Header graphic for print

Food Safety News

Breaking news for everyone's consumption

Publisher’s Platform: How to Break a Foodborne Illness Story

Daily, weekly, monthly and yearly, people get sick from eating food. The Centers for Disease Control and Prevention (CDC) estimates that at least 48 million of us get sick each year, with 125,000 hospitalizations and 3,000 deaths.

Many of those sickened are the “canaries in the coal mine” – people who are the first to become ill in an outbreak.

Promptly, but accurately, announcing early illnesses and the cause of those illnesses can help prevent more illnesses from happening. Letting consumers know what products are poisoning them helps the free market work. Simply put, transparency allows consumers to make decisions about what products to avoid based on which manufacturers and products have a history of problems.

But, how, when, or if ever public health officials announce outbreaks and recalls remains a mystery to many reporters who cover food-related pathogen outbreaks.

After practicing in this area for more than 20 years – yes, it has been more than 20 years since the Jack in the Box E. coli O157:H7 outbreak – I think I have learned a few things about how the foodborne illness surveillance system operates and how to get information to consumers that allows them to protect themselves and their families and to make better choices about the food they choose to buy.

Bugs are now reportable.

In November 1992, there was an ominous uptick of E. coli O157:H7 illnesses in the San Diego area. Ultimately, by the end of December of that year, there would be at least 40 sickened and a child dead. However, since E. coli O157:H7 infections were not reportable at the time, no one knew the cause of the outbreak, and the same tainted meat served at San Diego-area Jack in the Box restaurants was shipped to other Jack in the Box restaurants in the Western U.S., eventually sickening several hundred and killing three more children.

That outbreak launched both a legal career and mandatory reporting of the most common foodborne pathogens: E. coli O157:H7, Salmonella, Campylobacter, Listeria and Shigella, to name a few.

Foodborne surveillance operates, but a bit slowly.

Think of the foodborne illness surveillance system as a large funnel. At the opening on top are all of us who eat food and get sick, and at the very bottom are those who get diagnosed, reported and linked to others so that the events become an outbreak.

One thing to remember is that, for every person who gets counted as part of an outbreak, sometimes dozens of others are actually sick, but remain undiagnosed and therefore uncounted. Here are some startling statistics: for every one E. coli O157:H7 case counted in an outbreak, 26.1 go undiagnosed; for Salmonella, that ratio is 1:29.3; for Campylobacter, it is 1:30.3; for Listeria, it is 1:2.1, and for Shigella, it is 1:33.3. That amounts to a lot of uncounted ill people.

If you are uncounted, no one asks you what you ate or drank or when. 

First, why do sick people remain uncounted? Well, in order to be counted, you need to be stool-culture positive for a reportable pathogen.

  • If you are not sick enough to go to the doctor, you do not get counted.
  • If the doctor does not order a stool culture, you do not get counted.
  • If you do not test positive for a reportable pathogen due to lack of testing, there is no reporting requirement.
  • If you are not counted, your local health department will never call you to ask what made you ill.

Without that intervention, whatever made you sick may well remain on the market to sicken others.

Hint No. 1:

Ask your local, state and national health authorities frequently about unusual upticks in reportable illnesses.

When a stool culture returns positive for a reportable illness, that lab is required to contact the local health department in the location where the ill person lives. The length of time it takes for a positive culture to be reported can vary widely, location to location, state to state, potentially slowing the process of local health authorities even beginning an investigation.

Often a state’s Department of Health will become involved. The stool culture isolate will be provided to the state’s lab, which may perform Pulsed Field Gel Electrophoresis (PFGE), or genetic fingerprinting. The genetic fingerprinting allows for comparison between individual isolates (that may also include food samples). PFGE, along with good epidemiology, may well begin the process of finding connections between people and the food they consumed.

Again, this takes time, and time that keeps potentially tainted products on store shelves or allows a bad manufacturing process is continuing to sicken customers.

