Objective To investigate the risk factors for the occurrence of bronchiolitis obliterans (BO) in children with refractory Mycoplasma pneumoniae pneumonia (RMPP) and their predictive value of the factors. Methods A retrospective analysis was performed for the medical records of 156 children with RMPP who were admitted to the hospital from May 2020 to March 2024. According to the diagnostic criteria for BO, they were divided into a BO group (n=76) and a non-BO group (n=80). A logistic regression analysis was used to investigate risk factors for the occurrence of BO, and the receiver operating characteristic (ROC) curve was used to assess the value of the model established based on the risk factors in predicting BO. Results Compared with the non-BO group, the BO group had a significantly longer duration of fever, a significantly higher leukocyte count, and a significantly lower albumin level (P<0.05). Compared with the non-BO group, the BO group had a significantly lower proportion of children with initiation of macrolide antibiotic therapy within 5 days, initiation of glucocorticoid therapy within 2 weeks, or initiation of bronchoscopic therapy within <2 weeks (P<0.05). The ROC curve analysis showed that the logistic regression model established based on the above six indicators had an area under the curve of 0.901 (95%CI: 0.849-0.953, P<0.001) in predicting the occurrence of BO, with a sensitivity of 0.893 and a specificity of 0.827 at the optimal cut-off value of 0.341. Conclusions The logistic regression model established based on duration of fever, leukocyte count, albumin, initiation of macrolide antibiotic therapy within 5 days, initiation of glucocorticoid therapy within 2 weeks, and initiation of bronchoscopic therapy within 2 weeks has relatively high sensitivity and specificity in predicting the occurrence of BO in children with RMPP.