OEM Design of Dog Food for Suspected Chicken Allergy: Meat, Poultry Meal, Chicken Fat, and Egg

“A lot of dogs are allergic to chicken, so please take out every chicken-related ingredient.” When we are approached about dog food formulated with allergies in mind, the conversation often opens with exactly that sentence.

The question that always follows is how far “every” goes. Chicken meat, poultry meal, chicken fat and egg are all chicken-related ingredients, but they differ in what they are as raw materials and in the risks that need to be checked. On top of that, there are routes that are easy to miss if you only read the main ingredients on a formulation sheet, such as digests, hydrolysates and palatant coatings.

To put the conclusion first: when you design a product around a suspected chicken allergy, the first thing you have to define is the range of chicken-derived ingredients you are excluding. “No chicken meat,” “chicken-free” and “free from chicken-derived ingredients” do not cover the same scope. Until the exclusion scope is fixed, you cannot settle the recipe, the ingredient specifications, the checklist for the factory, or the wording on the pack.

The hardest call of all is chicken fat. For the protein that remains in chicken fat, we could not identify any common standard — within published regulations or primary research — that would let you say “below this level it will not trigger an allergic reaction.” It is not that the measurement is impossible; it is that no common standard linking a measured result to clinical safety in dogs has been published.

For that reason you cannot label chicken fat as uniformly “safe” or “unsafe.” Equally, you cannot conclude that heating, refining or hydrolysis on their own remove the concern. What the product development side can do is set out the intended use and the label wording, then specify the exclusion scope, ingredient specifications, manufacturing process, testing conditions and record-keeping in concrete terms.

What this article covers

• Chicken-derived ingredients need to be considered in four categories: fresh meat, poultry meal, chicken fat and egg. There are also ingredients that require a separate check, such as digests and coating materials.
• “No chicken meat,” “chicken-free” and “free from chicken-derived ingredients” each cover a different range of excluded ingredients.
• For residual protein in chicken fat and for the degree of hydrolysis, there is no common standard that would let you judge clinical safety in dogs.
• “Not used,” “no cross-contamination” and “clinically hypoallergenic” each require a different kind of evidence.

Think of chicken-derived ingredients in four categories

Chicken-derived ingredients need to be considered in at least four categories — groupings by the part or fraction that is taken out. For as long as they are all lumped together as “chicken,” no design decision is possible.

The single word “chicken” houses ingredients with very different characters: meat, meal, fat and egg.

Chicken-derived ingredients | fresh meat, meal, chicken fat, egg

In this article we treat chicken-derived ingredients in four groups: (1) fresh meat (chicken meat, chicken liver and so on), (2) poultry meal (a dried protein ingredient that has been through rendering — heating, defatting and grinding), (3) chicken fat (also traded under names such as chicken oil, though the name alone does not guarantee an identical process or specification), and (4) egg. Groups (1), (2) and (4) are protein-led; only (3) is fat-led. That asymmetry matters later.

First Reach’s view

In our view, this four-way split is a practical framework this article has reorganised for OEM recipe design; it is not an industry-standard classification. It matches neither the regulatory ingredient categories nor the academic allergen classifications.

Alongside it, put a fifth check route into your working practice: digests, hydrolysates, gravies and palatant coatings. It is not a fifth biological category, but if you only look at the main ingredients on the formulation sheet you will miss chicken-derived material that enters at the coating stage.

How often is chicken allergy reported?

As a matter of record, in a systematic review that collected elimination-diet and re-exposure studies (Mueller et al. 2016, BMC Vet Res 12:9), chicken was reported as involved in 45 (15%) of 297 dogs with confirmed adverse food reactions — third after beef (102 dogs, 34%) and dairy (51 dogs, 17%).

This is a reported frequency with confirmed food-allergic dogs as the denominator, not a prevalence figure for dogs in general; and the subject was chicken meat, not refined chicken fat.

About the protein contained in chicken fat

The first thing to be clear about is that, in the dog pet food field, we found no common standard for the protein remaining in chicken fat — nothing that says “below this level you need not declare it” or “below this level it will not trigger an allergic reaction.”

That does not mean residual protein cannot be measured. Measurement is possible; what is missing is a unified basis on which to judge the number.

In the dog pet food field, there is at present neither a published rule setting how much protein may remain in chicken fat, nor published data tying such a figure to clinical safety.

There is no obligation to declare crude protein

As a matter of record, the EU and Canadian feed ingredient provisions do not uniformly require crude protein to be measured or declared for animal fats.

For “Animal fat” in the EU Catalogue of feed materials, the declarations required are crude fat, and moisture where the content exceeds 1%. There is no obligation to declare crude protein.

In the Canadian feed ingredients table as well, what must be declared for animal fats are maximum levels for moisture, insoluble impurities, unsaponifiable matter and free fatty acids. No crude protein guarantee is required. Note, however, that this ingredients table covers livestock feed and does not apply directly to dog and cat pet food.

In other words, protein ingredients such as meals are required to declare crude protein, while animal fats are not. That does not mean “fat contains no protein at all”; it means no common basis for measuring and declaring residual protein has been established.

There is no published data on residual protein

In this review we could not identify peer-reviewed papers or official analytical data measuring how much protein remains in refined chicken fat.

