Analysis Of Influencing Factors And Establishment Of Risk Prediction Models Of Adolescent Depression Disorder

Objective:1.Investigate and analyze the influencing factors of depression disorder in adolescents,and provide data sources for the construction of the risk prediction model of depression disorder in adolescents.At the same time,it can provide a reference for for developing targeted interventions.2.The risk prediction model of adolescent depression disorder was constructed,and to verify and evaluate the model,to provide a basis for the further promotion and application of the model in clinical practice.3.The constructed risk prediction model of adolescent depression disorder was transformed into a wechat mini program for convenient application,to provide a simple and effective evaluation tool for the prediction and prevention of people at high risk of adolescent depression disorder,so as to achieve the primary prevention of the disease,reduce the burden of disease.Methods:1.A case-control study design was used,it is convenient to select adolescent depression patients in the outpatient and inpatient department of a Grade A hospital in Shanxi Province from March to June 2021(n=196)and healthy adolescents with matched school age(n=197)as the research subject of the modeling group.The modelling group divided adolescents into depression and control groups based on whether they developed depressive disorder or not.A large number of relevant literature on the influencing factors of adolescent depression disorder at home and abroad was searched,and the influencing factors related to adolescent depression disorder were comprehensively selected in combined with the existing research basis,expert opinion and clinical practice.The research tools were selected according to the 20 factors related to adolescent depression disorder summarized in the early stage,and the questionnaire was designed,systematic evaluation of adolescents on the six aspects of emotion,cognition,behavior,family,society,and special events.The general data and related variables of the depression group and the control group in the modeling group were collected and univariate analysis,Variables with statistically significant differences in the results(P<0.05)were included in the binary Logistic regression analysis to screen the independent influencing factors of adolescent depression disorder.2.Combined with the partial regression coefficientβand intercept of each independent influencing factor,the regression equation was constructed to construct the prediction model of adolescent depression disorder risk.Hosmer-Lemeshow test and ROC curve were used to test the goodness of fit and discriminative ability of the risk prediction model of adolescent depressive disorder,and the constructed model was evaluated.3.It is convenient to select adolescent depression patients in the outpatient and inpatient department of a Grade A hospital in Shanxi Province from July to August 2021(n=59)and healthy adolescents with matched school age(n=60)were used as the study subjects in the validation group.Adolescents were divided into depression and control groups based on whether they developed depressive disorder or not.The validation group data were substituted into the constructed risk prediction model,and the constructed risk prediction model of adolescent depression disorder was externally verified to examine the clinical utility and validity of the model.4.According to the constructed risk prediction model of adolescent depression disorder,We Chat mini program is developed,provide a simple and quick tool for the application of the model,it is beneficial to promote the application of the prediction model in the adolescent population.Results:1.Modeling group,univariate analysis results showed that depression group and control group,gender,family mental illness history,academic performance,weight index BMI,smoking,drinking,number of close friends,parents marital status,positive response scores,negative response scores,family function assessment scores,adolescent life event scores,childhood trauma questionnaire scores,social support scale scores,the difference was statistically significant(P<0.05).Binary Logistic regression results showed that gender,family history of mental illness,body mass index,number of close friends,parental marital status,coping style scores,family function assessment scores,and childhood trauma questionnaire scores were independent factors affecting adolescent depression disorder.2.The Prediction Model formula is P=1/{1+exp[-(1.368×gender-4.842×family history of mental illness+0.318×body mass index+3.478×number of close friends(1)+3.005×number of close friends(2)+1.257×number of close friends(3)+1.324×marital status of parents-0.32×positive coping Score+0.133×negative coping score+0.048×total score of family Functioning Rating Scale+0.073×total score of childhood trauma questionnaire-10.172)]}.model evaluation:Hosmer-Lemeshow test of modeling group shows that,x~2=5.538,P=0.699,model prediction accuracy is 92.1%;area under ROC curve is 0.972,maximum Youden index is 0.853,corresponding sensitivity is 92.9%,specificity is 92.4%and best cutoff value is 0.455.3.External verification:The data of the modeling group and the validation group were compared,and the difference was not statistically significant.the Hosmer-Lemeshow test of the validation group showed that,x~2=7.506,P=0.483,the model prediction accuracy is 86.6%;the area under the ROC curve is 0.935,when the cutoff value is 0.455,the sensitivity is 89.8%and the specificity is 83.3%.4.Based on the construction of risk prediction model developed a small program to predict the risk of depression,teenagers only need to fill in gender,family history of mental illness,body mass index,number of close friends,parents marital status,positive response scores,negative response scores,family function scale score,childhood trauma questionnaire score,click"test"can automatically calculate adolescent depression risk probability,when the probability value is higher than 45.5%,suggest teenagers may have the risk of depression.Conclusion:1.Gender,family history of mental illness,body mass index,number of close friends,parental marital status,coping scores,family function assessment score,and childhood trauma questionnaire score are independent influencing factors of adolescent depression,which can provide reference for psychological care of adolescent patients with depression disorder.At the same time,it can suggest that the above influencing factors of adolescent depression disorder need to attract high attention from parents,school,society and clinical medical staff.2.The constructed risk prediction model of adolescent depression disorder can effectively predict the occurrence of depression disorder in adolescents.The predictive model was externally validated,it has a good goodness of fit and discrimination degree,can be popularized in clinical practice,and can provide reference for early identification of high-risk groups and timely preventive intervention measures.3.The wechat mini program developed according to the risk prediction model of adolescent depression disorder is simple,convenient,fast and efficient,save the time and economic costs,it can accurately and efficiently calculate the risk probability of depression disorder in adolescents can be used to assess the risk of depression disorder in the adolescent group,suitable for the promotion and use in the adolescent population.