Objective To construction and validation of a predictive model for post-traumatic stress disorder(PTSD)in patients after motor vehicle collisions(MVC).Methods A convenience sampling method was used to select 279 MVC patients who visited the emergency department of Beijing Chaoyang Hospital,Capital Medical University,from April 2020 to December 2022.Patients were categorized into a PTSD group(n=96)and a non-PTSD group(n=183)based on whether they developed PTSD three months after the MVC.General data of both groups were analyzed,and binary Logistic regression analysis was conducted to identify influencing factors for PTSD onset in MVC patients.Based on these factors,a multi-indicator regression predictive/assessment model was developed,with its diagnostic value assessed using the area under the receiver operating characteristic curve(AUC).An additional 67 MVC patients admitted to the Emergency department from January to March 2023 were used to validate the model.Results Binary Logistic regression analysis identified high education level,poor family relationships,facial injuries,history of anxiety,history of depression,witnessing death during the accident,feelings of guilt,and alcohol use as significant risk factors for PTSD(P＜0.05).The 8-factor LogP model developed based on binary Logistic regression analysis demonstrated strong predictive performance,with an AUC of 0.898[95％CI(0.808,0.974)].Validation with 67 additional MVC patients showed comparable sensitivity,specificity,and accuracy(P＞0.5),indicating equivalence between the model and the validation results.Non-inferiority test yielded P=0.010,and equivalence test yielded P=0.020,confirming that the model and validation results are equivalent.Correlation tests between validation results and true samples revealed P＜0.01.Superiority test P=0.182.Conclusions The predictive model demonstrates high predictive value and is convenient for clinical use,offering valuable guidance for clinical nursing practice.
维普
万方数据