Nor could we identify studies feeding chicken fat alone to dogs to examine allergic reactions, or research directly testing whether chicken fat can be used in an elimination diet for food allergy. Ingredient specifications and general composition tables may carry an indicative crude protein figure, but we found no reliable published data that also states the measurement method and the limit of detection.

That said, this does not mean “protein remaining in fats and oils cannot be measured.” In human food research, trace cashew protein in heated vegetable oil has been reported as measurable by an analytical method called LC-MS/MS. In oil spiked with 100 ppm of cashew protein, over 90% was still detected and recovered after 30 minutes at 138°C, and over 50% at 166°C.

However, that study was an experiment using vegetable oil, not an investigation of animal fats such as chicken fat. It therefore cannot serve as direct evidence for how much protein remains in chicken fat or what the effect on dogs would be. It is simply a case showing that measuring trace protein in fats and oils at the ppm level is itself feasible.

Residual protein cannot be judged on “amount” alone

You cannot simply conclude that “chicken fat contains little protein, so the allergy concern is small too.” What matters is not only the total amount of protein but which kinds of protein remain.

In the study by Olivry & Bexley (2018), serum IgE from corn-reactive dogs reacted to corn flour in 20 of 30 samples (67%), while none of 40 samples reacted to cornstarch. The protein concentration of the extracts used in the test had been adjusted to the same 5 µg/mL in both cases.

The difference in reactivity therefore cannot be explained by a simple difference in protein quantity. The authors reason that the kinds and composition of the proteins present are likely to differ.

The study concludes that refined cornstarch is less allergenic than corn flour. But you cannot explain that solely as “refining reduced the amount of protein.” The kinds of protein that remain matter too.

For chicken as well, Olivry et al. (2022) identified seven major chicken allergens and one minor allergen recognised by canine serum IgE. One of them is serum albumin, which has the property of binding fatty acids.

However, that study was a blood-based test, not an analysis of chicken fat itself. It therefore cannot tell us whether these allergens remain in refined chicken fat, or to what extent if they do. We could not confirm this from published data either.

Note that the 2018 study was carried out by Avacta Animal Health with funding from Royal Canin. The authors’ relationships with both companies are disclosed, so this should be taken into account when assessing the findings.

Design decisions for each chicken-derived ingredient

The categories differ in regulatory treatment, in whether primary canine data exists, and in the design questions they raise. The dividing line for design quality is whether you can write “which category, and why we are excluding it” rather than just “remove chicken.”

Meat, meal, fat, egg and hydrolysis each require a separate judgement, because the strength of the evidence differs.

Category (fraction / processing)	Treatment in regulations and reference sources (jurisdictions checked)	Primary data in dogs	Design question
Fresh meat	Declared as an ingredient name	Seven major allergens plus one minor (serum IgE)	Findings on protein fractions cannot be extrapolated to fat
Poultry meal	Defined in AAFCO and the Canadian CFIT, but naming the species is optional	Non-hydrolysed meal is recognised by serum IgE	“Poultry” does not reveal whether it is single-species or mixed
Chicken fat	In the EU and Canadian provisions checked, no requirement to declare or guarantee crude protein	No published data identified	Define by specification in an area that has no common standard
Egg	Declared as an ingredient name	Only whole-egg extract-level reports identified	Treated as a separate item from chicken meat in several primary studies
Hydrolysis (processing)	In the EU provisions checked, no numerical molecular-weight requirement	Indicative molecular weights conflict between sources	No common standard; define by specification

(1) Fresh meat

This area has comparatively more research data. Care is needed, however, when switching to duck or turkey in order to avoid chicken.

In a study of 30 dogs with detectable chicken-specific IgE, all reacted to chicken meat, and 97% reacted to duck meat and to turkey meat respectively (Olivry et al., 2017). In other words, canine serum IgE recognised duck and turkey meat in much the same way as chicken meat.

This was a blood-based test result, however. No provocation test was carried out to see whether dogs actually develop symptoms when fed. Egg was also not examined in that study.

The finding suggests that simply switching chicken for duck or turkey may not differentiate a product sufficiently as a “novel protein.” If avoiding cross-reactivity with chicken is the priority, choosing a non-avian species is a more consistent approach to ingredient selection.

(2) Poultry meal

For this category, note that the ingredient name alone tells you little about what is inside.

Under the definitions on the AAFCO official site and in the Canadian CFIT, “Poultry Meal” and “Poultry By-Product Meal” are defined as separate ingredients. In both cases, however, stating the specific bird species — chicken, turkey and so on — is optional. So when a label says only “poultry,” you cannot determine whether a single bird species was used or several were mixed.

There is also data on protein after processing. In the serum IgE testing by Olivry et al. (2017), 73% of sera from chicken-sensitised dogs reacted to an extract of non-hydrolysed rendered chicken meal. Reactivity to chicken meat was 100%, and to the beef comparator 3%.

This result indicates that chicken-derived protein recognised by canine IgE may remain even in chicken meal that has undergone heat processing such as rendering. You therefore cannot say “it has been heated, so there is no allergy concern.”

What was used in the test, however, was the protein fraction of chicken meal after fat had been removed with acetone. It was not a study of protein remaining in refined chicken fat, so the result cannot be transferred as-is to chicken fat.

