AVMA vs Raw Food

Updated 8/1/26

In 2012, the AVMA adopted a policy position against raw meat diets for pets. Originally, it cited six studies that did not actually support its conclusions. This article was first written to rebut those conclusions.

Update (2026): The AVMA has since revised its policy. It still discourages feeding raw or undercooked animal-source proteins, but it now recognizes scientifically validated pathogen-reduction methods other than cooking or pasteurization. This is a welcome improvement over the original policy.

This article has been substantially revised to reflect developments in the scientific literature and changes to the AVMA policy since its original publication.

The most consistent problem with AVMA’s policy is that it conflates detectable pathogens and fecal shedding with actual harm. The two are not the same. Dietary contamination and asymptomatic shedding in raw-fed pets are frequently detected. We know that the U.S. meat supply is heavily contaminated with a variety of bacteria and other pathogens. But documented clinical illness in  pets eating raw meat (or humans in the household) is uncommon.

The mere detection of a microorganism in food does not demonstrate that it causes disease. Throughout this discussion, it is important to distinguish contamination, exposure, colonization, and clinical illness.

AVMA’s current policy statement: 

Animal-sourced proteins of concern include beef, pork, poultry, fish, and other meat from domesticated or wild animals, as well as milk* and eggs. Several studies reported in peer-reviewed scientific journals and product recalls have demonstrated that raw or undercooked animal-sourced protein may be contaminated with a variety of pathogens, including Salmonella spp, Campylobacter spp, Clostridium spp, Escherichia coli, Listeria monocytogenes, highly pathogenic avian influenza (HPAI) virus, Mycobacterium bovis, and enterotoxigenic Staphylococcus aureus. Dogs and cats may develop foodborne illness after being fed animal-sourced protein contaminated with the organisms listed above if adequate steps are not taken to reduce or eliminate pathogens. Apparently healthy dogs and cats can have subclinical infections with these organisms and pose a risk to other animals and people, especially young, elderly, pregnant, and immunocompromised individuals. People handling contaminated raw pet foods are also at risk of becoming sick.

My Response

The U.S. meat supply is contaminated with many bacteria. Uncooked meat does carry a real risk of bacterial contamination.[i] We are not arguing that point; in fact, we gladly stipulate to that fact. The following discussion focuses primarily on raw meat-based diets. Raw milk, eggs, fish, and wild-caught animal products carry additional source-specific hazards that are not fully reviewed here.

But which, if any, of these organisms presents a real risk to pets, people, or both?

Let’s take the bad guys named by AVMA, one by one:

Salmonella. The first study cited by the AVMA states that “Salmonella was isolated from 80% of the BARF diet samples (P < 0.001) and from 30% of the stool samples from dogs fed the diet (P = 0.105). Dogs fed raw chicken may therefore be a source of environmental contamination.”[ii] However, this is not the whole story. This study looked at only 20 dogs (10 raw-fed, 10 control), and of the three positive stool samples, only 1 was from a dog whose food contained the same serovar; another was from a dog whose food contained a different serovar, and the last was from a dog whose food tested negative. No dog or human illness was reported, despite confirmed contamination of the food.

The second AVMA-cited study states that 95% of human Salmonella cases are due to eating contaminated food. The authors speculate that some number of these could be due to contact with dogs or dog food, and cites data from 1987, which estimated that 1% of human salmonellosis could be due to contact with pets (but not pet food). The authors conclude, “To date, raw pet foods have not been associated with salmonellosis in humans.”[iii]

Research suggests that up to 36% of dogs and 18% of cats are asymptomatic carriers of Salmonella.[v] The possible source of the organism was not discussed, but may include the environment as well as diet, which of course was most likely to be heat-processed pet food.

Dogs and cats themselves appear to be quite resistant to illness due to this ubiquitous bacteria. In one case of illness in cats possibly linked to Salmonella, one cat was positive for a different serotype than the one in the food, and Salmonella was not proven to be the cause of either cat’s illness. Owner neglect was clearly the primary factor in their deaths.[vi]

In another case, two cats in a Dutch household developed gastrointestinal illness. One cat was ill for 10 days and lost almost three pounds before veterinary care was sought; the second cat developed similar signs several days later and was treated sooner. Both cats were diagnosed with salmonellosis. A commercial frozen poultry raw diet fed in the household contained two Salmonella serotypes. One cat was infected with one serotype and the other cat with the other. The authors appropriately described these as highly suspected, rather than proven, foodborne cases. [Giacometti F, Magarotto J, Serraino A, et al. Highly suspected cases of salmonellosis in two cats fed with a commercial raw meat-based diet: health risks to animals and zoonotic implications. BMC Veterinary Research. 2017;13:224. doi:10.1186/s12917-017-1143-z. PMID: 28738871.]

