August 2008

I am off to China to the International Food Safety & Quality Conference and Expo in a few weeks.  I am proud to be the lead sponsor of the event and am glad that I was able to convince our former Governor, Gary Locke, to attend as well.  China will become over the coming decades an even greater trading partner in food.  Coming up with strategies to help create an environment and culture where safe food is important is important to me.

Food safety is a worldwide issue that can benefit greatly from collaboration, standardized approaches, and common solutions. In many countries, food safety awareness is at an all-time high. New and emerging threats to the food supply are constantly being discovered, and our food supply is becoming increasingly global. Achieving food safety success in this changing environment requires novel prevention strategies, greater harmonization and more collaboration at the international level than ever before. As a responsible partner in the international food safety community, the General Administration for Quality Supervision Inspection & Quarantine (AQSIQ) is once again hosting the China International Food Safety & Quality Conference + Expo in 2008.

The Oklahoma State Department of Health (OSDH) reports today that at least 176 persons have become ill as a result of the E. coli O111 outbreak in northeastern Oklahoma. Cases include 128 adults and 48 children. Federal and state health officials say E. coli O111 is a rare type not normally associated with an outbreak this large. OSDH disease investigators, along with staff from Tulsa Health Department and area local county health departments, have interviewed more than 450 persons in an effort to identify the source of the outbreak. Interviews continue this weekend. While the source has not yet been identified, health officials continue to focus on the Country Cottage restaurant in Locust Grove, OK, after interviews with cases indicated most had eaten there during the time period Aug. 15 through Aug. 23.

The restaurant is closed while the investigation continues. Not all persons who ate at the restaurant have become ill. No other restaurant or food service outlet in the area has been linked to the outbreak. OSDH laboratory analysis of water samples taken from a private well on the restaurant property is continuing, however, health officials believe it is unlikely that any well water contamination is the source of the outbreak.

Prior Outbreaks of E. coli O111:

– Community Outbreak of Hemolytic Uremic Syndrome Attributable to Escherichia coli O111:NM — South Australia, 1995

– Outbreak of diarrhoea due to Escherichia coli O111:B4 in schoolchildren and adults: association of Vi antigen-like reactivity

– Escherichia coli O111:H8 Outbreak Among Teenage Campers – Texas, 1999

– Outbreaks of food poisoning in adults due to Escherichia coli O111 and campylobacter associated with coach trips to northern France

Several people have commented that switching from grain to grass feeding could be one of the solutions to the problem with foodborne pathogens in cattle and other livestock. Quotes like these are becoming more common on the Internet and in recent media reports:

“Products from grass-fed animals are safer than food from conventionally-raised animals.”  Eatwild, 2008

“Research has shown that the strains of E. coli most devastating to humans are the product of feedlots, not cows. This is due to the animals being forced to eat an unnatural diet, and not their natural choice, grass.”  Grass-Fed Beef: Safer and Healthier, Animal Welfare Approved, June 15, 2008

If true, changing the cow’s diet would be such a simple and cheap management practice to implement. Have we found the Holy Grail for food safety? Below is some research I did on the topic.


• Identification of on-farm management practices that would reduce or eliminate foodborne pathogens in cattle and other livestock (including diet changes) is an active area of research, but many study results are inconclusive. E. coli O157:H7, Campylobacter, Salmonella, and other dangerous pathogens have been repeatedly isolated from both grass and grain fed livestock, and the studies show conflicting results regarding whether the levels of pathogens are higher, lower, or the same when animals are fed grass- or grain-based diets.

• There is no clear and consistent definition in the literature of “grass-fed,” but the majority of papers describe animals that are on pasture or confined, but receiving only hay-based diets. Last year, the USDA Agricultural Marketing Service issued a standard for grass (forage) fed marketing claims. More research on this topic is needed that compares rates of foodborne pathogens among grain and grass fed animals using a specific definition such as the USDA standard or other accepted definition.

• The original study by Diez-Gonzalez published in Science in 1998, and since cited numerous times in the literature and media, suggested that cattle could be fed hay for a brief period before slaughter to significantly reduce the risk of foodborne E. coli infection. They based this conclusion on a hypothesis that grain feeding increases acid resistance of E. coli in cattle. Although they showed increased acid resistance in E. coli from grain-fed cattle, but the sample size was small, and they used “generic” E. coli stains, not E. coli O157:H7.