(3) Chicken fat

As explained in the previous section, there is no published common standard for the protein remaining in chicken fat. In product design, therefore, judgements should not rest on the ingredient name alone: checking the ingredient specification and controlling the manufacturing process become important.

It is possible that the amount of residual protein varies with the refining method. But as long as no published data confirming this can be identified, there is insufficient basis for claiming that “this chicken fat is low in residual protein.”

Two real formulation examples follow.

For one veterinary hydrolysed-protein diet sold in the North American market, the ingredient listing on the manufacturer’s official site runs “rice, hydrolysed soy protein, chicken fat,” with chicken fat third. This is no more than one example of chicken fat being used in a hydrolysed-protein diet. It is not data showing that chicken fat is safe or that it does not cause allergic reactions, and the inclusion rate is not published.

In another veterinary diet for the UK market, the oil is declared as “animal fats” without specifying the species, and “hydrolysed poultry liver” is listed separately as the protein source. Because EU and UK labelling rules do not require the source species of oils and fats to be stated, the label alone cannot tell you whether these “animal fats” are chicken-derived.

It should be added that using hydrolysed protein is a basic design approach for this kind of veterinary diet. The presence of poultry-derived ingredients in the listing is therefore not in itself inconsistent with the product’s design. These are simply real formulation examples; note that they do not demonstrate the safety or the allergenicity of chicken fat.

(4) Egg

In research, egg is treated as a separate ingredient from chicken meat. In the several primary studies collated by Mueller et al. (2016) as well, chicken meat and chicken egg were each challenge-tested as separate ingredients.

That, however, shows only that the individual studies examined chicken meat and egg separately. We were not able to identify a textbook or guideline stating that “egg must always be assessed separately from chicken meat.”

Nor could we identify research using actual feeding provocation tests to establish what proportion of dogs reactive to chicken meat also react to egg.

Accordingly, neither “dogs that react to chicken meat also react to egg” nor “egg is separate from chicken meat, so it can be used” can be determined from current canine research data.

(5) Hydrolysis

Hydrolysis is not a type of ingredient but a processing method that breaks protein into smaller pieces. It is nonetheless an important consideration when designing allergy-conscious products.

For hydrolysed protein, there is no common standard that would let you conclude “bring the molecular weight below this figure and it is safe.” The indicative molecular weights given in peer-reviewed papers also differ between sources.

Trace those figures back further and the sources include not only systematic reviews in dogs but also non-peer-reviewed conference abstracts and reviews in humans. In other words, they are not thresholds derived from trials that fed dogs protein at different molecular weights and doses to establish the level at which reactions begin (Lesponne et al., 2018 / Masuda et al., 2020).

Product design and allergy diagnosis are different things

When you design an allergy-conscious food, you do not investigate the allergic trigger of each individual dog before development. What the product development side decides is “which ingredients to exclude, and what kind of use the product is for.” Whether that product suits an individual dog is judged separately, through a diet trial under veterinary supervision.

The development side decides what the product excludes; the veterinarian judges what the dog reacts to.

What the development side decides

In product development you do not identify an allergic trigger common to all dogs and then build the recipe. You decide which ingredients to exclude and which protein sources to use on the basis of conditions such as the following.

• What intended use to assume
• Which ingredients to exclude
• Whether to adopt novel protein or hydrolysed protein
• What to declare and claim on the pack and in advertising
• Whether the factory can segregate ingredients and control cross-contamination
• How far the origin of ingredients and the manufacturing process can be verified
• How much risk remains even after those measures

The phrase “allergy-conscious” therefore does not by itself determine a recipe. The exclusion scope and control level have to be set out concretely, in line with the intended use and the label wording.

Suitability for an individual dog is judged by a diet trial

Reviews as of 2026 still hold that the most reliable way to diagnose adverse food reactions in dogs and cats is the combination of an elimination diet trial and a provocation test.

In an elimination diet trial, a diet free of the suspected ingredients is fed for a set period to see whether the signs improve. In the provocation test that follows, the suspect ingredient is fed again to see whether the signs recur.

Tests using serum IgE and IgG, saliva or hair are concluded to lack sufficient reliability and not to substitute for these diet trials (Mueller & Olivry, 2017 / Udraite Vovk & Mueller, 2026). This is an overall conclusion covering dogs and cats.

Diet trials carry their own uncertainty, however. In a double-blind, placebo-controlled trial in 12 dogs, half also reacted positively to placebo. The authors themselves note that doubts remain about interpreting provocation tests with single food items (Sofou et al., 2026).

In short, even a product designed as an allergy-conscious food will not necessarily suit every dog. Suitability for an individual dog cannot be settled from the product name or the ingredient list; it has to be confirmed under veterinary supervision.

Feeding duration also matters for a diagnostic elimination diet

In a systematic review pooling case data from 209 dogs, more than 95% of dogs showed improvement in skin signs within eight weeks of starting the elimination diet. To detect over 90% of adverse food reactions presenting with skin signs, continuing the elimination diet trial for at least eight weeks is recommended (Olivry et al., 2015).

