The diagnosis of cow's milk allergy (CMA) is particularly challenging in infants, especially with non-Immunoglobulin E (IgE)-mediated manifestations, and inaccurate diagnosis may lead to unnecessary dietary restrictions. The aim of this study was to assess the accuracy of the cow's milk-related symptom score (CoMiSS ) in response to a cow's milk-free diet (CMFD). We prospectively recruited 47 infants (median age three months) who had been placed on a CMFD due to persisting unexplained gastrointestinal symptoms. We compared data with 94 healthy controls (median age three months). The CoMiSS score was completed at recruitment and while on the exclusion diet. In 19/47 (40%) cases a response to the diet occurred. At recruitment CoMiSS was significantly higher in cases compared to controls (median score 8 vs. 3; -value: <0.05), 9 cases had a score ≥12 and 8/9 normalized on CMFD. An oral milk challenge was performed in all 19 responders and six of these had a positive reaction to cow's milk (CM). In eight infants IgE allergy tests were positive. The receiver operation characteristic (ROC) curve identified a CoMISS score of 9 to be the best cut-off value (84% sensitivity, 85% specificity, 80% positive (PPV) and 88% negative predictive value (NPV)) for the response to CMFD. We found CoMiSS to be a useful tool to help identify infants with persisting gastrointestinal symptoms and suspected CMA that would benefit from CMFD.