Longitudinal trajectories and networks of depression symptoms from adolescence to adulthood

Background: Depression is a complex and heterogeneous condition, in which individuals follow distinct symptom trajectories and experience fluctuations over time. This study integrates longitudinal trajectory modelling with symptom network analysis to examine symptom-level heterogeneity both between and within individuals from adolescence to early adulthood. Methods: We applied latent class growth analysis to identify distinct depressive symptom trajectories in the Avon Longitudinal Study of Parents and Children, a population birth cohort (N=5,730; age range 11-28 years). For each trajectory (persistent, adolescent-onset, adolescent-limited and stable low), we estimated symptom network models at four significant developmental milestones (ages 11, 17, 23, and 28). We computed four metrics capturing key network properties: symptom influence, structural similarity, global connectivity, and network stability. Results: Persistent and adolescent-onset trajectories showed higher symptom scores across domains, whereas the adolescent-limited trajectory was driven by a smaller subset of symptoms (distracted, unhappy, apathetic, restless). Cognitive-affective and social-withdrawal symptoms distinguished chronic from transient courses, while early somatic and mood symptoms were common across all trajectories. Symptom networks in the persistent and adolescent-onset trajectories strengthened and stabilised across adolescence, indicating intensified symptom interdependence, whereas the adolescent-limited trajectory showed declining symptom influence and connectivity by emerging adulthood, suggesting weakening symptom relations potentially linked to developmental transitions. Conclusions: Depression heterogeneity extends beyond severity, with distinct symptom profiles and dynamics characterising developmental trajectories. These findings highlight key developmental windows and specific symptoms within trajectories that could inform targeted interventions to support recovery.