That said, the researchers did not test 209 dogs under identical conditions. The analysis pooled, from several earlier studies, the cases for which the timing of improvement after starting the elimination diet was known. It therefore does not directly prove through a comparative trial that eight weeks is the optimal duration.

So if you are developing a product to be used as a diagnostic elimination diet, the pack size and sales unit that allow eight weeks of continuous feeding also become design considerations. For a general “allergy-conscious food,” by contrast, designing for eight weeks’ supply is not a requirement.

The role of the product development side is not to diagnose individual dogs. It is to make the intended use, the excluded ingredients, the label wording and the scope of manufacturing control explicit — and to state concretely what can be controlled and what cannot be guaranteed.

A “not used” claim is backed by process control

Since there is no common standard for residual protein or contamination levels, the ingredient name alone cannot tell you at which steps the ingredient in question is used, or how much contamination risk remains.

To make a “not used” claim, you need to substantiate — through ingredient specifications, process control, test results, manufacturing records and the like — that the ingredient concerned is not used at any stage of manufacture.

A “not used” claim does not, however, guarantee the complete absence of the substance including unintended cross-contamination, nor that no dog will develop allergic signs.

“What is not used” cannot be substantiated by ingredient names alone. Ingredient verification, contamination prevention, testing and records have to be combined to determine what can legitimately be declared.

Ingredients absent from the label are sometimes detected

In a 2018 systematic review pooling 18 studies on labelling discrepancies in dog and cat pet food, the proportion of products in which an unlabelled ingredient was detected ranged from 0% to 83% between studies, with a median of 45%.

Even restricting the view to elimination-diet products claiming “novel protein” or “limited ingredient,” unlabelled ingredients were detected in 33% to 83%.

Only a subset of ingredients was tested in each study, however. The authors note that the real rate of labelling discrepancy could be higher still if the range of ingredients tested were widened. Note also that the review covers studies published up to January 2018.

On the other hand, detecting an unlabelled ingredient does not mean that the product actually causes allergic signs. As of August 2026 we could not identify any study that fed products with confirmed labelling discrepancies to dogs or cats allergic to the ingredient concerned and examined whether signs occurred.

“A component derived from an ingredient was detected” and “a dog develops signs” therefore have to be kept apart.

A brand can set its own control criteria

In the study by Lesponne et al. (2018), a tolerance level of 0.5% was set for unintended incidental protein contamination in one extensively hydrolysed diet product line.

A corresponding total DNA concentration of 2.1 µg/g was then given a safety margin, and 1.2 µg/g adopted as the product acceptance criterion.

However, 0.5% and 1.2 µg/g are internal criteria set for that product line. They are neither a statutory standard nor a common standard applicable across the pet food industry. The authors themselves state that there is no clear consensus on how much contamination is acceptable. Note also that employees of Royal Canin / Mars are among the authors of that study.

As this shows, if the analytical method and the acceptance criteria are set out concretely, it is possible for a brand owner and an OEM factory to establish their own contractual control criteria for lot-by-lot release decisions.

What you cannot do is position those figures as “a safety threshold at which no dog will develop allergic signs.” What can be set are control criteria for ingredients, manufacturing and testing — not a common threshold guaranteeing clinical safety.

Residual protein may be reducible

The possibility that manufacturing steps can reduce the protein remaining in oil has been reported in human food research.

In a small-scale trial close to real production, 70–130 ppm of cashew protein was detected in shared oil after 15 batches of cashews had been fried. When that oil continued to be used, 23.0 ppm was detected from peanuts and 193.5 ppm from potato chips.

On the other hand, treating the oil with 11 µm or 25 µm filters or with diatomaceous earth was reported to bring protein levels of over 200 ppm down to below 10 ppm (Chen, Baumert & Downs, 2026).

This indicates that process controls such as filtration may reduce the protein contained in oil.

The subject of that research, however, was shared vegetable frying oil used in human food, not chicken fat. The same study also concludes that the concentrations detected could pose a health risk to people with food allergies.

You therefore cannot transfer this result directly to chicken fat, nor conclude that “filtration makes it safe.” For chicken fat, the treatment method, test target, limit of detection, testing frequency and acceptance criteria all have to be set individually.

What a specification can fix, and what it cannot decide

Setting figures and test conditions in a specification does not by itself rule out every risk. You need to separate what can be controlled from what the results cannot tell you.

Issue	What a specification can fix	What the result alone cannot decide
Chicken fat	Analytical method, target protein, limit of quantification, testing frequency, reporting format	That no allergen is present, or that no dog will react
Species of origin of the ingredient	Single-species declaration, specifying species-specific testing	A single-species guarantee based on a declaration such as “poultry” alone
Line changeover	Cleaning procedure, changeover records, handling of the preceding flush lot	That cross-contamination is completely zero
Testing	Target species, analytical method, limit of detection, testing frequency, acceptance criteria	Treating a negative result as complete absence of the ingredient concerned
Hydrolysis	Origin of the ingredient, definition of the degree of hydrolysis, measurement method	That below a given molecular weight no dog will react

Ten points to check with an OEM factory

Verification supporting “free from chicken-derived ingredients” splits into two parts: checking the recipe and ingredients, and checking the manufacturing process.

