Dermatology
RCT
● RCT
Study finds placebo effect complicates food allergy testing in dogs with skin condition
Veterinary dermatology
Published April 1, 2026
Sofou Evi I, Samuel Elisa, Aleksandrova Svetlina, Chatzis Manolis, Saridomichelakis Manolis N
A small study looked at how to identify food allergies in dogs with atopic dermatitis, a common itchy skin condition. Researchers tested 12 dogs by adding small amounts of eight common food items, like beef and chicken, to their special diet for one week each. They also gave the dogs a placebo, which was just their regular diet mixed with water, to see if reactions were real.
They found that most dogs (11 out of 12) reacted to between one and six of the food items. However, half of the dogs that reacted to real foods also reacted to the placebo. This means some of the 'allergic reactions' might have been due to the placebo effect, where a dog or its owner expects a reaction and sees one, rather than a true food allergy.
The study was very small and did not report specific numbers for how strong the reactions were. The main reason to be careful is that this placebo effect creates doubt about how accurate food allergy tests are, both in this study and in previous ones that didn't use a placebo. Readers should understand that identifying food allergies in pets can be tricky, and this research highlights why controlled testing is important, even though more work is needed.
View Original Abstract ↓
BACKGROUND: Identification of offending foods in dogs with adverse food reactions is usually based on "deterioration" during open food challenges.
OBJECTIVES: To examine the placebo effect during double-blinded, placebo-controlled food challenges using a predefined set of criteria for relapse.
ANIMALS: Twelve dogs with atopic dermatitis and adverse food reactions.
MATERIALS AND METHODS: Dogs were serially challenged with 40 g/day of eight food items (beef, chicken, codfish, corn flour, cow's milk, hen's egg, lamb, wheat), for 1 week, each mixed with their elimination diet and water. An additional two challenges were placebo (elimination diet mixed with water). Owners and investigators were blinded to the challenges and the order of the 10 challenges was randomised for each dog. Relapse was defined as moderate-to-severe owner global assessment of challenge deterioration and/or > 100% increase of Canine Atopic Dermatitis Extent and Severity Index, 4th iteration (CADESI-04) score (with the score at the end being > 9) and/or > 100% increase of pruritus using a Visual Analog Scale (with the score at the end being > 1.9/10).
RESULTS: Most (91.7%) dogs were positive to one to six challenges with food items, yet half of them also were falsely positive in one placebo challenge. Two dogs had only one placebo challenge. The number of positive challenges to foods did not differ between dogs with positive and negative placebo challenges.
CONCLUSIONS AND CLINICAL RELEVANCE: The placebo effect during double-blinded food challenges creates doubts about the accuracy of the results of challenges with food items in this and in previous studies where open food challenges were used.
