes to the current US food safety system, what would they be?

A: (1) Better information-sharing among government agencies. FDA, for example, has restrictions on what it can share with state and local officials during the course of an investigation. We don’t always have the best coordination among agencies.

 (2) Greater commitment to aggressive epidemiologic investigations in all levels of government. One of the limiting factors is always getting patients interviewed. If you have a multi-billion dollar outbreak, you may be trying to coordinate interviews across multiple agencies with competing priorities. If we had some way to centralize interviews it would help us get them done rapidly and thoroughly.

(3) More effective use of traceback investigations. Typically, FDA and other regulatory agencies start the traceback process only after they have a good idea what the contaminated food is. But let’s say there’s a possibility the contaminated food could be lettuce or cilantro or tomatoes, and you don’t know which one. In that case, the traceback investigation may be needed to disentangle the co-mingled evidence
so we can use that information as part of our epidemiological analysis to identify the source of the outbreak. We have to make better use of traceback investigation than we are right now.

Q: How can food safety laboratories strengthen their collaborations with food epidemiologists? With academia?

A: Public health laboratories and epidemiologists need to work closely together. There should be regular information-sharing between them. They should feel comfortable talking with one another on a daily basis to build the relationships that allow that free flow of information to happen. Partnering with academia brings new eyes and new ideas into the mix, so we can do the kind of applied, systems-based public health research that will help us understand the gaps in our efforts and ways to improve our work. Unfortunately, there isn’t a lot of federal support for academia to work with state and local public health agencies. This is an area where additional investment could provide big pay offs. One model is the integrated food safety centers of excellence established under FSMA. Minnesota is part of that, but it’s just the tip of the iceberg.

Q: Are there any foods that are so risky (from a food safety standpoint) that you won’t eat them?

A: The only food I totally avoid solely for food safety reasons are sprouts. That’s because the seed is produced in open field conditions, and the growing conditions for the sprouts are identical to good incubating conditions for bacteria. I don’t drink raw milk, but I do eat some raw milk cheeses.

Q: Go ahead and brag: What are your favorite Minnesota eats?

A: Our seasonal abundance of sweet corn and tomatoes. And of course, when the state fair rolls around, anything on a stick.