| Objective To analyze and identify potential biological markers for the onset of major depressive disorder（MDD） in adolescents, and further construct a risk prediction model, aiming to identify high-risk factors for the development of MDD in adolescents and biological markers for preventing the occurrence of the disease. Methods A total of 360 adolescent patients with（MDD） from Tangshan Central Hospital from October 2023 to June 2024 were selected as the study group, and 360 healthy adolescents during the same period were selected as the healthy control group. General information, mental health status, and serum markers were collected from both groups. After univariate screening, a logistic regression model was constructed to analyze the independent influencing factors of MDD onset in adolescents. Based on this, a nomogram prediction model was developed. The receiver operating characteristic（ROC） curve was used to visually evaluate the discriminatory power of the prediction model, the calibration curve was used to evaluate the calibration of the model, and the decision curve analysis was used to assess the clinical application value of the prediction model. Results The research group and the healthy control group eliminated 8 and 12 invalid questionnaires, respectively, with effective recovery rates of 97.78% and 96.67%. The 24-item Hamilton Depression Scale（HAMD） scores for 352 adolescent MDD patients were（27.96±4.00） points, with 114 cases（32.29%） classified as mild, 159 cases（45.17%） as moderate, and 79 cases（22.44%） as severe. Significant differences were observed in the levels of serum brain-derived neurotrophic factor（BDNF）, neuropeptide Y（NPY）, cortisol（CORT）, and C-reactive protein（CRP） among patients with different levels of illness severity（P＜0.05）. Pearson correlation analysis showed that HAMD scores were negatively correlated with serum NPY and BDNF levels（r=-0.568,-0.812, both P＜0.05）, and positively correlated with serum CORT and CRP levels（r=0.489, 0.675, both P＜0.05）. Univariate and logistic multivariate regression analyses revealed that factors such as the Adolescent Self-Rating Life Events Checklist score, academic performance, parental divorce/widowhood, interpersonal relationships, Childhood Trauma Questionnaire（CTQ） score, Family Assessment Device score, serum NPY, CORT, BDNF, and CRP levels were independent factors influencing the onset of adolescent MDD（P＜0.05）. Based on the results of logistic multivariate regression analysis, a nomogram prediction model for the onset of adolescent MDD was constructed. The receiver operating characteristic（ROC） curve indicated that the area under the curve（AUC） of this model for predicting the onset of adolescent MDD was 0.863（95% confidence interval: 0.836～0.890）. Evaluation and validation results showed that the model’s predictive risk ability index（Cindex） was 0.846, indicating good predictive ability and clinical validity, as well as good precision discrimination. Conclusion The levels of serum NPY, BDNF, CORT, and CRP are closely related to the occurrence and severity of MDD in adolescents. By combining various factors such as CTQ scores and interpersonal relationships, a nomogram prediction model is constructed, which has good predictive value for MDD in adolescents. This model can be clinically applied to predict the risk of MDD occurrence in adolescents, guiding clinical intervention. |