• Studies by other researchers worldwide have since found little difference in acid resistant E. coli O157:H7 among grain- verses grass-fed cattle, and some even found more E. coli O157:H7 shed by grass-fed animals.

• It has been discovered that E. coli O157:H7 and Salmonella can rapidly switch from being “acid sensitive” to “acid resistant” within minutes after entering an acidic environment (such as the human stomach). Thus, even if the grass-fed/E. coli acid-resistance hypothesis were true, manipulating the diet may not have any effect since pathogens can adapt quickly to new environments like the human stomach.

• Outbreaks have traced back to grass-fed and pastured animals, as well as animals in feedlots. Notably, the E. coli O157:H7 spinach outbreak strain in 2006 was isolated from grass-fed cattle. Another outbreak of E. coli O157:H7 was linked recently to raw milk and colostrum from cattle raised organically on grass.

• In summary, the scientific evidence at this time does not support a broad conclusion that grass feeding significantly and consistently reduces the risk of E. coli O157:H7 or other dangerous foodborne pathogens entering the food chain. However, more research is needed into the influence of food animal diets. For example, preliminary experimental data shows a possible association between feeding dried distiller’s grains and shedding of E. coli O157:H7 in cattle feces.


A systematic approach is necessary to combat the emerging challenges in food safety such as the unexplained “uptick” of E. coli O157:H7 outbreaks and recalls linked to beef products. Interventions to protect the food supply should ideally occur across the continuum from “farm to fork.” The “Holy Grail” of pre-harvest (farm-level) food safety would be to find an effective, affordable, and practical means to prevent or reduce food animals from shedding foodborne pathogens in the first place so the dangerous bacteria never enter the human food chain. Since cattle or other livestock may be located near drinking water sources or vegetable crops, a farm-level intervention could also help to protect nearby water and crops from contamination by manure via runoff, transport by wildlife/insects, or other mechanisms.

Oliver et al (2008) published a comprehensive review of developments and future outlooks for pre-harvest food safety this month. Examples of potential farm-level management practices that have been studied for E. coli O157:H7 and other foodborne pathogens in livestock include:

• Antibiotics
• Bacteriophages (viruses of bacteria)
• Dietary changes
• Immunization
• Probiotics or prebiotics in animal rations
• Sanitation/hygiene (feed, water, environment)
• Wildlife and insect control

Unfortunately, the best approaches for on-farm control of foodborne pathogens in livestock remain elusive. No single management practice, or even a combination of methods, has proven to be very effective or reliable in preventing foodborne pathogen colonization in livestock. Clearly, sanitation including clean feed/water sources and insect control are important, but difficult to maintain in a farm environment. Livestock immunizations are not available for most foodborne pathogens with the exception of an E. coli O157:H7 vaccine under development (and some ask “who would pay for such a program?” since cattle do not become ill from E. coli O157). Use of antibiotics is problematic because it can lead to resistance.


Definition of “Grass-Fed”

The majority of cattle are fed grass or other forage at some time during their lives. For the purpose of marketing, the USDA Agricultural Marketing Service issued a voluntary standard for grass (forage) fed marketing claims last year that states: “grass fed standard states that grass and/or forage shall be the feed source consumed for the lifetime of the ruminant animal, with the exception of milk consumed prior to weaning. The diet shall be derived solely from forage and animals cannot be fed grain or grain by-products and must have continuous access to pasture during the growing season.”

Note that most papers in the literature do not specifically define grass-fed using this new standard or any other specific definition, but differentiate, in general, between animals on forage (grass) only verses diets containing grain.

The Study that Started the Controversy

The original study that launched the controversy over grain feeding was published in Science in 1998 by researchers from Cornell (Diez-Gonzalez et al). They described potential dietary effects on the acid resistance of E. coli in cattle fed grain- versus hay-based diets. This study has since been cited numerous times in the literature and media, but later studies have not been able to reproduce the findings. This may be due, in part, to several limitations in the original study design including: 1) small sample size and 2) “Generic” E. coli levels were measured, not E. coli O157:H7.

