DDELICATE +38 063 545-32-19
Choice and composition · 2026-11-09

Grain-free dog food: who needs it and who does not

Serhii Tverdyi, DELICATE production technologist · published 2026-11-09 · updated 2026-11-09

Grain-free food is needed by a dog with a confirmed reaction to cereals specifically, and there are few such dogs. Food allergy affects 1–2 % of dogs brought to a vet (Olivry, Mueller, 2017), and among 297 dogs with a confirmed diagnosis a cereal was the allergen in 19 % of cases (Mueller, Olivry, Prélaud, 2016): recalculated, that is 2–4 dogs in 1,000 against 8–16 in 1,000 for animal protein. Between 2014 and November 2022 the US regulator FDA received 1,382 reports of dilated cardiomyopathy in dogs: in the report of 27 June 2019, over 90 % of those foods were grain-free and 93 % contained peas or lentils.

Key points

Who actually needs grain-free food?

Grain-free food is food without corn, soy, wheat, rice, barley or other cereals: that is exactly how FDA defines the term. It is needed in two cases: when a vet has confirmed a reaction to a specific cereal, and when cereals are removed for the duration of an elimination diet.

Food allergy is a reaction to a protein in food, and only an elimination diet with the food reintroduced confirms it. A blood test does not, and neither does the label on the bag.

The word “hypoallergenic” and the claim “grain-free” are not related. The FEDIAF Code of Good Labelling Practice of 2019 (clause 5.2.2.2) requires a “free from Y” claim to mean the absence of even traces of Y, and adds that such claims must not give the impression that foods containing that substance are unsafe.

How many dogs are allergic to cereals specifically?

Roughly 2–4 dogs in 1,000 brought to a vet are allergic to any cereal. Among 297 dogs with a confirmed diagnosis, wheat was the allergen in 38, corn in 13, rice in 5 and beef in 102 (Mueller, Olivry, Prélaud, 2016).

How many dogs in 1,000 seen by a vet are allergic to a given source
SourceShare among dogs with allergy, %Per 1,000 dogs seen by a vet, estimate
Beef343.4–6.9
Chicken151.5–3.0
Wheat131.3–2.6
Corn40.4–0.9
Rice20.2–0.3
Any cereal, at most191.9–3.8
Any animal protein, at most797.9–15.9

The share from Mueller, Olivry, Prélaud (2016) multiplied by the 1–2 % prevalence from Olivry, Mueller (2017), own calculation. Upper-bound estimate: some dogs reacted to several foods.

Animal allergens are four times more common among confirmed cases than cereal onesBeef6.9 dogs per 1,000Any cereal3.8 dogs per 1,000Dairy3.4 dogs per 1,000Chicken3 dogs per 1,000Wheat2.6 dogs per 1,000Corn0.9 dogs per 1,000Rice0.3 dogs per 1,000
Animal allergens are four times more common among confirmed cases than cereal ones
Source: Olivry, Mueller 2017; Mueller, Olivry, Prélaud 2016; own calculation

FEDIAF (2025, clause 7.6.4) cautions that a high frequency of individual allergens may simply reflect the fact that dogs eat those proteins most often.

The myth that cereals are the main allergen: what does the arithmetic show?

The claim that cereals are the main allergen and that grain-free food therefore works as prevention does not survive the arithmetic. Take a Border Collie of 3 years, 18 kg, healthy, with no confirmed reaction to food. The chance that cereals are her problem, per the table above, is 0.2–0.4 %.

Now the owner picks a grain-free bag based on chicken. The food removes rice, which 2 % of dogs with allergy react to, and keeps chicken, which 15 % react to.

The claim breaks down where a cereal is swapped for another protein: the rarer allergen is gone and the more common one stays. If a reaction is already there, the order is different: examination, diet, and only then the choice of composition. The signs are covered in the article on symptoms of food allergy in dogs.

Now a different dog, one with a confirmed reaction to wheat. A “grain-free” bag based on peas removes both wheat and rice, which only 2 % of dogs with allergy react to. A food without wheat, with the cereals left in, is enough.

