Will the "Spontaneously Hatched" Pre-Slaughter Meat Poultry Testing Guidance Actually Take Wing in the Dakotas?
The hasty APHIS 72-hour testing guidance, likely pressured (in hindsight) by yet another raw chicken cat mortality, now meets the actual Fall 2025 H5N1 outbreak - this will be interesting to watch...
Back on August 26th, APHIS, seemingly out of the blue, announced a revised policy for enhanced HPAI preslaughter surveillance, first reported via Feedstuffs, an industry news magazine: APHIS updates bird flu surveillance recommendations
Here is the link to the actual APHIS guidance:
This document describes additional premovement actions and testing that USDA APHIS recommends for poultry flocks prior to moving to slaughter and clarifies its indemnity policies that pertain to flocks in which HPAI has been detected that have been moved or partially moved to slaughter. This guidance is in addition to the laboratory obligations to perform avian influenza testing as per Title 9, Code of Federal Regulations, Part 145.15 (9 CFR 145.15).
The gist of the recommendations “recommend” that producers outside of HPAI control areas in “affected states” quarantine and collect HPAI targeted bird samples at 72 hours prior to shipment in turkey and broiler flocks scheduled for harvest, especially in flocks with increased mortalities above base-lines. Additionally, flocks shipped over multiple days are to be retested with negative results reported prior to subsequent shipments. The policy document details recommendations for sampling and clarifies indemnification policies related to flocks diagnosed as positive via the surveillance process.
At the time this was announced, I wondered what had prompted this policy change with no recent major news events or research findings publicized to prompt the announcement. In retrospect, the entire process was likely a relatively rapid, but carefully choreographed negotiated response between FDA, APHIS, and CA Department of Public Health to the most recent H5N1 in raw chicken cat food mortality episode in San Francisco CA.
Around September 1st, APHIS announced a confirmed positive cat from San Francisco on the HPAI Detections in Mammals web page:
Note that the “collection date”, likely from the owner’s veterinary clinic, was July 13th; however, NVSL did not classify the case as officially confirmed until August 29th. This was the first public indication of a feline H5N1 mortality in CA, coming after the new testing policies were announced.
The APHIS preharvest testing policy change and the listing of the positive San Francisco cat (7/13 collection and 8/29 “detection-confirmation” preceded an FDA announcement regarding contaminated cat food in California on September 3:
Following up on a case of H5N1 Highly Pathogenic Avian Influenza in a cat, testing performed by the U.S. Food and Drug Administration, state and local public health and agriculture partners, and federal partners suggests a link between the strain of H5N1 virus detected in the cat and in certain lots of RAWR Raw Cat Food Chicken Eats, a product the cat consumed before falling ill. FDA is sharing information about the testing for public awareness. The agency continues to investigate and will update this notice should new information become available.
Summary
FDA has found that certain lots of RAWR Raw Cat Food Chicken Eats sliders tested positive for H5N1. The affected lots are Lot CCS 25 077 (Sell By 09/18/26) and Lot CCS 25 093 (Sell By 10/03/26).
The San Francisco Department of Public Health (SFDPH) was notified a cat that ate product from Lot CCS 25 093 became ill with H5N1 and was euthanized. After initial polymerase chain reaction (PCR) testing of the open product sample from Lot CCS 25 093 collected from the pet owner by SFDPH detected H5N1, confirmatory PCR testing and subsequent whole genome sequencing (WGS) of a diagnostic sample from the cat and the open product sample from Lot CCS 25 093 were performed by USDA National Veterinary Services Laboratories (NVSL).
FDA collected and tested two retail samples of the same RAWR Chicken Eats product with a different lot number (CCS 25 077) and Sell By date (09/18/26). Both samples were positive for Influenza A Virus, and WGS was performed on one sample, which was also positive for H5N1.
