Introduction: Accurate assessment of adolescent depression requires instruments that are both developmentally sensitive and clinically robust. While many tools exist, there is a continuous need for scales that integrate ICD-11 criteria with language suitable for the adolescent population.
Purpose: This study describes the development and validation of a new 23-item depression scale, evaluating its factor structure, reliability, and demographic nuances.
Method: Item generation followed a rigorous process: 16 core symptom domains based on ICD-11 were identified, followed by expert evaluation (n=6) and a pilot study. After initial testing, one item was removed due to low factor loading and network centrality. The final 23-item version was validated using a sample of 688 adolescents (627 community, 61 clinical; ages 13-18). The clinical group consisted of adolescents with a confirmed diagnosis of depression or a clinical psychologist's assessment.
Results: EFA supported a four-factor structure: F1: Emotional-Cognitive; F2: Suicidal Ideation; F3: Interpersonal Distress; and F4: Somatic Symptoms. While the 4-factor model was consistent for girls, a simpler 2-factor model emerged for boys. The scale demonstrated excellent reliability (omega = 0.96). Significant demographic trends were observed: scores progressively increased with age, and girls reported significantly higher average scores compared to boys. Criterion validity was robust, with the clinical group (M=106.0) scoring significantly higher than the community sample (M=44.1); this was confirmed by a Mann-Whitney U test (U = 5536, p < .001, r rb = 0.711). A strong negative correlation with the WHO-5 Well-being Index was also observed (r = -0.51).
Conclusions: The 23-item scale is a psychometrically sound tool for assessing adolescent depression. It effectively differentiates clinical cases from non-clinical populations, with 19.9% of the community sample exceeding the optimal Youden-indexed threshold. The findings highlight the scale's sensitivity to developmental and gender-based symptom presentations.