Establishment and validation of a predictive model for severe pneumonia in children.

Background: This study aimed to develop a model for the early identification of severe pneumonia in children by comparing common laboratory indicators between children with ordinary pneumonia and severe pneumonia. Methodology: Children aged 1 month to 14 years, diagnosed with pneumonia and admitted to our hospital between January 2017 and June 2022, were included in the study. Participants were divided into two groups based on the severity of their pneumonia. Data, including demographic information, medical history, clinical symptoms, laboratory indicators, and treatment outcomes, were collected from the hospital's medical records system. Results: The single-factor analysis revealed significant differences (P < 0.05) between the two groups in various parameters, including age, length of hospital stay, repeated hospitalization within 90 days, invasive ventilation, Intensive Care Unit (ICU) stay time, birth history, temperature, respiratory rate, blood pressure, procalcitonin, and C-reactive protein. Binary logistic regression analysis indicated that high body temperature and high respiratory rate were independent risk factors for severe pneumonia (P < 0.05). Conclusion: A predictive model for severe pneumonia in children can identify the risk of progression to severe disease, enabling prompt treatment and improving patient prognosis. This reduces the burden on families and social security.