FDA is concerned about the lots of RAWR Raw Cat Food Chicken Eats described above because whole genome sequencing suggests the H5N1 detected in the now-deceased cat and in Lots CCS 25 093 and CCS 25 077 of the Chicken Eats product originated from a common source of contamination.
WGS results also indicated that H5N1 from all three samples were within the same WGS cluster, indicating relatedness. The cluster involves a virus lineage that was detected from about November to December 2024 and is no longer circulating, supporting that the cat became ill from eating Lot CCS 25 093 of the Chicken Eats product.
NVSL testing of the cat, Lot CCS 25 093, and Lot CCS 25 077 identified the H5N1 as genotype B3.13. The B3.13 genotype virus has previously been found in other brands of raw poultry-based pet foods that were associated with the illness or death of cats.
FDA is not aware of any human cases of HPAI contracted through exposure to contaminated pet food.
This announcement then led reporters to the story, including interviews with the raw chicken company involved in the recall. Here is CIDRAP’s reporting on the incident:
FDA warns of H5N1 avian flu detection in raw cat food | CIDRAP
The US Food and Drug Administration (FDA) yesterday warned pet owners after tests found a link between H5N1 avian influenza samples from a sick cat in San Francisco County and a brand of raw cat food containing chicken.
Cat illnesses and deaths from H5N1 have been linked before to raw food and raw milk consumption, which prompted earlier recalls from different companies.
In a statement, the FDA said the San Francisco County Department of Health learned that a cat became ill with H5N1 after eating Rawr Raw Cat Food Chicken Eats and was euthanized. County officials tested a sample from an open container of the product collected by the cat owner, revealing H5N1.
The product is sold in 2.5-pound resealable plastic bags in retail and online stores nationwide. The two lots have sell-by dates of September 18, 2026, and the brand’s 40-count 1-ounce slider product expires October 3, 2026. The company has a production facility based in Grass Valley, Calif.
Samples from cat and food related, from B3.13 genotype
Whole-genome sequencing on samples from the cat and the food was performed by the US Department of Agriculture (USDA) National Veterinary Services Laboratory (NVSL). Earlier this week, the USDA posted a notification of a detection of H5N1 in a domestic cat from San Francico County with a collection date of July 13 and a detection date of August 29.
FDA testing on samples from the same company but a different lot number were positive for influenza A, and sequencing on one of the samples was positive for H5N1.
The FDA said it was concerned because testing of products from two different lots and the sample from the cat originated from a common contamination source and are related. Sequencing from all three samples were within the same cluster and involve a lineage detected in November and December 2024 that is no longer circulating.
NVSL testing of all three samples found that they belong to the B3.13 H5N1 genotype that circulated in dairy cows, then spread to commercial poultry farms across several states. The same genotype has been detected in raw cat food before and was implicated in cat illnesses and deaths.
So far, no human illnesses have been linked to exposure to H5N1-tainted cat food, the FDA said.
Company removed products but questions FDA communications
The company today posted a safety alert, saying it was deeply saddened by the illness and death of any cat, but claimed that the FDA waited weeks before providing information to help the company understand the investigations.
It said it uses only USDA-inspected human food-grade chicken and that the H5N1 finding should be addressed at the USDA level. Rawr said the sample from the sick cat’s home was tested 6 weeks after the product was opened, raising the possibility that it might have been contaminated in the home environment. It added that products from the other lot were produced from a different poultry source.
Out of an abundance of caution, the FDA said it removed products from both lots from the market weeks ago, well before the FDA’s announcement.
Finally, local news in California provided some additional context, with this piece from NBC television in the San Francisco area:
San Francisco cat sickened by bird flu linked to raw pet food – NBC Bay Area
…San Francisco's Department of Public Health told NBC Bay Area that on July 17 they learned about a "suspected feline bird flu case." The department said a San Francisco cat, after showing breathing and neurologic symptoms, tested positive for influenza A with a local veterinarian. After that, SFDPH said the specimen was then sent to the California Department of Public Health, which confirmed it tested positive for H5N1 bird flu.
