The study included 132 hospitalized children with lobar pneumonia caused by Mycoplasma pneumoniae, of which 44 had plastic bronchitis (PB) and 88 did not. The aim was to identify independent risk factors for PB and develop a predictive model for early diagnosis. Independent risk factors were platelet-lymphocyte ratio (PLR), systemic immune inflammation index (SII), D-dimer and treatment with macrolide antibiotics. Continued treatment with macrolides after two cycles did not significantly reduce the incidence of PB. The combined model of PLR, SII and D-dimer had a higher predictive value than the individual factors, which was confirmed by the ROC curves. This triad model allows a simple clinical calculation to predict PB. The model is more sensitive and specific in predicting PB in children and provides guidance for the length of macrolide therapy.