In 2006, Hancock and Besser wrote a summary of the evidence surrounding the hypothesis that feeding hay instead of grain would reduce the problem with E. coli O157:H7, purportedly because the stomachs of grain-fed cattle are more acidic. They concluded: “while one cannot rule out a role of cattle diet on affecting exposure and infectivity of E. coli O157:H7 to humans, the data available at present demonstrate that cattle on a wide variety of diets (including 100% forage diets) are regularly and similarly colonized with this pathogen.”

Another interesting study from a research group in The Netherlands discovered that E. coli O157:H7 and Salmonella can rapidly switch from being “acid sensitive” to “acid resistant” within minutes after entering an environment with reduced pH (such as the human stomach). Thus, even if the grass-fed hypothesis were true, manipulating the diet may not have any effect since E. coli O157:H7 can adapt quickly to new environments like the human stomach.

Recent Findings in the Literature

In searching through the literature since Hancock and Besser’s review, several new papers relevant to the discussion were found.

1. Nutritional aspects of grass-fed beef.

Leheska, J. M., L. D. Thompson, J. C. Howe, E. Hentges, J. Boyce, J. C. Brooks, B. Shriver, L. Hoover, and M. F. Miller. 2008. Effects of conventional and grass feeding systems on the nutrient composition of beef. J Anim Sci.

• This paper explores the question about whether there are differences in nutrient composition of grass-fed beef compared with conventional (grain)-fed beef. Researchers have previously found higher omega-3 fatty acids and CLA (conjugated linoleic acid) in forage-fed beef, and lower fat content overall. Some consumers prefer eating grass-fed meat because they believe it is “healthier,” and/or tastes better than conventional beef.

• The authors of this study enrolled only producers that were marketing grass-fed beef and confirmed that “100% of the diets were made up of native grasses, forages, or cut grasses or forages.”

• Fatty acid composition of grass-fed and conventional-fed beef was found to be different, but the authors conclude “the effects of the lipid differences between grass-fed and conventional raised beef, on human health, remains to be investigated.”

2. Papers continue to be published about possible effects of diet on E. coli O157:H7 prevalence and concentration.

For example, a research team from Kansas State University reported that feeding distillers grains, a co-product of ethanol production, to feedlot cattle may have a positive association with fecal shedding of E. coli O157. The mechanism is unknown, but they hypothesize that the grains change the ecology of the hindgut where E. coli O157 is most likely to colonize cattle. The authors report that larger studies are underway to investigate this possible link.


In summary, the scientific evidence at this time does not support a broad conclusion that grass feeding significantly reduces the risk of E. coli O157:H7 or other dangerous foodborne pathogens from entering the food chain. However, more research is needed to better understand the influence of diet, especially the use of different types of grains in animal feed.


Continue Reading Grass-Fed vs Grain-Fed Beef and the Holy Grail: A Literature Review

My wife and 16-year-old daughter spent the week at the Democratic National Convention.  Me, I took our other 13-year-old and 9-year-old daughters to Hawaii for a few days in the surf and sun.  Yes, we did go to a laua – picture below the main course.

OK, I get asked all the time what I eat and do not eat – yes, I ate it.  My kids, however, kept asking why they killed and cooked Wilbur?  So, a little bit of an experiment in food illness surveillance?  It has been 48 hours since I ate Wilbur, let’s see how it goes.  If the pig is contaminated with listeria, I have a long time to wait with the incubation period running up to a month. 

As the Canadian Public Health Agency says, 29 cases of listeriosis have been confirmed nationally, and another 31 suspected cases are being investigated.  Maple Leaf, the manufacturer of the contaminated product, has ordered the return of all products made at the plant from nursing homes, hospitals, restaurants and stores, in one of Canada’s biggest food recalls.  For more information on Listeria, see

And for more worries North of the border, Salmonella kills one, leaves 87 ill in Quebec
Cheese Recall.
  A salmonella outbreak in Quebec has left one person dead and 87 others sick, prompting the Ministry of Agriculture, Fisheries and Food to recall three cheeses manufactured by Fromages La Chaudiere Inc.  Blocks of hard cheese, as well as cheese curds labelled La Chaudiere, Polo and Tradition, manufactured between July 24 and Aug. 24, have been pulled off store shelves as they may be contaminated with salmonella.  The outbreak has centred in three regions of Quebec — Chaudiere Appalaches, Estrie and Mauricie Centre du Quebec — but the cheeses have a wide distribution throughout the province, Horacio Arruda, Quebec’s director of public health, said in a news conference yesterday in Montreal.  Over the past week, a total of six cheeses have been pulled from store shelves across the province. In addition to the three cheeses recalled on Thursday, three other cheeses were recalled earlier this week because they may contain listeria.  For more information on salmonella, see