The FDA said the cat was ultimately euthanized and that the cat ate RAWR Raw Cat Food Chicken Eats before getting sick…
Sabrina Simmons, the owner and founder of the RAWR raw cat food company, said in an interview with NBC Bay Area that RAWR was made aware that the cat had gotten sick at the end of July.
"Our understanding was that there was an ill cat, it had eaten our food, and there were just a few simple questions the FDA asked for," Simmons said. "We responded, and then it was quite some time before we heard follow-up."…
After reading the reporting with comments that the genotype was B3.13, matching dairy and poultry outbreaks from Oct-Dec. 2024, I searched the HPAI Poultry Confirmations Site for that time period in California, the most likely state source for the infected chicken product:
I counted 62 poultry H5N1 outbreaks in Oct-Nov-Dec., including 29 commercial chicken flocks (17 broiler flocks and 12 layers and pullet flocks - it’s possible that cull layers could be utilized for raw poultry pet food product). APHIS does not provide the genotype for outbreaks, but according to the California State Veterinarian in Congressional testimony as part of her plea for dairy herd vaccinations, almost all California poultry cases in late 2024 were B3.13 due to concurrent dairy herd outbreaks in the same areas.
Given the genotype signature, FSIS-passed infected poultry product was likely frozen and stored post-harvest somewhere in the pet food chain until purchase at retail and fed in early July to the cat who succumbed to the infection. It’s notable that the second lot of positive samples was separate and genotypically distinct, indicating likely slaughter of multiple infected flocks. I’m not aware if anyone has independently confirmed the phylogenetic information outside of NVSL and possibly FDA.
Regardless, given repeated findings of infectious H5N1 in raw poultry leading to feline mortalities, it’s likely that this latest incident or incidents discovered in mid to late July forced USDA to require some sort of action by APHIS to address the ongoing issue. Interestingly but not surprisingly, we saw the results of that pressure (the testing memo release) before we were made aware of the events that precipitated its hasty drafting…
The question now becomes how this new “suggested” policy plays out as outbreaks resume this fall. As of this weekend, South and North Dakota are both commercial poultry HPAI-affected states. I would assume that Georgia, California, and New York may not fall under the new guidance since their outbreaks were detected in non-commercial poultry operations:
Will all shipping turkey producers and the NAHLN labs in those 2 states begin testing at 72 hours pre-shipment and again prior to each subsequent multi-day load-out from each site? Will turkey processors expect producers to certify negative testing results under this guidance? Frankly, I’d be surprised if this new policy is enforced or even encouraged at this early date.
However, this policy brings several issues to the fore:
It is not in the CFR and therefore not mandatory: This is likely a major reason why it was written as suggested guidance. Short of an emergency rule by the Secretary due to dire threats to animal or human health, mandatory code changes cannot be promulgated or enforced.
Can NPIP infrastructure keep up and consistently meet a 72-hour turn-around? Existing testing requirements will likely be continued, despite irrelevant results, because they are codified. This memo will require additional testing, with 1 set of pooled PCR tests for each day of shipment per farm in “affected states”. I’m not well-versed on volumes of shipments by site in the turkey and broiler industries; however, someone will calculate the volume of official testing required in meat-poultry dense states to determine if this approach is even physically possible on a state-wide basis in broiler and turkey prominent states.
This guidance is written on an “affected state” basis. An “affected state” metric is not an adequately granular measure for increased risk of HPAI infection. What about bordering states in poultry dense areas (e.g. eastern IN and western OH)? Also, what about high- risk locations like meat poultry sites near H5N1-infected dairy farms, or farms near documented high levels of migrating bird movements?
Shipped asymptomatic infected birds from newly positive herds are not removed from the food supply. APHIS states this is not their issue; however, the testing will provide the processor (and FSIS) with potentially actionable information regarding likely H5N1-infected product already in distribution channels passed for human consumption. Who owns this headache?
