Objective This work was to establish an early risk prediction model for non-suicidal self-injury (NSSI) among undergraduate students.Methods Firstly,the distribution characteristics of NSSI were obtained from the current situation survey.Then,the risk factors were screened based on logistic regression.Finally risk score method was used to establish non-suicidal self-injury early risk prediction model.The results of the model were evaluated by the area under the ROC curve and the positive predictive value.Results Among the 3854 college students,393 (10.2％) were diagnosed as probable NSSI-positive individual.Logistic regression analysis showed "childhood maltreatment" (OR =3.493)," parents punished meeven for minor negligence " (OR =1.995),"nagging what I did in front of others made me feel embarrassed" (OR =1.835),"non-biological mother"(dummy variables,P=0.009),"beat and scolded by my parents" (OR=l.651),"be disgraced"(OR=1.602),"often impulse to want to throw things" (OR =1.601),"failure in exams" (OR =1.553),"often feeling nervous"(OR =1.527) were the main risk factors of non-suicidal self-injury among undergraduates.The ROC curve of the subjects was calculated by the risk score model,and the area under the curve was 0.738,95％CI (0.707,0.770).Among 70％ of the samples,the sensitivity was 72.3％,the specificity was 64.8％ and the positive predictive value was 19.0％ if the cut point was set 22.8.Conclusion The early risk prediction model based on risk score can be used to predict NSSI among undergraduates.
NETL
NSTL
维普
万方数据
