The purpose of this retrospective study was to compare home-prepared and chicken hydrolysate diets in the diagnosis of canine adverse food reactions (AFR). Seventy-two dogs were fed home-prepared diets and 109 were fed hydrolysate. Owners chose the type of diet at presentation, and ingredients of home-prepared diets were selected depending on each dog's dietary history. Ectoparasitic infestations and microbial infections were treated during the trials. Cutaneous and gastrointestinal signs and pruritus scores were recorded before starting the diet, 6 weeks into the trials and after provocation with the original diets. AFR was diagnosed if pruritus resolved during the trial and recurred on dietary provocation. The dropout rate was lower for home-prepared diets although not statistically significant (18.1% home prepared; 24.7% hydrolysate, P=0.377). AFR alone was diagnosed in 10 dogs (17%) using home-prepared diets and in 15 (18.3%) fed the hydrolysate. Gastrointestinal problems were more frequent in dogs with AFR than in dogs without AFR (P=0.001). Another 11 dogs (18.6%) in the home-prepared diet group and 20 (24.4%) in the hydrolysate diet group had AFR concurrent with other pruritic diseases, mainly atopy. The similar frequencies of AFR diagnosis in the two groups (P=0.837 AFR; P=0.416 concurrent AFR) indicate that the chicken hydrolysate diet may be a valuable alternative to home-prepared diets in the diagnosis of canine AFR. Prospective cross-over studies are warranted to confirm these findings.