Multicenter development and validation of a predictive model for macrolide treatment nonresponse in school-aged children with Mycoplasma pneumoniae pneumonia

Background Mycoplasma pneumoniae pneumonia (MPP) is a common condition among school-aged children, with the potential to develop into macrolide-unresponsive Mycoplasma pneumoniae pneumonia (MUMPP). Objective To develop and validate a predictive model for the early identification of MUMPP in school-aged children. Methods A retrospective dual-center study involved children aged 6–14 years diagnosed with MPP and admitted between March 2022 and April 2025. Participants were randomly allocated into derivation and validation cohorts in a 7:3 ratio. The primary outcome was as MUMPP.Predictors were selected via Least Absolute Shrinkage and Selection Operator regression (LASSO), followed by multivariate logistic regression to identify independent predictors and construct the model, which was presented as a nomogram. The model performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, a calibration curve, and decision curve analysis(DCA).Differences in AUCs between the predictive model and individual predictors were evaluated using the DeLong test. Results A total of 323 patients were enrolled, including 130 (40.2%) macrolide non-responders. Multivariable analysis identified duration of fever before admission, LDH, and CT score as independent predictors (all P  < 0.05). The model showed excellent discrimination with AUCs of 0.834 (95% CI 0.779–0.890) in the derivation cohort and 0.860 (95% CI 0.784–0.936) in the validation cohort. The calibration curve showed satisfactory agreement, and DCA indicated strong clinical applicability across threshold probabilities. DeLong test showed superior AUC values compared to individual biomarkers (all P  < 0.05). Conclusion We developed and validated a predictive model for MUMPP in school-aged children, providing a practical tool for early risk stratification and guiding timely therapeutic adjustment.