Comparing Depression and Anxiety Symptom Networks in Chinese Students and Teachers

Depression and anxiety often co-occur in schools, but few studies have compared their symptom networks in students and teachers. We analysed the same PHQ-9 and GAD-7 symptoms in 52,093 Chinese secondary-school students and 7222 teachers. Ordinal Gaussian graphical models were estimated from polychoric correlations, and 5000 group-label permutations directly tested group differences. Network structure differed (maximum edge difference = 0.142, p < 0.001), but global strength did not (teachers = 7.736, students = 7.634, p = 0.543). Nine of 120 edges remained significant after Holm correction. Across 16 expected-influence comparisons, sleep problems had a higher value in students; trouble concentrating and psychomotor symptoms had higher values in teachers. Across 16 bridge expected-influence comparisons, worthlessness, excessive worry, trouble relaxing and irritability had higher values in students. The observed network estimates were unchanged at γ values of 0.50, 0.25 and 0, and the same inferential pattern was reproduced in 20 matched-size analyses, each based on 7222 participants per group and the full 5000-permutation procedure. In conclusion, this study revealed selective student-teacher differences in how depressive and anxiety symptoms are interconnected and positioned within symptom networks, underscoring the value of considering population context when examining depression-anxiety symptom organisation across educational groups.