Food (Safety) Fight Blog
By: Richard Raymond

Checking the math on foodborne illnesses

(The views and opinions expressed in this blog are strictly those of the author.)

I keep hearing elected and appointed officials say 76 million Americans were ill with a foodborne illness last year. That number is misleading.


The number comes from a CDC journal called "Emerging Infectious Diseases", published ten years ago in September, 1999. For that full report, go to: Food-Related Illness and Death in the United States. In this report, the authors used  FoodNet data from 1996-98 to reach the conclusion that 76 million Americans suffered a foodborne illness.

Last week the Centers for Disease Control (CDC) released its "Preliminary FoodNet Data on the Incidence of Infection with Pathogens Transmitted Commonly Through Food--10 States, 2008.”

In this report, comparing 2008 rates with rates from 1996-98, Yersinia had decreased 48%, Shigella had decreased 40%, Listeria had decreased 36%, Campylobacter was down 32%, and E coli 0157:H7 was down 25%. Salmonella did not change over that period, but had been down by 8% last year until the recent high number outbreaks.


 If the rates have dropped over the last 10 years, shouldn’t the number being thrown around also drop? 


In reading the 1999 report, I question the accuracy of the 76 million number because of two assumptions and one fact. The one fact is that the illnesses include Salmonella from petting zoos, E coli from livestock shows, Giardia from mountain streams and Hepatitis A from food service workers.


The first assumption is that for every reported case of Salmonella and Campylobacter, the two most common pathogens, 37 go unreported so the number is multiplied by a factor of 38, a number that is 10 times higher than used in the UK.  Personally, if I get bloody diarrhea, I am heading straight for my doctor's office.


The second assumption is that there are 130 million cases of Norwalk-like viral gastrointestinal illnesses per year, and that 40% of those are foodborne. The authors admit that is a speculative number, and that if they had chosen 30%, the number of foodborne illnesses would "only" be 63 million per year.


So if we only looked at foodborne illnesses that are the result of the meat, poultry and produce slaughter and/or processing practices, numbers which have fallen the last ten years but not reflected in the repeated use of the 76 million number, and do not count 52 million viral illnesses and other illnesses not related to industry, what is the "real" number of foodborne illnesses last year that the USDA and/or FDA could have an impact on? Now that would be an interesting number.


The FSIS is responsible for making certain that labels are "accurate and not misleading". Shouldn’t the same be demanded of the speech writers for the people who keep saying "76 million Americans suffered a foodborne illness last year" and our system "is a hazard to public health?"

4/14/2009 1:04 PM 


Main Page

setstats            Copyright (C) All rights reserved under

            If you have any comments, please send your email to