Comorbidity of anxiety and depression disorder among clinical referral patients: a longitudinal study based on network analysis
Sensen Zhang,Fengqin Ding,Jiatai Chen,Zhang, Sensen,Chen, Jiatai
DOI: https://doi.org/10.1007/s12144-024-05856-2
IF: 2.8
2024-04-02
Current Psychology
Abstract:The COVID-19 pandemic has exacerbated anxiety and depression disorders. We used network analysis to identify core symptoms and comorbid pathways between the two disorders and reveal temporal changes in symptoms that traditional assessment tools fail to capture. Data was collected from 533 clinically comorbid patients ( Mean age = 43.4; female = 363) who completed the Hamilton Anxiety Scale and Hamilton Depression Scale before and after treatment (T1→ T2). Node and bridge strengths were calculated, and analyses included a network comparison test (NCT) and cross-lagged network analyses (CLPN) to examine the interconnectedness and changing features of anxiety and depression comorbidity. Results indicate that (a) In contemporaneous networks before treatment, the most central nodes were tension , somatization , anxiety ( somatic ), depressed mood , psychomotor retardation , somatic symptoms ( GI ), autonomic symptoms , and hopelessness . Additionally, depressed mood , psychomotor retardation , somatic symptoms ( GI ), and hopelessness were also bridge nodes associated with psychological and physical symptoms of anxiety. (b) Although the total score of the assessment tools decreased during treatment and NCT confirmed overall network strength remained stable during treatment ( p = 0.38), specific symptoms changed significantly ( p < 0.05 ), underscoring the importance of nuanced evaluations beyond total scores. Specifically, nodes such as fear , insomnia , intellectual , suicide , psychomotor agitation , anxiety ( psychic ), loss of weight , and helplessness exhibited a significant reduction ( p < 0.05). In contrast, nodes including genitourinary symptoms , insomnia ( late ), work and interests , and somatic symptoms ( general ) demonstrated a significant enhancement ( p < 0.05). (c) Genital symptoms , hypochondriasis , and paranoid symptoms positively predicted other symptoms, while insight negatively predicted tension in longitudinal networks. In summary, our results contribute additional perspectives on the mechanisms underlying comorbid symptomatology and its dynamics throughout the treatment process, identifying meaningful targets for intervention. Thus, recommending the integration of network analysis into current diagnostic, treatment, and follow-up procedures promotes individualized interventions and improves patient recovery.
psychology, multidisciplinary