Online publication only

Introduction
As a physician nutrition specialist, I find myself recommending food as part of my medical therapies. Therefore, as with any therapy, I began considering its potential side effects. My search highlighted the food safety concerns presented here. The knowledge may help reduce food-related illness and facilitate appropriate treatment.

Sentinel Surveillance
Sentinel surveillance works as in the "chicken salad at the church picnic" outbreak investigation, often used in introductory epidemiology classes: The sentinel case notifies a public health official of symptoms consistent with a foodborne illness. Through word of mouth, public health investigators, health-care resources, shopper loyalty cards, dining cards, or credit cards, additional cases are sought. (It's a good idea to use "plastic" to buy food; in case of an outbreak, you can be more easily reached.) Laboratory tests are conducted on as many cases as possible. Simultaneously, the potential source of exposure such as a restaurant or grocery store is investigated, including extensive food sample testing. More cases now involve interstate and even international commerce due to our complex food sourcing, so federal agencies can get involved very quickly as can the news media. However, the entire system is only as effective as the ability to identify the sentinel case.

Food Safety and Public Health Infrastructure
Case detection requires a prepared team of experts. Why have several recent outbreaks been detected in one state – Minnesota? Perhaps food safety favors a prepared health department. Could the steady trend of reducing health department resources make food look safer than it is?

Not All Acute Foodborne Illness Is Infectious
Most food safety graphs omit heavy-metal and related exposures. One of my longtime mentors at Johns Hopkins is Dr. Timothy Baker. After the infectious causes had been ruled out, he persevered in diagnosing an acute foodborne illness as cadmium poisoning from a refrigerator shelf used as an improvised barbeque grill (Baker 1961; Public Health Report). Gleaning from outstanding mentors like Tim, I suspected methemoglobinemia in a bottle-fed infant who presented to the emergency room with acute respiratory symptoms. The child's infant formula had been prepared from powder and boiled well-water. While boiling removed infectious agents, it also concentrated the nitrate that had entered the groundwater probably from a nearby chemical plant.

On Detecting Nonacute Foodborne Illness
Sentinel surveillance presupposes that a foodborne exposure makes people acutely sick. While that inherent limitation may seem obvious, I've heard many statements, and you may have, too, which overreach these tenets.

One fallacy is, food is much safer now from tapeworms than it used to be. The medical literature provides evidence for the opposite assertion. A ready example is neurocysticercosis, the larval state of the pork tapeworm in humans. The increasing rates may be the result of better case detection with imaging studies, but could also be a true increase as the US becomes more international both in its people and its food sources.

The general statement about fewer tapeworms in the US is probably referring to the adult stage of the tapeworm, which, in the case of the pork tapeworm, comes from ingesting the viable larvae found in undercooked "measly" (infected) pork. In such a case the ova and parasites (O&P) stool analysis could potentially be positive. But when is it detected? Exposure to measly pork doesn't result in clusters of acute diarrheal disease, which might prompt stool O&P testing. Even when this testing is conducted, detection rates are low when specimens are not examined promptly because the proglottids and eggs break down over time. DNA testing, if available, tends to be more specific than sensitive. The important, critical point sometimes overlooked in clinical practice is that neurocysticercosis cannot be detected by a stool O&P. Neurocysticercosis comes from food contaminated by the eggs usually from someone else involved in food handling or harvesting. The eggs do not appear in the stool but rather enter the body through the gastrointestinal tract and deposit as cysts (the larval stage) in muscle or brain.

Mold doesn't seem to make it onto the food safety charts at all. When mycotoxin-related foodborne illness is addressed, there seems to be a misunderstanding that moldy food will look moldy to visual inspection. For the most part, mycotoxins aren't about the glass of coconut water that someone drank before cracking the coconut open to notice that it was moldy, or the brown fuzzy raspberry at the bottom of the bowl. The primary concern centers on crops processed en masse, introducing hidden mycotoxins in foods such as peanut butter and corn chips. Mycotoxin-related illness has far-reaching long-term consequences that do not lend themselves to food safety surveillance. A case cluster of neural tube defects in the US was linked to maternal exposure to fumonisin-contaminated corn. Similarly, foodborne aflatoxins are associated with hepatic cancer.

Are Some People More Vulnerable Than the Sentinel Cases?
The word outbreak suggests that many people are affected. What if only one or two people become ill? My clinical impression is that not only is undetected illness common, but also some people have preexisting illness: they may have inadequate stomach acid to break down parasite eggs inadvertently ingested at a salad bar, methylation defects impair removal of foodborne mycotoxins, and a fatty liver is a risk factor for Vibrio infections.
 
Is Infection-Mediated Foodborne Illness at Odds with Nutrition?
Irradiating food exposes the food, workers, and environment to ionizing radiation but makes food safer from an infectious disease standpoint. Trans fats reduce the rates at which food becomes rancid, because the trans double-bond configuration reduces the fat's interaction with oxygen. Fruits and vegetables, including some with the organic label, can be mass-produced in a highly controlled environment where infection risk and nutrient levels are minimized.

My preference is for more dialogue about these decisions, which are often made without societal input. In most situations, options can be developed, usually with some modest investment, where animal welfare, the environment, food safety, and nutrition can all benefit.

Conclusion
Don't be incorrectly reassured about food safety. Current public health surveillance is helpful and needed, but it is limited, especially in its detection of food and water exposures to heavy metals, parasites, and mold.

Unfortunately, metals, parasites, and mycotoxins are also difficult to detect in clinical practice. This doesn't mean that they are not present. On the contrary, when patients don't get better with conventional treatment, astute clinicians consider the potential for metal toxicants, parasite-derived biotoxins, and mycotoxins. Treatment resources are available: several can probably be found in this edition of Townsend Letter, which focuses on inflammation, chronic infections, and Lyme disease.

Ingrid Kohlstadt, MD, MPH, has been elected a Fellow of the American College of Nutrition and the American College of Preventive Medicine, and is an associate at the Johns Hopkins Bloomberg School of Public Health. She recently completed a two-year appointment at the FDA working in the Office of the Commissioner, and published one of Medscape Public Health's "Top Ten CMEs for 2010." Dr. Kohlstadt is the founder and chief medical officer of INGRIDients Inc., editing Food and Nutrients in Disease Management (CRC Press; Jan 2009) and Scientific Evidence for Musculoskeletal, Bariatric and Sports Nutrition (CRC Press; 2006).