A recently developed confirmatory LC-MS method has been applied to the quantification of five major beta-lactam antibiotics in suspect raw bovine milk samples that gave a positive response with receptor-based (BetaStar) and rapid microbial inhibitory screen tests (Delvotest SP). In total, 18 presumptive positive raw milk samples were reanalyzed; 16 samples showed traces of antibiotic residues that could be identified and quantified by the LC-MS method, ranging from the limits of confirmation up to 38 microg/kg. Of the positive samples, only five (approximately 30%) were found to be violative of EU maximum residue limits. The most frequently detected antibiotic residues were cloxacillin and penicillin G, the former often in combination with amoxicillin or ampicillin. This study compares the results obtained by the three methods on identical samples and addresses how these relate to certain criteria such as sensitivity and selectivity. Furthermore, the limitations of the LC-MS method and the potential impact of the presence of frequently more than one residue in the same milk sample on the response of the rapid test methods are discussed.