Finding Salmonella in food is not the same as demonstrating disease. Most studies of commercial raw diets measure contamination or fecal shedding rather than clinical illness. Those are important observations, but they answer different questions.

As of 2026, this has remained the general pattern: Salmonella is periodically isolated from commercial and homemade raw diets, yet reports of clinical illness in pets or people associated with those diets remain uncommon.

Campylobacter. AVMA’s reference on Campylobacter examines the public health risks of feeding raw diets to dogs.[vii] As evidence, this review cites a Hungarian study that included free-roaming dogs with access to aborted livestock fetuses and raw poultry that were found to be contaminated with Campylobacter. One person became sick, but it is unknown whether or not he handled infected tissues.

Campylobacter is a common pathogen of wildlife, livestock and poultry, is also zoonotic. However, research suggests that only a “small proportion of human infections are acquired from dogs…usually [from] puppies that are themselves suffering from diarrhoea. Only four infections associated with cats (all kittens with diarrhoea) have been reported.” The study concluded that dogs and cats are unlikely to be the source of human infections.[viii] Most human campylobacteriosis is associated with poultry and other food-animal reservoirs, with environmental exposures (including contaminated water) also contributing. [Cody AJ, Maiden MCJ, Strachan NJC, et al. A systematic review of source attribution of human campylobacteriosis using multilocus sequence typing. Eurosurveillance. 2019;24(43):1800696. doi:10.2807/1560-7917.ES.2019.24.43.1800696. PMID: 31662159.]

Importantly, one study on Campylobacter states that “…if a person and a pet have concurrent campylobacteriosis, the veterinarian must consider whether they both obtained it from some other common source (for example, food or water) or if the pet obtained it from the human being.” The authors conclude that dogs and cats are unlikely to be the source of human infection.[x]

Clostridium perfringens. Among the Clostridium species, commonly found in decaying vegetation, are common and contagious causes of gastroenteritis in dogs and cats. Infections are usually self-limiting. The study previously cited by AVMA to support its claim that Clostridia in meat is a significant health hazard states, “the exact relationship between C perfringens infection and gastroenteritis in dogs is not clear.” It also states that if anyone handled that meat–and did not wash their hands–they probably would have become infected. Another study reported, “No associations between C. difficile, MRSA or VRE and consumption of raw meat were detected.”[xiv]

Detection of Clostridium species or their spores does not by itself establish that a food caused disease. That depends on the species, toxin production, bacterial load, clinical findings, and other possible sources.

E. coli (Escherichia coli). The U.S. meat supply, particularly beef, is highly contaminated with E. coli. The organism is also ubiquitous in the environment.

Most strains of E. coli are harmless, normal inhabitants of the intestinal tract of humans and animals. Their presence in raw meat primarily indicates fecal contamination during slaughter or processing. But there are a few pathogenic types, particularly Shiga toxin-producing E. coli (STEC), which can cause severe disease in people.

European studies have occasionally found STEC O157 in commercial raw pet foods, whereas two large U.S. surveys (616 samples combined) found no O157 isolates. A Canadian study that looked for a pathogenic strain, E. coli O157, in raw-meat fed dogs and their environment found none. [xv]

One UK outbreak investigation found that three of four people infected with an outbreak strain of STEC O157 had exposure to dogs fed raw tripe. Although STEC was isolated from the tripe sampled during the investigation, it was not the same strain responsible for the human illnesses. The authors concluded that the raw tripe was a potential source of infection, but did not establish direct transmission from the food or from the dogs. Because green tripe is stomach tissue containing gastrointestinal contents, it presents a substantially greater bacterial contamination risk than skeletal muscle meat.

Listeria monocytogenes. Listeriosis has occasionally been reported in dogs and cats, but confirmed cases associated with raw pet diets remain rare. In people, Listeria monocytogenes is most commonly associated with contaminated ready-to-eat foods such as deli meats, hot dogs, soft cheeses, and similar refrigerated products, where the organism can survive and multiply during storage. Commercial raw pet foods have occasionally been found to contain Listeria and have been recalled for that reason. However, documented clinical illness in pets attributable to those products has remained uncommon.

