Abstract
<jats:p>Road traffic accidents remain a leading cause of mortality and disability among children worldwide and represent a major public health concern. The anatomical and physiological characteristics of the paediatric body—including a relatively larger head-to-body ratio, incomplete development of the musculoskeletal system, and insufficient myelination of the central nervous system—render children particularly susceptible to mechanical injuries in autumobile collisions. This vulnerability underscores the need for effective preventive strategies and reliable risk assessment tools. Objective: To develop and validate a mathematical algorithm for predicting injury risks to child passengers in automobile collisions, accounting for age, anthropometric characteristics, and use of child restraint systems. Materials and Methods: A retrospective analysis was performed of 218 cases of child passenger injuries involving individuals aged 29 days to 18 years during the period 2014–2024. Data collected included age, sex, anthropometric parameters, type and correctness of child restraint system use, road traffic accident characteristics, and severity of traumatic injuries. The study was conducted in accordance with basic bioethical principles. Statistical methods applied were Spearman correlation analysis, multiple regression, discriminant analysis, and receiver operating characteristic (ROC) analysis. Results and Discussion: A statistical model with high predictive performance was developed (R² = 0.738, p < 0.001). Correct use of child restraint systems was found to reduce the risk of severe injuries by 71% (β = –0.508, p < 0.001). The area under the ROC curve was 0.912, with model sensitivity of 86.8%, specificity of 81.5%, and overall classification accuracy of 84.2%. The high validity metrics support recommendation of the model for practical application. Conclusion: The developed algorithm enables highly accurate assessment of individual risk of injury to child passengers and can be implemented in pediatric practice and forensic medical examination.</jats:p>