Determinants of depressive symptoms at individual, school and province levels: a national survey of 398,520 Chinese children and adolescents

Objectives: The aims of this study were to describe the national distribution of depressive symptoms in Chinese children and adolescents, to examine the determinants of depressive symptoms at individual, school and province levels and to assess the gender and age differences in the effect of school factors on depressive symptoms.

Study design: This was a national cross-sectional study.

Methods: A school-based online survey was conducted in mainland China from between December 1, 2021, and January 1, 2022. A total of 398,520 eligible participants were included in the analysis. School-level data were drawn from students, headteachers and Baidu Maps, and province-level data were obtained from the national human development report. The Patient Health Questionnaire-2 was used to measure depressive symptoms.

Results: Areas with the highest mean scores for depressive symptoms were in the northeastern, inner central and southwestern regions of China. At the individual level, younger age, male sex, being an only child, Han ethnicity, lower body mass index, more days of exercise, less drinking and smoking behaviours, higher subjective family socio-economic status (SES) and popularity in school were related to fewer depressive symptoms; however, objective family SES and maternal education were not related to fewer depressive symptoms. The school-level variables of public status, psychological activities and psychological courses and province-level variable of higher Human Development Index were associated with fewer depressive symptoms. The effect of psychological courses and activities on depressive symptoms was greater in females.

Conclusions: The results showed multilevel factors related to depressive symptoms and emphasised the importance of implementing school-based psychological activities to ameliorate depressive symptoms in Chinese children and adolescents across age and gender.