关键词: 肺炎支原体肺炎, 儿童, 肺功能异常, 危险因素, 风险分层, 预测模型

Abstract: Objective To investigate the modifiable environmental factors and social determinants of post-Mycoplasma pneumoniae pneumonia (MPP) pulmonary dysfunction in preschool and school-aged children, and to develop a risk stratification model and corresponding screening strategy. Methods This prospective cohort study consecutively enrolled 1 594 pediatric patients diagnosed with MPP between February 2024 and February 2025. Participants were recruited from a tertiary pediatric specialty hospital, the pediatric department of a secondary general hospital, and two community health service centers. Data on clinical characteristics, modifiable environmental factors, and social determinants were systematically collected. The primary outcome was the presence of pulmonary function abnormalities at six months post-clinical resolution. Three nested logistic regression models were constructed to evaluate the incremental predictive value of the collected factors. A nomogram was subsequently developed for risk stratification. The public health utility of the model was assessed using decision curve analysis (DCA) and cost-effectiveness analysis. Results The overall incidence of post-MPP pulmonary function abnormality was 28.7%. Multivariate logistic regression analysis identified several independent risk factors: radiological involvement of ≥2 lobes, a febrile duration exceeding 7 days, absence of systemic corticosteroid therapy, significant secondhand smoke exposure, poor follow-up adherence, low parental educational attainment, and coverage by the New Rural Cooperative Medical Scheme (NRCMS). The full model (Model 3) demonstrated superior predictive performance, with an area under the curve (AUC) of 0.846, compared to Model 1 (clinical factors only; AUC=0.819) and Model 2 (clinical and environmental factors; AUC=0.835). The incidence of pulmonary dysfunction was 47.6% in the high-risk stratum versus 9.6% in the low-risk stratum. Compared with universal screening, the model-guided screening strategy yielded a cost saving of 260 RMB per case of abnormality detected. Furthermore, the cohort with pulmonary function abnormalities exhibited significantly lower quality of life scores, prolonged recovery periods, and a higher incidence of recurrent respiratory tract infections (all P<0.001). Conclusion The inclusion of modifiable environmental factors and social determinants significantly enhances the predictive efficacy for pulmonary dysfunction following MPP in children. The developed risk stratification model demonstrates robust discrimination and calibration. In conjunction with the cost-effectiveness analysis, these findings provide a solid evidence-based foundation for the formulation of regional pediatric respiratory health management strategies.

Key words: Mycoplasma pneumoniae pneumonia, Child, Pulmonary function abnormality, Risk factors, Risk stratification, Prediction model