Key Depression and Anxiety Symptoms in Hospice Patients: A Network Analysis Combined with Latent Profile Analysis

Background/Purpose: Depression and anxiety are highly prevalent among hospice patients, significantly impairing their quality of life during end-of-life care. In China, where hospice services are rapidly developing amidst cultural stigmas surrounding death and mental health, understanding the interplay of psychological symptoms is critical. Traditional approaches often treat depression and anxiety as distinct disorders, yet emerging network theory emphasizes their interconnectedness, where symptoms dynamically influence one another. A nuanced exploration of symptom comorbidity, particularly central and bridging symptoms that drive distress or connect depression and anxiety clusters, could inform targeted interventions. However, research on symptom networks in Chinese hospice populations remains scarce. This study addresses this gap by investigating depression and anxiety symptom networks and comparing their structures across subgroups with distinct symptom profiles. Findings aim to guide precision-based care strategies for this vulnerable population.

Methods: A cross-sectional study was conducted with 388 Chinese hospice patients (mean age = 73.62). Depression and anxiety symptoms were assessed using the Patient Health Questionnaire-9 and the 7-item Generalized Anxiety Disorder Scale, respectively. Psychometric network analysis was employed to model symptom interactions, with centrality indices (e.g., strength and expected influence) identifying pivotal symptoms. Bridge strength metrics were calculated to pinpoint symptoms linking depression and anxiety. Latent profile analysis (LPA) categorized participants into subgroups based on symptom severity patterns. Network comparison tests evaluated structural differences between subgroups. Analyses were conducted in R using “qgraph” for network estimation, “bootnet” for stability checks, and “mclust” for LPA.

Results: The network analysis revealed hopelessness and anhedonia as the most central depression symptoms, while excessive worry and nervousness dominated in anxiety. Bridging symptoms linking depression and anxiety included hopelessness, nervousness, and irritability. LPA identified two distinct profiles: a “mild-symptom” group (46.13%) with lower severity and a “severe-symptom” group (53.87%) exhibiting elevated scores across all symptoms. Networks differed significantly between profiles (global strength: S = 0.683, p = 0.032) while no difference existed in network invariance. In the mild-symptom network, central symptoms included hopelessness, excessive worry, reduced concentration, and anhedonia, with excessive worry, insomnia, and hopelessness acting as bridges. The severe-symptom network showed stronger connections overall, with hopelessness, difficulty relaxing, reduced concentration, and nervousness as central symptoms, while hopelessness, irritability, and nervousness served as bridges.

Conclusions and Implications: This study highlights hopelessness as a critical intervention target due to its dual role as a central and bridging symptom in the networks. Addressing hopelessness could disrupt the overall symptom burden and mitigate comorbidity. Secondary targets include anhedonia, excessive worry, and nervousness, which amplify distress within their respective disorders. The distinct network structures across symptom profiles underscore the need for tailored approaches: mild-symptom patients may benefit from early interventions targeting cognitive symptoms, while severe-symptom patients require intensive, integrated care for emotional and somatic symptoms. Social workers can use these insights into recognize and address central and bridging symptoms to reduce the psychological burden at the end of life and prioritize symptoms that maximally influence patients’ psychological well-being.