HPAI (H5N1 Avian Infuenza or “Bird Flu”)

In the last two years, a major new risk, especially for cats, has arisen: H5N1 “Bird Flu.” This is a special case because cats are extremely susceptible to this virus, and cats have contracted it from raw meat and milk, both from homemade diets and commercial raw cat foods. (They can also get it from predation on infected birds and rodents.) Poultry is the riskiest raw product. Please read this article to understand the serious risks and how to mitigate them.

Staphylococcus aureus and Bacillus cereus. The original AVMA policy cited a study on Staph aureus that states, “Dogs are also susceptible to illness caused by Staphylococcus aureus and Bacillus cereus.” However, AVMA has chosen to focus solely on S. aureus while ignoring B. cereus. Its reasoning may become clear in a moment.

Both of these toxin-producing bacteria may be present in foods, but illness results from ingestion of toxin produced when these organisms multiply in food. Improper storage and prolonged holding at room temperature increase that risk. Any moist pet food—raw, canned, or dry food moistened with water or broth—can support bacterial growth if improperly handled. Safe storage and prompt refrigeration of uneaten food are far more important than whether the food is raw or cooked.

B. cereus, on the other hand, may be a greater risk with dry kibble than for raw food. Unpublished observations by UC Davis Professor Emeritus James Cullor (personal communication) found Bacillus cereus on the surface of every one of more than 50 commercial dry dog foods tested. He attributed many cases of “garbage gut” to B. cereus rather than Staphylococcus aureus. [xvii]

Many idiopathic digestive upsets may be related to bacterial growth from both of these organisms in moist foods left out for some time before they are eaten.[xvi] Many people mix water, broth, or other liquids with dry foods–this is clearly risky and should be avoided.

Mycobacterium bovis (bovine tuberculosis)

This is a peculiar addition to AVMA’s list of pathogens. The practical risk of contamination with Mycobacteria in commercially inspected U.S. beef is extremely low thanks to longstanding eradication and meat inspection programs. This organism remains an important concern in other countries, as evidenced by an outbreak in the U.K. among 50 cats in 2018. In the United States, risk from commercially inspected domestic beef is very low. Greater concern applies to raw milk, imported products from endemic regions, and raw meat from susceptible wildlife—particularly venison harvested in areas where bovine tuberculosis persists.

Toxoplasma gondii. Strangely, AVMA removed Toxoplasma gondii from its previous list, although somewhat contrarily, I still consider it a (minor) risk. So let’s take a quick look for the sake of completeness. Regarding this protozoal parasite, which may encyst in raw meat, researchers concluded that, “Because of their fastidious nature, the passing of non-infective oocysts, and the short duration of oocyst shedding, direct contact with cats is not thought to be a primary risk for human infection.”[xi] Given that about 41% of U.S. cats are allowed outdoors where they can readily hunt and consume prey, it is unlikely that raw food diets significantly increase the risk.[xii] The primary cause of human toxoplasmosis comes from humans themselves consuming raw or undercooked meat, milk, or shellfish.[xiii] Having 3 or more kittens was also a risk factor, but a sick litter of kittens also implies an outdoor cat–and irresponsible owners.

In Canada and some other countries, Toxoplasma tissue cysts are found relatively commonly in pork. Freezing and sourcing from controlled production systems markedly reduce the risk. However, outdoor and backyard pigs may carry a higher risk. Many commercial raw manufacturers freeze ingredients before processing. For homemade diets, all meat should be pre-frozen for at least 72 hours to kill Toxoplasma and other tissue cysts.

Cats become infected primarily by eating infected prey or raw meat containing tissue cysts. Humans, however, most commonly acquire infection from eating undercooked meat or from environmental exposure to sporulated oocysts (for example, while gardening without gloves), rather than through direct contact with pet cats.

Reducing and Eliminating Pathogens

AVMA does have a good point about the need to reduce the risk of illness from foodborne pathogens.  The heat processing used in commercial pet food production is intended to kill all the organisms specified by AVMA’s policy statement.

However, every method of feeding carries risks. The relevant question is not whether a diet is risk-free, but what the risks are, how common they are, and whether they can be mitigated.

The fact is that heat-processed pet foods have accounted for many more recalls than raw foods, despite the FDA’s heightened scrutiny of raw diets.

Dry food is heat-processed twice: first, the meat ingredients are rendered (a slow-cooking process that boils animal parts for hours at 250^oF), and then again during extrusion (a quick process, but done at extremely high pressure and temperature). Canned foods are retorted at high heat for up to two hours.

Heat processing carries its own risks. It denatures (distorts) proteins, which could become allergenic. High and/or prolonged heat also creates AGEs (advanced glycation end products) that have been associated with oxidative stress and inflammation, potentially leading or contributing to many degenerative diseases.

