BACKGROUND Provocation-tests with contrast media are not standardized, and our main objective was to evaluate a rapid provocation-test, including patients with a history of anaphylaxis. As secondary objectives, we phenotyping our population and proposed a predictive methodology for the outcomes of allergy tests. METHODS An allergy study using iohexol, iodixanol, ioversol, and iobitridol was conducted in patients over 18 years of age with previous hypersensitivity reactions to iodinated contrast media. A rapid provocation-test (100 cc administered in 12 minutes) was performed using a non-involved iodinated contrast medium that had tested negative on skin tests. A statistical analysis was carried out, including binary logistic regression and cluster analysis. RESULTS A total of 130 patients were enrolled. Ninety-six patients (74%) showed cutaneous symptoms exclusively, while 17 patients (13%) experienced anaphylaxis. Nine patients (7%) had positive skin-tests, and 20 of 141 provocation-tests performed were positive, all exhibiting mild cutaneous symptoms, including in those with a history of anaphylaxis. A safe alternative contrast medium was recommended to 122 patients (94%), with good tolerance in 50 patients who required a new radiological examination. We identified three patient phenotypes, each associated with a different risk of a positive drug provocation-test. A predictive model for the outcomes of allergy tests was obtained, but it exhibited a low predictive capacity. CONCLUSIONS We confirm the efficacy and safety of a protocol including rapid provocation-tests in patients with hypersensitivity reactions to iodinated contrast media of varying severity. Three patient clusters were identified, each showing a different risk level for a positive provocation-test.