What replaces cereals, and does the food end up with more meat?

Cereals are replaced with legumes and potato, and that does not add protein. The Vet-LIRN laboratory, commissioned by FDA, compared grain-free and grain-inclusive foods: protein, fat, taurine and resistant starch on a dry matter basis “were similar”.

Legumes are peas, lentils, beans and chickpeas. In the FEDIAF labelling code of 2019 they fall under the “Vegetables” category and cereals under “Cereals”, so a pea-based food is formally grain-free. According to FDA, such foods usually contain more legumes.

DELICATE is a manufacturer of dry food for adult dogs from Dnipro: by the label, meat and animal by-products make up at least 60 % of the mix, followed by rice and corn, crude protein is 32 %, and energy is 397 kcal per 100 g (calculated per NRC 2006). There is no wheat, pea, lentil or potato in the composition, so the chicken food is not grain-free as FDA defines it. A 10 kg bag costs UAH 1,188 with delivery: for an 18 kg Border Collie that is 246 g a day, about UAH 29 per day and 40 days per bag. It suits a healthy adult dog from 10 months of age; with a confirmed reaction to chicken or beef the composition is chosen without that protein.
See the food — UAH 1,188

What did FDA find about grain-free foods and dogs' hearts?

FDA established no causal link, but it did collect figures. In July 2018 the regulator announced an investigation into reports of dilated cardiomyopathy in dogs that ate foods with a high share of peas, lentils or potato among the first ten ingredients. The report of 27 June 2019 recorded 524 reports for 2014–2019: 560 dogs and 119 deaths.

FDA reports of dilated cardiomyopathy in dogs by period
PeriodReportsMonthsPer month
01.2014–11.2018329595.6
12.2018–04.2019190538.3
05.2019–07.20206081540.4
08.2020–01.11.2022255279.4
Total1382

Figures from the Q&A of the FDA Center for Veterinary Medicine (current as of 19.08.2024); reports per month are our own calculation. FDA does not name a cause for the spike.

The regulator's position verbatim: “FDA has no definitive information indicating that the diets are inherently unsafe”. The independent data are contradictory: a 2025 narrative review sees the clearest link with legumes, peas above all (Mornard et al., 2025), while an 18-month trial in 60 dogs found no difference in cardiac biomarkers (Morris et al., 2025; funded by Hill's). Questions about a dog's heart are for a vet.

Does a dog digest rice and corn?

Starch digestibility after extrusion exceeds 98 % in all six diets tested by Carciofi et al. (2008): on rice, corn, sorghum, cassava, peas and lentils. Dogs absorbed dry matter and energy best from rice and cassava, and worst from peas and lentils.

The ability to digest starch is inherited. In 392 dogs the median was 10 diploid copies of the pancreatic amylase gene AMY2B against 2 copies in 51 wolves, and each extra copy adds about 5.4 % of serum amylase activity (Arendt et al., 2016). There are exceptions: the Siberian Husky has a median of 3 copies, the Greenland Dog 2.5 and the dingo 2.

How to read a bag labelled “grain-free”?

A food is judged by its ingredient list, and the claim on the front of the bag does not do that job. Three things worth looking at.

What to do if you are choosing a bag right now?

Questions and answers

Does my dog need grain-free food?

If a vet has not confirmed a reaction to cereals, there is no need for it: food allergy affects 1–2 % of dogs seen by a vet (Olivry, Mueller, 2017), and any cereal accounts for 19 % of those cases (Mueller, Olivry, Prélaud, 2016), that is 2–4 dogs in 1,000.

How many dogs are allergic to wheat, corn and rice specifically?

Wheat is an allergen for 1.3–2.6 dogs in 1,000 seen by a vet, corn for 0.4–0.9, rice for 0.2–0.3, while any animal protein accounts for 7.9–15.9 (own recalculation from the 2016 and 2017 reviews).

Is it true that grain-free food harms a dog's heart?