The first thing to check is the recipe and ingredient specification for each product. Even within the same “hydrolysed diet” range, the ingredients used differ by SKU. In practice there are cases where one product lists hydrolysed chicken liver and chicken gizzard while another product in the same range does not.

So a range name or product category such as “hydrolysed diet” or “limited ingredient diet” cannot tell you whether poultry-derived ingredients are present. The ingredient declaration, formulation sheet and ingredient specification have to be checked SKU by SKU.

It is not only main ingredients such as meat and meal that need checking. Sub-ingredients and processing aids like the following are also in scope.

• Main and sub-ingredients
• Carriers
• Digests
• Coating materials

Next comes the manufacturing process. Even where the recipe contains no chicken-derived ingredient, the same factory or equipment may be producing other products that do. The ingredient declaration alone cannot tell you about the potential for cross-contamination via shared equipment, production sequence, cleaning or rework.

So check, at every step — intake, storage, weighing, charging, mixing, forming, drying, coating and packing — whether chicken-derived ingredients may be used or introduced.

The ten points below are not a checklist laid down uniformly by law. They are practical examples of how to substantiate that “the ingredient concerned is not used at any stage of manufacture,” adapted to the factory’s equipment and production conditions.

Point to check	Why it matters
(1) Dedicated tank or shared tank	If shared, you need to check the residue risk from the previously manufactured product
(2) Extent of shared pipework and weighing systems	Without knowing the shared extent, you cannot define what to clean or what to record
(3) Sharing of coating equipment	Coating materials such as digests are an easily overlooked contamination route
(4) Production changeover sequence	Residue risk changes depending on whether it follows a product containing chicken-derived ingredients
(5) Cleaning procedure and cleaning validation	You need to confirm not just that a cleaning procedure exists but that it works
(6) Handling of flush material	Disposal, re-charging or use in other products each send the residue somewhere different
(7) Whether rework may be charged, and its records	Rework can bring in ingredients that are meant to be excluded
(8) Notification conditions for change control	If an ingredient or process changes, the basis for the free-from claim changes
(9) Sampling plan	Without fixed sampling locations, timing and number of samples, test results cannot be evaluated properly
(10) Record retention period and how records are provided	Check that the evidence supporting the claim is retained for the prescribed period and can be produced on request

If testing is carried out, writing “PCR tested” is not enough. The target species, whether DNA or protein is being examined, the analytical method, the limit of detection, the testing frequency and the acceptance criteria all have to be specified.

In addition, Article 11 of the Fair Competition Code on Pet Food Labelling sets retention periods for the material underpinning a claim, such as formulation design documents, spec sheets and analytical data.

It does not, however, prescribe the format in which that material is to be provided. So OEM contracts and quality specifications should set out, alongside the retention period, how the material will be provided, by when, and how changes will be notified.

These ten points are examples of the evidence and controls that support a “not used” claim. Rather than demanding every item uniformly, work back from the wording you want to use and from the factory’s equipment and production situation to decide the necessary scope of verification and level of control.

The line a claim must not cross

Allergy-related claims require different evidence depending on what they say. To state “chicken-free,” for example, you need not only to keep chicken out of the recipe but to confirm that it is not used at any stage of manufacture.

“Not used,” “zero cross-contact” and “clinically hypoallergenic” are separate propositions. The evidence that lets you write each of them is different.

Four types of claim

The following is organised on the basis of the Fair Competition Code on Pet Food Labelling (Japan Fair Trade Commission / Consumer Affairs Agency Notification No. 4 of 30 September 2024, in force 1 October 2024) and its Enforcement Regulations. It binds businesses participating in the Code. It governs labelling for the Japanese market; other markets are governed by their own rules.

Type of claim	Typical product type	Basis in the Code	What must be satisfied
(1) Emphasising the use or inclusion of chicken	General products	Article 7 of the Code	Unless the ingredient accounts for 5 per cent or more of the net content, its use may not be indicated in the product name, illustrations, photographs, descriptive text and the like
(2) Chicken / chicken-derived ingredients not used	Limited-ingredient products / chicken-free products	Article 8(4) of the Code + Article 7(4)(a) of the Enforcement Regulations	The name of the ingredient stated as not added must be clearly indicated alongside the claim, and it must be verifiable that the ingredient is not used at any stage of manufacture of the pet food
(3) Hypoallergenic / allergy-conscious	General products claiming allergy consideration	No direct definition in the Fair Competition Code; judged case by case under the Premiums and Representations Act and the general prevention-of-misleading-claims principle in Article 10 of the Code	Prepare reasonable grounds corresponding to the claim, and do not imply treatment, prevention or improvement of disease
(4) Treatment or improvement of disease	Elimination diets and veterinary diets used under veterinary supervision	Article 3(4) and Article 6 of the Code / Article 5 of the Enforcement Regulations / Article 8(3) of the Enforcement Regulations	There is a route to state the disease or health condition for which the product applies as a “veterinary diet” and to report it to the Fair Trade Council. Conversely, labelling that claims or implies the efficacy of a veterinary medicinal product, or comparable efficacy, is a category of improper representation

Notes on the table’s categories and labelling terms

“Typical product type” is a classification used for explanation

The “typical product type” column is a classification created to make the explanation clearer in this article; it is not a formal category under the Code. The formal purpose categories under Article 4(2) of the Fair Competition Code on Pet Food Labelling are the following four.

