Association between sleep quality trajectories and depression among college students: computer-simulated network analysis

Sleep quality is a well-documented risk factor for depression, yet existing evidence predominantly relies on static assessments. The dynamic trajectories of sleep quality and distinct depressive symptom networks across trajectory subgroups remain poorly understood in college students. This prospective cohort study recruited undergraduates from multiple universities in Ningxia, China, with follow-ups every 6 months. Sleep quality and depressive symptoms were assessed using the PSQI and PHQ-9, respectively, and depression was defined as a PHQ-9 score ≥ 10. A group-based trajectory model was applied to identify sleep trajectories. modified Poisson regression was employed to analyze the association between different sleep quality trajectories and depression risk. The Ising network analysis and computer-simulated network intervention analysis were performed to identify core symptoms and potential intervention targets. Two distinct sleep quality trajectories were identified. The persistently poor trajectory (24.54%) and the gradually improving trajectory (75.46%). The former exhibited a significantly higher incidence of probable depression compared to the latter ( RR = 5.31, 95% CI [3.51–8.03], P < 0.001). In both sleep trajectory groups, PHQ-4 (Fatigue) showed the highest strength centrality, indicating that it was the most structurally central symptom within the depression network. However, the NodeIdentifyR algorithm yielded different intervention-sensitive targets across groups. In the persistently poor group, PHQ-6 (Guilt) showed the largest predictive effect. However, in the gradually improving group, PHQ-3 (Sleep disturbance) produced the largest change in the projected total symptom burden. Sleep quality trajectories are heterogeneous and strongly associated with probable depression risk in college students. Symptom-specific intervention targets vary across sleep trajectories, supporting the need for tailored mental health prevention strategies.