On July 16th, 2008 the Connecticut Department of Agriculture began an investigation of a possible link between several reported illnesses and the consumption of Retail Raw Milk (unpasteurized milk).  Recently we concluded that investigation. The investigation was prompted when the Department was notified by Connecticut Department of Public Health (DPH) Epidemiologists of 2 reported illnesses in which both patients had consumed Retail Raw Milk from a dairy licensed to produce Retail Raw Milk and pasteurized milk and milk products. The patients were aged 2 and 7, one was on dialysis. After notifying the dairy of the investigation, the dairy voluntarily stopped sale of all milk. Soon after the initial 2 reported illnesses, DPH reported 2 additional cases linked to the dairy. By the time we concluded our investigation a total of 7 known individuals were sickened from consuming Retail Raw Milk and several were hospitalized. The Retail Raw Milk implicated in this incident was purchased from 2 separate national, natural food, chain store locations and directly from the farm. None of the reported illnesses were linked to pasteurized milk and milk products produced at this dairy.

The individuals sickened had acquired a condition known as Hemolytic Uremic Syndrome (HUS) and one case of Thrombotic Thrombocytopenic Purpura (TTP). HUS is a disorder that occurs when an infection in the digestive system produces toxic substances that destroy red blood cells. It often effects the kidneys. This disorder is most common in children. It often occurs after a gastrointestinal (enteric) infection, often caused by a type of Escherichia coli (E. coli) bacteria, O157:H7.  Unpasteurized (Raw) milk has been associated with several outbreaks of E. coli O157:H7 infections in the U.S. Other outbreaks of E. coli O157:H7 have been associated with undercooked or raw hamburger (ground beef), unpasteurized fruit juices, alfalfa sprouts, dry-cured salami, lettuce, game meats and from transmission from animals to humans from contact with infected animals. HUS also can be caused by other enteric infections, including Shigella and Salmonella, and some non-enteric infections. Patients with TTP have clinical and pathologic features similar to patients with HUS.

In addition to Department of Agriculture staff, the investigation involved the Connecticut Department of Public Health and local health departments. After extensive testing of milk, milk contact surfaces, water sources, the environment in and around the farm and processing plant and, analysis of feces from each milking aged animal, the department obtained a genetic fingerprint match between E. coli O157:H7 recovered from the feces of 1 cow and E. coli O157:H7 isolated from 3 patients.

Approximately 170 separate samples and specimens of milk, water, feces and swabs of milk contact surfaces were analyzed by the DPH Public Health Laboratory in a 3 week period. A review of scientific literature reveals that E. coli O157:H7 as well as other food borne pathogens most likely are introduced into milk by contamination from animals shedding the organism in their feces. Direct introduction of pathogens into the milk from the bloodstream is unlikely but can not be ruled out. The department has concluded that the most likely cause of this food borne illness outbreak was the consumption of Retail Raw Milk contaminated with E. coli O157:H7. While good sanitation and management practices can lower the incidence of pathogens in raw milk we believe and studies support the position that pasteurization is the only proven way to eliminate pathogens from raw milk.

The Connecticut Department of Public Health and the Food and Drug Administration, and other public health authorities such as the American Medical Association, the American Academy of Pediatrics, the National Conference on Interstate Milk Shipments, the National Association of State Departments of Agriculture, the Association of Food and Drug Officials, and National Association of State Public Health Veterinarians all oppose the consumption of unpasteurized milk because of the health risks.

The food borne illness outbreak in northeastern Oklahoma that has sickened more than 115, hospitalized 50 and taken one life is the latest emergence of the virulent and highly toxic E. coli bacterium. Most E. coli outbreaks in North America are subtypes of E. coli O157:H7, but the CDC has just revealed that this outbreak is a rare serotype: E. coli O111.

“This is highly unusual,” said food borne illness attorney William Marler. “We have been involved in every major US outbreak in the last 15 years, and we have only seen this serotype twice before—once traced to apple cider in New York, and once connected to water or salad in Texas.”