• Complete and balanced food
• Treats
• Veterinary diet
• Food for other purposes

Article 10(2) of the Code also prohibits, as an improper representation, any labelling that does not conform to the labelling standards in Article 7 or Article 8.

For the EU, do not swap “intolerance” for “allergy”

For products aimed at the EU, the terms you may use also require care.

Part B, entry 13 of the Annex to Commission Regulation (EU) 2020/354 sets out the official wording “Reduction of ingredient and nutrient intolerances.”

Footnote (6) permits the words “ingredient and nutrient” to be replaced with the name of the specific intolerance concerned. It does not, however, extend to replacing “intolerance” with “allergy.”

The term “dietetic” is likewise not a freely usable expression. Under Article 18 of Regulation (EC) No 767/2009, it is reserved for feed intended for particular nutritional purposes.

“No chicken meat” and “no chicken-derived ingredients” cover different scopes

“No chicken meat” normally indicates that the flesh of the chicken is not used. “Free from chicken-derived ingredients,” by contrast, is a broader expression covering all ingredients derived from chicken.

If you are going to claim “free from chicken-derived ingredients,” you have to define in advance whether the exclusion extends beyond chicken meat to the following ingredients.

• Chicken fat
• Egg
• Poultry meal
• Digests
• Coating materials
• Other chicken-derived sub-ingredients

Rather than fixing the label wording first, decide the range of ingredients to exclude and then check the ingredient specifications and the manufacturing process. This is the point at which the factory checklist set out in the previous section matters.

In the EU market there are in fact products claiming a “single animal protein source” while listing poultry oil and poultry hydrolysate among the ingredients (verified 3 August 2026).

Under EU dietetic feed rules, oils and hydrolysates may be treated as outside the scope of a “single protein source” claim. The labelling alone therefore cannot be judged as false.

What this example shows is that a product category or claim such as “single protein source” cannot by itself tell you that poultry-derived oils and sub-ingredients are absent. Note that the inclusion rates for the product concerned are not published.

“Not used” is a process requirement

This is the most practically consequential point in this article. Whether you can write “chicken-free” is not a recipe question but a manufacturing process question.

Article 7(4)(a) of the Enforcement Regulations provides that terms such as “no added” or “not used” may be applied to non-additive ingredients only where the name of the ingredient stated as not added is clearly indicated alongside, and where “it can be verified that the ingredient concerned is not used at any stage of manufacture of the pet food.” What the provision requires is a state of being verifiable; it does not lay down a uniform list of specific tests. The ten points in the previous section are practical examples for assembling that substantiation.

“The ingredient concerned is not used,” “cross-contact is zero” and “the product is clinically hypoallergenic” are separate propositions. Article 7(4)(a) requires the first; it does not guarantee the second or the third.

Note also that emphasising “not used” for an ingredient or additive that would not normally be used, in a way that misleads consumers into thinking the product is superior in quality, may fall within the category of improper representation under Article 8(1)(a) of the Enforcement Regulations (relating to Article 10(6) of the Code), even where Article 7(4)(a) is satisfied.

Where “hypoallergenic” stands

“Hypoallergenic” is a term for which no definition exists in the EU feed labelling Regulation (EC) 767/2009, the PARNUT list Regulation (EU) 2020/354, the FEDIAF labelling code (October 2019 edition), or the AAFCO model regulations for pet food (2023 approved version; verified against the AAFCO public PDF version).

Whether it may be used is therefore judged case by case: in the EU under the general prevention-of-misleading-claims and claim requirements of Articles 11 and 13 of 767/2009, and in the United States within the framework of misbranding and drug claims.

Japan’s Fair Competition Code contains no definition either. Article 7(1) of the Enforcement Regulations sets criteria for terms such as “low,” but these apply to the presence or amount of particular nutrients and do not apply directly to allergens.

Regulatory note

Expressions suggesting the treatment, prevention or improvement of disease cannot be used in Japan, the EU or the United States. In Japan, Article 8(3) of the Enforcement Regulations makes labelling that claims or implies the efficacy of a veterinary medicinal product, or comparable efficacy, a category of improper representation.

In the United States, the FDA has formally taken the position that express or implied claims to treat, prevent or mitigate disease require approval as a new animal drug. In the EU, Article 13(3)(a) of Regulation (EC) No 767/2009 prohibits claiming that feed will prevent, treat or cure a disease.

Retention of supporting material

Article 7(2) of the Premiums and Representations Act provides that, where the Consumer Affairs Agency has requested submission of reasonable material substantiating a representation in connection with an order for action, and the business does not submit it, the representation is deemed — for the purposes of that order — to be a misleading representation of superior quality (falling under Article 5(1)).

Article 11 of the Pet Food Fair Competition Code likewise provides that material underpinning a representation — formulation design documents, spec sheets, analytical data and the like — is to be retained for two years from the final manufacturing date of the product concerned where the best-before period is within two years, and until the expiry of the best-before date of the final manufactured product where it exceeds two years.

In practice it makes sense to design the process records for a free-from claim into that same retention scope.

Frequently asked questions (FAQ)

Q1. Can chicken fat be included in an elimination diet?

