PTSD CHARACTERISTICS OF RESCUERS BASED ON DECISION TREE

Post-traumatic stress disorder (PTSD) is a mental disorder caused by traumatic events, characterized by recurrence of memories and nightmares, avoidance of stimulation, difficulty in regulating emotions, and persistent hypervigilance. Due to the particularity of their work, rescuers are often more likely to suffer from PTSD. Identifying and clarifying the influencing factors of PTSD, establishing an effective prediction model for rescuers, and providing effective evaluation tools to provide targeted treatment and support measures for rescuers are of great significance in helping rescuers cope with and prevent PTSD. Based on the data set provided by the PLA Medical College, this study used a decision tree classification model to construct an effective prediction model for rescuers, identify important variables that affect rescuers' PTSD, and evaluate the prediction efficiency of the model by the receiver operating characteristic curve (ROC). Results: The accuracy was 95.56%, the sensitivity was 93.75%, the specificity was 95.69%, the false positive rate was 4.31%, the F1 score was 75%, and the AUC value was 94.72%. The features classified by the model were ranked according to their importance. The top eight features of the decision tree model were: ASD alertness, ASD avoidance, ASD re-experience, ASD separation, ASD nature, smoking status, psychological resilience, and age. Conclusions: The decision tree model has high accuracy and stability and can be used to guide clinical prevention and treatment.