Although many strains of E. coli can be present in the body with no ill effects, strains like E. coli O111 and E. coli O157:H7 produce a deadly shiga toxin (stx) which ravages the digestive system and kidneys. By the time symptoms emerge—abdominal cramping, vomiting, and bloody diarrhea—the bacteria is already entrenched. Although there is no cure or antidote, immediate health care is critical to support the systems under attack, keep the patient hydrated, and try to alleviate the intense pain that accompanies the illness as the body works to rid itself of the toxic bacteria.

In those with compromised or immature immune systems, E. coli can progress to Hemolytic Uremic Syndrome, or HUS. Children, whose immune systems are not as developed as adults’, are especially vulnerable. HUS is a cascading complication resulting in kidney failure; at the moment several children in Oklahoma are on dialysis. Even when they are able to recover from the potent E. coli toxin (considered by the CDC to be one of the most toxic substances known to man), victims often have permanent kidney damage. It is not unusual for E coli victims infected as children to need multiple kidney transplants over their lifetime.

“Regardless of the strain of toxic E. coli, it produces a devastating illness.” continued Marler. “Under the best circumstances, it can take months to recover. Some victims are affected for the rest of their lives. We need to support the families going through this nightmare, and do everything we can to help them.”

According to Kim Archer of the Tulsa World, Oklahoma state health officials have determined that a relatively rare and virulent form of E. coli infected dozens of patrons of Country Cottage, killing one and sickening more than 73 people.  More than 50 of those who fell ill were hospitalized.  Five children remain in the pediatric intensive care unit at Children’s Hospital at St. Francis.  Four are on dialysis.  Two other children were sent to OU Children’s Hospital.  Ms. Archer quoted State epidemiologist Dr. Kristy Bradley as saying the E. coli strain is not the commonly known E. coli O157:H7.  Non-O157 strains are more common in South America and parts of Europe, according to the Pediatric Infectious Disease Journal.

I could not agree more. Shiga toxin is one of the most potent toxins known to man, so much so that the Centers for Disease Control and Prevention (CDC) lists it as a potential bioterrorist agent (CDC, n.d.). It seems likely that DNA from Shiga toxin-producing Shigella bacteria was transferred by a bacteriophage (a virus that infects bacteria) to otherwise harmless E. coli bacteria, thereby providing them with the genetic material to produce Shiga toxin.

Although E. coli O157:H7 is responsible for the majority of human illnesses attributed to E. coli, there are additional Stx-producing E. coli (e.g., E. coli O121:H19) that can also cause hemorrhagic colitis and post-diarrheal hemolytic uremic syndrome (D+HUS). HUS is a syndrome that is defined by the trilogy of hemolytic anemia (destruction of red blood cells), thrombocytopenia (low platelet count), and acute kidney failure.

Stx-producing E. coli organisms have several characteristics that make them so dangerous. They are hardy organisms that can survive several weeks on surfaces such as counter tops, and up to a year in some materials like compost. They have a very low infectious dose meaning that only a relatively small number of bacteria (< 50) are needed “to set-up housekeeping” in a victim’s intestinal tract and cause infection.

I represented three Hemolytic Uremic Syndrome victims in an outbreak of E. coli O21:H19 linked to a Utah Wendys in 2006. Non-O157:H7 was also likely responsible for some of the illnesses in the Dole Spinach outbreak of 2006.

I also recently gave a speech in front of the USDA/FSIS on making all Shiga-toxin producing E. coli’s adulterants under the US Meat Inspection Act. Here is that text:

Another issue facing, not only the meat industry, but all of us, is the extent to which non-O157 E. coli may be present in food products – FSIS regulated or not. It is clear that Non-O157 Shiga toxin producing E. coli have emerged as a public health issue. Some non-O157 possess the same range of virulence factors as E. coli O157:H7 and are capable of causing serious illnesses, or death. Numerous serotypes, including O26, O103, O111 and O145 have been identified as agents of food borne disease.

I have seen their nasty work in the Dole spinach outbreak and an outbreak in Utah involving Wendy’s. Since 1990, 13 outbreaks of non-O157 E. coli have been reported in the US. While E. coli O157 is the principal strain isolated from implicated food and clinical isolates in the US, non-O157 predominate in other countries, including several of our beef trading partners like Australia, Brazil and Canada.