Within published regulations and primary research, we could not identify a generalisable declaration threshold or clinical threshold for residual protein in chicken fat. It is not that measurement is impossible; it is that, in the dog pet food field, no common standard converting a measured result into clinical safety has been published. Whether to include it is therefore a matter for the attending veterinarian. On the product side, the realistic step is to make the presence or absence of chicken fat clear in the ingredient declaration.

Q2. Can an ingredient declared as “poultry meal” be assumed to be chicken only?

No, it cannot be determined. Under both the AAFCO and Canadian CFIT definitions, naming the species is optional (CFIT is an ingredients table for livestock feed). If you need to work on a single-species basis, you have to specify the species in the ingredient specification and require evidence.

Q3. Can I put “chicken-free” on the pack?

Beyond not using chicken as an ingredient, verification of the process is a requirement. Article 7(4)(a) of the Enforcement Regulations of the Fair Competition Code provides that terms such as “no added” or “not used” may be applied only where the name of the ingredient stated as not added is clearly indicated alongside, and where it can be verified that the ingredient concerned is not used at any stage of manufacture of the pet food.

What the provision requires is a state of being verifiable; it does not uniformly designate particular tests. Note also that “not used” does not mean zero cross-contact or clinical hypoallergenicity.

Q4. How often do animal species not on the ingredient list turn up in the product?

In a 2018 systematic review pooling 18 reports, the proportion of products in which an unlabelled ingredient was detected ranged from 0–83% between studies (median 45%), and 33–83% even when restricted to “novel / limited ingredient” products for elimination diets. The authors themselves state, however, that the real non-conformity rate is very likely to be higher than this. As set out in the main text, detection and the onset of signs are not linked.

Q5. Can I write “hypoallergenic” on the pack?

“Hypoallergenic” is a term for which no definition exists in the EU’s 767/2009 or 2020/354, the FEDIAF labelling code, or the AAFCO model regulations. Japan’s Fair Competition Code contains no definition either.

It is not a prohibited word, but the supporting evidence is yours to prepare. In Japan, Article 7(2) of the Premiums and Representations Act imposes a regime on unsubstantiated advertising, so you need material you can submit on request (for retention periods, please see “Retention of supporting material” in the main text).

Used in a context suggesting the treatment or improvement of allergy, however, it may amount to an improper representation under Article 8(3) of the Enforcement Regulations. Because whether a claim is permissible is judged case by case, please consult the Pet Food Fair Trade Council before finalising the wording.

