Published online Mar 19, 2024. doi: 10.5498/wjp.v14.i3.388
Peer-review started: December 12, 2023
First decision: January 2, 2024
Revised: January 12, 2023
Accepted: February 27, 2024
Article in press: February 27, 2024
Published online: March 19, 2024
Processing time: 98 Days and 2.3 Hours
Depression is one of the most severe diseases affecting the mental health of adolescents. Most adolescents with depression have suicidal ideation (SI). However, few studies have focused on the factors related to SI, and there is a lack of effective predictive models.
This study determined the factors influencing SI in adolescent patients with depression and construct a risk prediction model to provide a theoretical basis for prevention and intervention.
This study aimed to construct a risk prediction model for SI in adolescents with depression and provide an assessment tool for early screening.
Based on a retrospective analysis of social factors and laboratory indicators of 150 adolescent patients with depression and SI, this study constructed and internally validated a risk prediction model.
Studies have shown that trauma history, predisposing factors, and serum ferritin levels (SF), high-sensitivity C-reactive protein levels (hs-CRP), and high-density lipoprotein (HDL-C) levels influence SI in adolescents with depression. The AUC of the nomogram prediction model was 0.831 (95%CI: 0.763–0.899), the sensitivity was 0.912, and the specificity was 0.678. The high net benefit of the DCA and the average absolute error of the calibration curve were 0.043, indicating that the model had a good fit.
The nomogram model based on trauma history, predisposing factors, SF, hs-CRP levels, and HDL-C levels can effectively predict the occurrence of SI in adolescents with depression, which can help in implementing early clinical measures to reduce suicide mortality in adolescents with depression.
According to the general data and laboratory indicators of adolescents with depression, we identified risk factors for SI and used them to develop an effective predictive model for quick detection.