I will leave this to scientists and public health officials to sort all out. However, perhaps one needs to look no further than the FEDERAL MEAT INSPECTION Act and look at the term ”adulterated" for and answer. A product is adulterated: (1) if it bears or contains any poisonous or deleterious substance which may render it injurious to health." If non-O157 E. coli fits the bill, then to me that answers the question. However, what do we then do about salmonella, listeria, campylobacter, and shigella – especially those with particular virulence or antibiotic resistance?

One thing to remember, whether a product is considered to be an adulterant under the FMI or not, if a food product contains a bacteria or virus that sickens or kills, civil liability can, and often will attach. My vote is to simply get pathogens out of your product.

People in Oklahoma are understandably concerned about what is causing their friends and neighbors to become sick – some severely so. Unfortunately, pin-pointing an E. coli-contaminated food item in a buffet is difficult. Most people at buffets eat many items, and the same items, making in difficult to determine which item independently is the vector for the illnesses.

Cross-contamination is always an issue, but clearly, some food item (my suspicion – steak or hamburger) allowed the deadly E. coli to hitch a ride into the restaurant. It is not the first time that a buffet was the source of E. coli–related illnesses. In most, the vector is found, however, sometimes the best that can be determined is that it occurred at the restaurant. Here are a few that we have been involved with over the years.

Captain’s Galley Seafood Restaurant E. coli Outbreak – North Carolina

China Buffet E. coli Outbreak – Minnesota

Finley Elementary School E. coli Outbreak – Washington

Golden Corral E. coli Outbreak – Nebraska

Habaneros E. coli Outbreak – Missouri

King Garden Restaurant E. coli Outbreak – Ohio

Olive Garden E. coli Outbreak – Oregon

Sizzler E. coli Outbreak – Wisconsin

For more on Food Litigation generally and E. coli Litigation in particular, follow the links.  For more information on E. coli O157:H7, follow the link here.  And, for information on Hemolytic Uremic Syndrome, follow the link here.

Outbreak of Severe Diarrheal Illness in Northeastern Oklahoma

The Oklahoma State Department of Health (OSDH) has narrowed the focus of its investigation into a severe diarrheal outbreak in northeastern Oklahoma to the Country Cottage restaurant in Locust Grove, OK. The OSDH is trying to determine how food served at the restaurant might have become contaminated. The restaurant continues to remain closed while the outbreak investigation continues.

Yesterday the OSDH confirmed that an unusual type of E. coli is probably responsible for the severe illness experienced by at least 73 persons sickened by the outbreak, although disease investigators say that number could increase as they interview others who may have become ill as a result of the outbreak. The OSDH public health laboratory is collecting and analyzing specimens from patients and has sent some specimens to the federal Centers for Disease Control and Prevention for further analysis.

At least 50 persons have been hospitalized and one person has died.

Health officials warn that the strain of E. coli associated with this outbreak can be easily spread among family household members or close contacts, providing the potential for additional persons to become ill.

“We strongly recommend to the public that they wash their hands frequently, particularly after using the bathroom, changing diapers, and preparing food. Persons who are sick with diarrhea should not be involved in food preparation,” said State Epidemiologist Dr. Kristy Bradley.

As the investigation continues in the search for the source of the outbreak, health officials warn that the upcoming holiday weekend as well as tailgate parties and other recreational activities provide the opportunity for foodborne disease transmission. “Take the time to clean hands and food preparation surfaces frequently, keep raw meats separate from ready-to-eat foods, cook foods to proper temperatures, and refrigerate leftover foods quickly,” Bradley advised.

It is good to see an Agency do something that makes sense.

The U.S. Department of Agriculture announced a proposed rule to amend the Federal meat inspection regulations to initiate a complete ban on the slaughter of cattle that become non-ambulatory after initial inspection by Food Safety and Inspection Service (FSIS) inspection program personnel. This proposed rule follows the May 20 announcement by Secretary of Agriculture Ed Schafer to remove the provision that states that FSIS inspection program will determine the disposition of cattle that become non-ambulatory disabled after they have passed ante-mortem, before slaughter, inspection on a case-by-case basis. Under the proposed rule, all cattle that are non-ambulatory disabled at any time prior to slaughter, including those that become non-ambulatory disabled after passing ante-mortem inspection, will be condemned and properly disposed of. 

Good job.