Hint No. 2:

CDC is important, but most of the real investigation happens at the state and local level.

If a PFGE is performed at the state lab, the digital image (it looks like the bar code on the back of a cereal box) is uploaded to CDC’s PulseNet. PulseNet, along with CDC’s OutbreakNet, are potentially seeing these uploads from a variety of states. If the PFGE patterns are “indistinguishable,” and if patients can be linked in time and by a product, there is likely a multi-state outbreak occurring.

Remember, CDC is compiling the information from the states that are supplying the information. They are leaders among equals, but they are only as good as the information they receive. Again, each state has its own interests and time constraints. These either enhance or detract from the ability of a state public health agency or CDC to move more or less promptly on a potential foodborne illness outbreak.

Once local, state and national health authorities determine that an outbreak is happening, or has happened, what next – Part One?

Here is where transparency may get a bit tricky. My bias is full disclosure, but only after it is clear which product or manufacturer is the likely source of the contamination. In my view, once a public health official determines the likely cause of an outbreak, transparency demands that the public be notified.

At this point, there are no legitimate reasons for nondisclosure, only excuses. That the business may lose business is not a reason. That the product is no longer out in the market is not a reason. The public has a right to know – period.

Hint No. 3:

If a health official refuses to name “nationwide fast-food Mexican restaurant A,” ask them under what legal authority they are making that decision. If they continue to refuse, file a Freedom of Information Act request and do not back down.

Once local, state and national health authorities determine that an outbreak is or has happened, what next – Part Two?

Traceback – once a product has been identified as the source of illnesses, a potentially complex agency process begins. Side note: the U.S. Department of Agriculture’s Food Safety and Inspection Service (FSIS) is generally in charge of all meat – cows, pigs, chickens, turkeys and lamb. The U.S. Department of Health and Human Service’s Food and Drug Administration (FDA) deals with fish (except catfish; FSIS does that), dairy and all other food items. FDA now also does eggs. There is joint jurisdiction on meat pizzas with cheese.

Depending on the food involved, FSIS or FDA (sometimes both) will take the outbreak information from CDC and/or state health officials and attempt to trace the outbreak to the source or “root cause.” The idea is to both stop the outbreak and to learn the cause to prevent the next outbreak. Again, transparency becomes paramount. Of course, state Departments of Agriculture will be involved in partnership with FSIS and FDA as “boots on the ground.”

If the outbreak is local in nature – a local restaurant or church picnic, for example – CDC, FSIS or FDA may never be involved. Local health and environment health professionals will likely do all the real work.

Food that causes an outbreak seldom tests positive for the pathogen that caused the outbreak.

Simply put, the victims ate the evidence. That is why they became ill and why leftovers did not test positive for the same pathogen. Although there have been some successes – the spinach E. coli outbreak of 2006 and the peanut butter Salmonella outbreak of 2009 – most often there is no food left over to test.

Even if there is food from the same lot available to test, pathogens are not uniformly spread throughout a lot, or, by the time food is tested, other bacteria may have out-competed the pathogens.

Bottom line, even without a positive food sample, the good work of epidemiologists can link people locally, nationally and even internationally to a common pathogen and the food on which it hooked a ride.

Hint No. 4: 

Develop relationships with local, state and national public health and environmental officials. Do the same with people at FSIS, FDA, CDC and state Departments of Agriculture. Help them tell the story about how important what they do is to the health of the public.

Hint No. 5:

Develop relationships with reporters who know what they are doing. Talk to the good people at Food Safety News, Lynne Terry of The Oregonian, JoNel Aleccia at NBC and Elizabeth Weise of USA Today. Spend a few moments with the experts.

© Food Safety News
  • Marcos Avila

    Thank you Bill. This is a great post, that clarifies the logic and uses behind food health reporting. Thanks again!

  • Jim Mann

    Excellent review. Thanks.
    Do you have the ratio of counted to uncounted in norovirus?