A long record of problems with processed pet food provides more than enough proof. Recalls have occurred due to:

• Formulation errors
  • Thiamine deficiency
  • Excess vitamin D
  • Copper excess
• Manufacturing contamination
  • Melamine
  • Aflatoxin (mold)
  • Salmonella
  • Listeria
  • Foreign materials (plastic, metal, chemicals)

Now, let’s look at the positive side of raw meat diets. We won’t belabor the point with the thousands of stories told by veterinarians and pet owners alike of near-miraculous cures when a pet’s diet was changed from heat-processed to raw.

Since this article was first written, research has demonstrated measurable physiologic differences between animals fed raw and conventional diets, including changes in digestibility, the intestinal microbiome, and metabolic by-products. Whether these differences translate into improved long-term health remains an active area of investigation.

There are multiple studies showing that raw meat diets for dogs and cats are more digestible than heat-processed foods. Here are a few of them:

• Beloshapka AN, Duclos LM, Vester Boler BM, et al. Effects of inulin or yeast cell-wall extract on nutrient digestibility, fecal fermentative endproduct concentrations, and blood metabolite concentrations in adult dogs fed raw meat-based diets. Am J Vet Res. 2012 Jul;73(7):1016-23.
• Kerr KR, Vester Boler BM, Morris CL, et al. Apparent total tract energy and macronutrient digestibility and fecal fermentative end-product concentrations of domestic cats fed extruded, raw beef-based, and cooked beef-based diets. J Anim Sci. 2012 Feb;90(2):515-22.
• Murray SM, Patil AR, Fahey GC, et al. Raw and rendered animal by-products as ingredients in dog diets. J Anim Sci. 1997;75:2497-2505.
• Vester BM, Burke SL, Liu KJ, Dikeman CL, et al. Influence of feeding raw or extruded feline diets on nutrient digestibility and nitrogen metabolism of African wildcats (Felis lybica). Zoo Biol. 2010 Nov-Dec;29(6):676-86.

What are the takeaways from all this?

For veterinarians, good client education and an emphasis on safe meat-handling procedures is adequate to prevent human illness related to raw (or any other) pet diets.

Hundreds of veterinarians recommend raw meat diets, and dozens of companies make them, but nobody recommends abandoning all reasonable safe meat-handling principles when feeding them.

It is absolutely true that raw meat has a high likelihood of bacterial contamination. But simply having raw pet food in the kitchen carries no more risk than preparing meat loaf for your kids: safe, as long as one uses safe-handling and cleanliness precautions.

As one study on zoonotic diseases states, “The transmission of pet-borne zoonoses is complex and usually requires close contact between susceptible human beings and animals or their excretions. Such contact frequently involves lack of common sense and gross breach of sound hygienic practice.”[xx]

With specific reference to Campylobacter, but broadly applicable to all potential pet-food related zoonoses, another researcher stated, “The application of simple hygienic measures should prevent such infections.”[xxi]

For pet parents using homemade diets, sticking to a well-formulated, balanced, varied recipe is essential.

The biggest risk I see from raw diets comes from the homemade side. The vast majority of books, internet sites, magazine articles, and other sources provide all kinds of yummy recipes, but almost none of them are balanced for long-term feeding. Even when people start with a balanced recipe, there is a tendency toward “diet drift.” You run out of calcium powder or B vitamins and forget to buy more, or your cat doesn’t like liver so you skip it, or your cat prefers turkey so variety disappears from the diet. These gradual changes can lead to severe nutrient deficiencies and severe health problems over time. Failing to add calcium, for example, can result in your cat breaking a leg while walking across the kitchen.

Commercial raw diets also require scrutiny: a nutritional adequacy statement does not necessarily guarantee that a product meets the AAFCO nutrient profiles. States that buy commercial pet foods and test them against their own guaranteed analyses frequently find deficiencies. The standard six-month AAFCO feeding trial may fail to detect deficiencies or excesses that will manifest later. The current issue of copper excess is a good example of this problem. 

On a personal note: over more than three decades of feline practice, I observed that most cats showed noticeable improvements in coat quality, body condition, and overall vitality when even a small amount of fresh raw meat was added to a commercial diet. These observations cannot establish cause and effect, but they are consistent enough to have influenced my clinical recommendations. They deserve further study.

---

References

NOTE: The Roman numerals may look weird due to some bug in WordPress…they are not live links, but they are accurate, and they do correspond correctly to their placement in the article above. New references in this update were added inline with the text.