Purpose: To analyze as the primary endpoint the accuracy, sensitivity, and specificity of the SNOT-22 assessing CRS severity and to compare the results with a version of the SNOT-8 obtained from the nasal domain items.Methods: Data were obtained from a prospective multicenter controlled study of dupilumab in adults with moderate–severe CRSwNP. EQUATOR and STROBE network guidelines were adopted. A multivariate logistic regression model was used to evaluate the accuracy of the model with the full (SNOT-22) and reduced (SNOT-8) item set to predict the severity outcome.Results: SNOT-22 demonstrated an AUC of 0.885 (95% CI 0.825, − 0.945), and sensitivity and specificity of 91.49% (83.92–96.25%) and 69.23% (48.21–85.67%), respectively. Interestingly, after stepwise items elimination good outcomes were reported for SNOT-8, with an AUC of 0.818 (95% CI 0.744–0.892), achieving a sensitivity of 93.51% (85.49–97.86%) and specificity of 57.14% (40.96–72.28%).Conclusion: Psychometric analyses support the accuracy, sensitivity, and specificity of the nasal domains of SNOT-22 to assess the impact on HRQoL in patients with CRSwNP.