72-hour sampling still provides a 3-day window for the virus to establish itself in flocks and replicate to high levels of viremia without clinical illness, especially in turkeys. 72-hour testing turn-around is a much-improved metric versus current H5N1 monitoring protocols, which basically depend upon increased mortality and FAD testing to find HPAI in near market-ready birds. (Current serology or even PCR sampling 18+ days pre-shipment provides essentially no assurance of acute HPAI status at slaughter.) While the proposal provides for improved monitoring, the window is still wide open for late infections less than 72 hours prior to shipping. These may be caught later on farms with multi-day shipping; however, see #4 for the drawbacks on the initially shipped lots of birds. And what do producers do when mortality spikes 12-18 hours prior to shipment after a negative test?
I’d suggest the entire meat protein processing industry is on thin ice, not because of the current iterations of H5 2.3.4.4b, but due to prospective future ones. Some sort of certification/testing scheme can likely be instituted for product entering raw pet food channels to greatly lessen risk for that niche market. The meat industry will not turn itself inside out just to dramatically cut risk of H5N1 in all meat for the sake of raw pet food!
However, while the current virus is zoonotic, no one seems concerned enough to truly treat it as high risk at this point, especially in the agricultural environment. The prevailing attitude seems to be that industry can slip the risk of this virus into the food chain by ignoring it with a “just cook it and be grateful to farmers” attitude. That will certainly work - until it doesn’t…
If slaughter animals (chickens, turkeys, swine, cattle, etc.) suddenly risk becoming preclinically infected by widely known and feared highly infectious influenza viruses with elevated risks for zoonotic spillover, then who will:
load and drive the trucks
inspect the animals pre and post slaughter
harvest and process the carcasses
restore lines of credit and liability insurance after they are pulled for “force majeure” clauses due to uncontrollable and forces of nature risks?
In today’s U.S. business climate, where government regulation and controls are in rapid retreat, I believe it is time to make the case for a One Health approach based on raw industry economic risk. It’s one thing to fear government over-regulation and interference as a risk to business continuity. It’s quite another to fear loss of business continuity from lack of credit and labor to just feed and harvest the animals due to panic from actual and perceived health risks, leading to catastrophic business shutdown. I think we all remember the human death losses to Nipah virus in pigs in SE Asia. It’s hard to assume what might happen if we lose a few farm or plant workers here.
Scenarios like these may well be beyond worst case. However, we are playing with fire by letting H5 continue to seep unmonitored into domestic livestock, peridomestic species, wildlife, and the human population through low level exposure by multiple routes, including sub-clinically infected slaughter animals.
Those who have regularly read my writings know that I would much prefer that industry take it upon themselves to monitor their on-farm populations and high-risk points of concentration through cost-effective (pooled) targeted point of care (POC) molecular testing. The old fear that POC testing will be used on farm to “screen out positives” is no longer valid for most of production agriculture. In many cases the farms and plants today are related or under the same ownership, making slipping infected animals into slaughter counterproductive.
Accredited labs serve vital functions in validating tests and confirming non-negative screening results. However, we are wasting tremendous resources, under-sampling populations, and losing timeliness by requiring that all testing must occur via official processes at NAHLN’s for HPAI. On-site screening for slaughter loads could be accomplished within 24 hours of shipping, with non-negatives held for confirmatory testing at NAHLN’s. Whatever we may lose in sensitivity and (perhaps) specificity, we will more than pick up with timelier and repeated testing opportunities. Many details remain to develop and negotiate, but the vision is too compelling to continue to reject.
H5N1 2.3.4.4b is now seasonal and endemic. We cannot continue to test for it as an official foreign animal disease and expect to keep up with it. This newest guidance is a valiant last-ditch effort, but it just doesn’t solve the shipping distance, turn-around time, and sample volume problems. It will be interesting to see how well this iteration of guidance wears as outbreaks pick up this fall.
John