There is no proven link. FDA received 1,382 reports of dilated cardiomyopathy between 2014 and November 2022, and over 90 % of the foods in them were grain-free, but the regulator has no definitive data on such diets being unsafe.

Are rice and corn harmful in dog food?

The data say no. Starch digestibility after extrusion exceeds 98 % for both rice and corn, and dry matter digestibility was the highest of the six sources tested for rice (Carciofi et al., 2008).

Does “grain-free” mean there is more meat in the food?

No. The Vet-LIRN laboratory compared grain-free and grain-inclusive foods and found similar levels of protein, fat and taurine on a dry matter basis.

Sources
  1. FDA Center for Veterinary Medicine. FDA Investigation into Potential Link between Certain Diets and Canine Dilated Cardiomyopathy, update of 27.06.2019 (page current as of 23.12.2022). — The investigation was announced in July 2018: the reports concerned foods, many of them labelled “grain-free”, with a high share of peas, lentils, other legumes and/or potato among the first ten ingredients; some cases were in breeds with no known genetic predisposition. Table 1: from 01.01.2014 to 30.04.2019 there were 524 reports (515 about dogs, 9 about cats), 560 dogs and 119 deaths. “Grain-free” as FDA defines it means without corn, soy, wheat, rice, barley or other cereals. “More than 90 percent of products were "grain-free", and 93 percent of reported products had peas and/or lentils”; potato appeared in a far smaller proportion. The most common proteins in these foods were chicken, lamb and fish, yet “No one animal protein source was predominant”. Product Testing section: Vet-LIRN compared grain-free and grain-inclusive foods, and protein, fat, total taurine, cystine, methionine and resistant starch on a dry matter basis “were similar”. “Adverse event numbers… by themselves, do not supply sufficient data to establish a causal relationship”. We give no link: fda.gov is not on the list of domains allowed by CLAUDE.md.
  2. FDA Center for Veterinary Medicine. Questions & Answers: FDA's Work on Potential Causes of Non-Hereditary DCM in Dogs (page current as of 19.08.2024). — Reports by period: 01.2014–11.2018 — 329; 12.2018–04.2019 — 190; 05.2019–07.2020 — 608; 08.2020–01.11.2022 — 255; 1,382 in total. “FDA does not know the specific connection between these diets and cases of non-hereditary DCM”; “FDA has no definitive information indicating that the diets are inherently unsafe”; legumes have long been used in pet foods, “with no evidence to indicate they are inherently dangerous”; grain-free foods usually contain more legumes than grain-inclusive ones. Since December 2022 FDA has published no updates pending “meaningful new scientific information”. We give no link for the same reason.
  3. Olivry T, Mueller RS. Critically appraised topic on adverse food reactions of companion animals (3): prevalence of cutaneous adverse food reactions in dogs and cats. BMC Veterinary Research, 2017;13:51 (systematic review, 22 papers on dogs). — Food allergy is found in 1–2 % of dogs brought to a vet for any reason, and in 9–40 % of dogs with itching. Europe PMC
  4. 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, 2016;12:9. — 297 dogs with food allergy confirmed by reintroduction of the food: beef 102 (34 %), dairy 51 (17 %), chicken 45 (15 %), wheat 38 (13 %), soy 18 (6 %), lamb 14 (5 %), corn 13 (4 %), egg 11 (4 %), pork 7 (2 %), fish and rice 5 each (2 %); barley a single case. 309 cases in 297 dogs in total: some reacted to more than one source. Europe PMC
  5. FEDIAF Nutritional Guidelines, September 2025, clause 7.6.4. — Any product with intact protein can potentially trigger a reaction; it is not always clear whether the high frequency of individual allergens is simply explained by dogs eating those proteins most often. FEDIAF 2025