References and sources

1. Mueller RS, Olivry T, Prélaud P. “Critically appraised topic on adverse food reactions of companion animals (2): common food allergen sources in dogs and cats,” BMC Veterinary Research 12:9, 2016 https://doi.org/10.1186/s12917-016-0633-8 (accessed 2026-08-01)
2. Olivry T, Mueller RS, Prélaud P. “Critically appraised topic on adverse food reactions of companion animals (1): duration of elimination diets,” BMC Veterinary Research 11:225, 2015 https://doi.org/10.1186/s12917-015-0541-3 (accessed 2026-08-01)
3. Mueller RS, Olivry T. “Critically appraised topic on adverse food reactions of companion animals (4): can we diagnose adverse food reactions in dogs and cats with in vivo or in vitro tests?,” BMC Veterinary Research 13:275, 2017 https://doi.org/10.1186/s12917-017-1142-0 (accessed 2026-08-01)
4. Olivry T, Mueller RS. “Critically appraised topic on adverse food reactions of companion animals (5): discrepancies between ingredients and labeling in commercial pet foods,” BMC Veterinary Research 14:24, 2018 https://doi.org/10.1186/s12917-018-1346-y (accessed 2026-08-01)
5. Olivry T, Pucheu-Haston CM, Mayer U, Bergvall K, Bexley J. “Identification of major and minor chicken allergens in dogs,” Veterinary Dermatology 33(1):46-e16, 2022 https://doi.org/10.1111/vde.13029 (accessed 2026-08-01)
6. Olivry T, Bexley J, Mougeot I. “Extensive protein hydrolyzation is indispensable to prevent IgE-mediated poultry allergen recognition in dogs and cats,” BMC Veterinary Research 13:251, 2017 https://doi.org/10.1186/s12917-017-1183-4 (accessed 2026-08-01)
7. Olivry T, Bexley J. “Cornstarch is less allergenic than corn flour in dogs and cats previously sensitized to corn,” BMC Veterinary Research 14:207, 2018 https://doi.org/10.1186/s12917-018-1538-5 (accessed 2026-08-01. This study was conducted by Avacta Animal Health with funding from Royal Canin)
8. Masuda K, Sato A, Tanaka A, Kumagai A. “Hydrolyzed diets may stimulate food-reactive lymphocytes in dogs,” Journal of Veterinary Medical Science 82(2):177-183, 2020 https://doi.org/10.1292/jvms.19-0222 (accessed 2026-08-01)
9. Lesponne I, Naar J, Planchon S, Serchi T, Montano M. “DNA and Protein Analyses to Confirm the Absence of Cross-Contamination and Support the Clinical Reliability of Extensively Hydrolysed Diets for Adverse Food Reaction-Pets,” Veterinary Sciences 5(3):63, 2018 https://doi.org/10.3390/vetsci5030063 (accessed 2026-08-01)
10. Sofou EI, Samuel E, Aleksandrova S, Chatzis M, Saridomichelakis MN. “Randomised, Double-Blinded, Placebo-Controlled Challenge Test With Single Food Items in Dogs With Atopic Dermatitis and Adverse Food Reactions,” Veterinary Dermatology 37(2):247-256, 2026 https://doi.org/10.1111/vde.70041 (accessed 2026-08-01)
11. Udraite Vovk L, Mueller RS. “Adverse Food Reactions in Dogs and Cats,” Veterinary Clinics of North America: Small Animal Practice 56:579-589, 2026 https://doi.org/10.1016/j.cvsm.2025.11.003 (accessed 2026-08-01)
12. Pagani E, Soto del Rio MdlD, Dalmasso A, Bottero MT, Schiavone A, Prola L. “Cross-contamination in canine and feline dietetic limited-antigen wet diets,” BMC Veterinary Research 14:283, 2018 https://doi.org/10.1186/s12917-018-1571-4 (accessed 2026-08-01)
13. Chen S, Downs ML. “Development of a Mass Spectrometry-Based Method for Quantification of Total Cashew Protein in Roasting Oil,” Journal of AOAC International 107(3):443-452, 2024 https://doi.org/10.1093/jaoacint/qsae019 (accessed 2026-08-03)
14. Chen S, Baumert JL, Downs ML. “Quantification, Risk Assessment, and Mitigation of Cashew Allergen Cross-Contact in Shared Roasting Oil Systems,” Journal of Food Protection 89(1):100678, 2026 (online ahead of print 2025) https://doi.org/10.1016/j.jfp.2025.100678 (accessed 2026-08-03)
15. Canadian Food Inspection Agency, “Canadian Feed Ingredients Table (CFIT)” (a document incorporated by reference in the Feeds Regulations, 2024), date modified: 2026-06-26 https://inspection.canada.ca/en/animal-health/livestock-feeds/documents-incorporated-reference/canadian-feed-ingredients-table (accessed 2026-08-03)
16. Canadian Food Inspection Agency, “RG-1 Regulatory Guidance: Feed Registration Procedures and Labelling Standards,” date modified: 2026-04-15 https://inspection.canada.ca/en/animal-health/livestock-feeds/regulatory-guidance/rg-1 (accessed 2026-08-03)
17. European Commission, “Commission Regulation (EU) 2017/1017 amending Regulation (EU) No 68/2013 on the Catalogue of feed materials (Annex Part C Chapter 9),” 2017 https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32017R1017 (accessed 2026-08-01)
18. European Commission, “Commission Regulation (EU) 2022/1104 amending Regulation (EU) No 68/2013 on the Catalogue of feed materials,” 2022 https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32022R1104 (accessed 2026-08-01)
19. European Parliament and Council, “Regulation (EC) No 767/2009 on the placing on the market and use of feed (consolidated text 02009R0767 — 26.12.2018),” 2018 https://publications.europa.eu/resource/celex/02009R0767-20181226 (accessed 2026-08-01)
20. European Commission, “Commission Regulation (EU) 2020/354 establishing a list of intended uses of feed intended for particular nutritional purposes (consolidated text 02020R0354 — 22.04.2025),” 2025 https://publications.europa.eu/resource/celex/02020R0354-20250422 (accessed 2026-08-01)
21. AAFCO, “What’s in the Ingredients List?” (definitions of Poultry Meal / Poultry By-Product Meal) https://www.aafco.org/consumers/understanding-pet-food/whats-in-the-ingredients-list/ (accessed 2026-08-01)
22. AAFCO, “Model Regulations for Pet Food and Specialty Pet Food Under the Model Bill (20230403 final; approved by members 2023-07-31),” 2023 https://www.aafco.org/wp-content/uploads/2023/04/9._FINAL_PFC_MBRC_for_Pet_Food_and_Specialty_Pet_Food.pdf (accessed 2026-08-01) (Public PDF version; the text carries a “DRAFT!!” marking. The authoritative version is the paid AAFCO Official Publication, 2026 edition)
23. U.S. Food and Drug Administration, “Animal Food Labeling and Pet Food Claims,” content current as of 2026-03-02 https://www.fda.gov/animal-veterinary/animal-food-feeds/animal-food-labeling-and-pet-food-claims (accessed 2026-08-03)
24. FEDIAF, “Code of Good Labelling Practice for Pet Food,” October 2019 https://europeanpetfood.org/wp-content/uploads/2022/02/FEDIAF_labeling_code_2019_onlineOctober2019.pdf (accessed 2026-08-01)
25. Pet Food Fair Trade Council, “Fair Competition Code on Pet Food Labelling (Japan Fair Trade Commission / Consumer Affairs Agency Notification No. 4 of 30 September 2024, in force 1 October 2024) and its Enforcement Regulations” https://pffta.org/pdf/regulations.pdf / Consumer Affairs Agency, “Act against Unjustifiable Premiums and Misleading Representations (Act No. 134 of 1962)” https://laws.e-gov.go.jp/law/337AC0000000134 (accessed 2026-08-01)