Distinguishing Chronic and Non-Chronic Depression: A Clinical Profile and Symptom Networks Approach.

INTRODUCTION: Different disorder courses contribute to the large heterogeneity within depressive syndromes. Chronic depression (CD) warrants further investigation given its high prevalence and poor treatment outcomes. This study aimed to identify a clinical profile and symptomatic phenotype associated with CD versus non-chronic depression (NCD) to provide a better characterization of CD.

METHODS: The preregistered analysis used cross-sectional data from a large German cohort (N = 994; 64.9% female) with retrospective information on depression trajectories. We assessed associations of nine psychological and social characteristics with CD in logistic regression models. We further examined symptom networks across disorder courses using Bayesian network analysis of Beck Depression Inventory (BDI-I) item-level data.

RESULTS: In our sample 18.5% of patients with depression met criteria for CD (nCD = 184), whereas 81.5% showed a non-chronic course (nNCD = 810). The characteristics childhood maltreatment, age of onset, neuroticism, extraversion, social networks, social support, psychosocial functioning, and psychiatric comorbidities were associated with CD in single regressions. In a combined model (R2 = 0.25), greater exposure to childhood maltreatment, lower extraversion, and lower psychosocial functioning were associated with CD, defining a clinical profile distinguishing CD from NCD. Symptom networks based on BDI-I items were highly similar across trajectories, differing only by the connection between past failure and self-contempt, which was unique to the CD group.

CONCLUSION: Psychosocial factors, but not symptom profiles, distinguished CD from NCD. Thus, characteristics beyond classical depression symptoms seem more informative for differentiating depression trajectories. If confirmed longitudinally, these findings may improve prediction of depression trajectories and inform early intervention.