  6. FEDIAF Code of Good Labelling Practice for Pet Food, 2019, clause 5.2.2.2 and the Annex (ingredient categories). — A “free from Y” claim means the absence of even traces of Y; such claims must not give the impression that products containing that substance are unsafe (Regulation 767/2009, Art. 11.1 b). In labelling terms legumes belong to the “Vegetables” category (“all types of vegetables and legumes”) and cereals to “Cereals”. FEDIAF Labelling Code 2019
  7. Carciofi AC, Takakura FS, de-Oliveira LD et al. Effects of six carbohydrate sources on dog diet digestibility and post-prandial glucose and insulin response. Journal of Animal Physiology and Animal Nutrition, 2008;92:326–336 (abstract). — 36 dogs, six extruded diets (cassava, broken rice, corn, sorghum, peas, lentils): starch digestibility over 98 % in all; dry matter, organic matter and energy digestibility highest for rice and cassava, then corn and sorghum, lowest for peas and lentils. PMID 18477314
  8. Arendt M, Cairns KM, Ballard JWO, Savolainen P, Axelsson E. Diet adaptation in dog reflects spread of prehistoric agriculture. Heredity, 2016;117:301–306. — 392 dogs: median 10 diploid copies of AMY2B (range 1–22); 51 wolves: median 2 (2–3), that is 5 times fewer. Each extra copy gives about +5.4 % of serum amylase activity (per Axelsson 2013 and Arendt 2014). Few copies: Siberian Husky (median 3), Greenland Dog (2.5), dingo (2); Alaskan Malamute 9.5. Europe PMC
  9. Axelsson E, Ratnakumar A, Arendt ML et al. The genomic signature of dog domestication reveals adaptation to a starch-rich diet. Nature, 2013;495:360–364 (abstract). — Dog domestication was accompanied by selection for starch digestion genes. PMID 23354050
  10. Mornard L et al. Diet-associated dilated cardiomyopathy in dogs. Veterinary Sciences, 2025;12:1106 (narrative review). — The link between diet and dilated cardiomyopathy is clearest for legumes, peas above all; the mechanism is unclear. Europe PMC
  11. Morris EM et al. Effects of grain-free and grain-inclusive diets with and without pulses on cardiac biomarkers in healthy adult dogs. Journal of Animal Science, 2025;103:skaf225. — 60 healthy dogs, 18 months, four diets (grain-free with peas and potato, grain-inclusive with peas, grain-inclusive without peas or potato, grain-free with potato): no clinically meaningful differences in cardiac biomarkers, taurine within range. All authors are employees of Hill's Pet Nutrition and the study was funded by Hill's. Europe PMC
  12. Smith CE, Parnell LD, Lai CQ, Rush JE, Freeman LM. Investigation of diets associated with dilated cardiomyopathy in dogs using foodomics analysis. Scientific Reports, 2021;11:15881. — Metabolomics of 9 foods with three or more legumes or potato among the main ingredients against 9 others: 88 compounds higher, with peas the most strongly associated ingredient. Partly funded by Nestlé Purina. Europe PMC
  13. DELICATE in-house data: composition by the label (website, /kuryatyna/ and /telyatyna/ pages, 2026-09-12) — meat and animal by-products at least 60 % of the mix, rice, corn, animal fat, hydrolysed liver, flax, beet pulp, yeast, vitamins and minerals, DL-methionine, L-lysine, glucosamine, chondroitin; no wheat, peas, lentils or potato. Crude protein 32 %, energy 397 kcal per 100 g — metabolisable energy, DELICATE calculation from the label composition (modified Atwater formula, NRC 2006); no batch laboratory analysis yet. Price of a 10 kg bag UAH 1,188 (website, 2026-09-12). Daily amount = 70 × kg^0.75 × factor 1.3 / 1.6 / 1.9 — the method of the site calculator; for an 18 kg Border Collie at a factor of 1.6 that is 979 kcal, 246 g of food, UAH 29 per day and 40 days per bag. The “per 1,000 dogs seen by a vet” recalculation and the FDA reports per month were computed by script on 2026-09-30, and the portion was recalculated to 397 kcal on 2026-11-09.
Free delivery to your door

10 kg DELICATE — UAH 1,188

Courier delivery to your home across Ukraine. We'll call you back within 15 minutes.

Order with free deliveryOrder by phone