Core symptoms in a network of depressive and sleep symptoms in bipolar disorder based on Gaussian graphical models and directed acyclic graphs

ObjectiveTo construct a network of depressive and sleep symptoms in patients with bipolar disorder and to explore core symptoms and potential causal relationships. MethodsA total of 212 inpatients with bipolar disorder hospitalized at the Fifth People’s Hospital of Nanning from January 2022 to December 2024 were enrolled. Depressive symptoms were assessed using the depression dimension of the Symptom Checklist-90 (SCL-90), and sleep status was evaluated using the Pittsburgh Sleep Quality Index (PSQI). A Gaussian graphical model was used to analyze correlations between depressive symptoms and sleep symptoms, and a directed acyclic graph (DAG) was constructed to analyze potential causal relationships. ResultsAmong the 212 patients with bipolar disorder, the median (IQR) score for the SCL-90 depression factor was 21.50 (14.00, 38.75), and the median (IQR) PSQI score was 9.00 (5.00, 14.00). Partial correlation network analysis showed a positive correlation between depressive symptoms and sleep symptoms. Centrality analysis identified three core symptoms: feeling depressed (S30), excessive worry (S31), and feeling worthless (S79). The directed acyclic graph (DAG) indicated a potential influence of depressive symptoms on sleep symptoms, with feeling worthless (S79) serving as a key mediating node that could exert direct or indirect effects on other symptoms. ConclusionsFeeling worthless (S79) occupies a central position in the symptom network and may readily trigger other symptoms. Clinical treatments and interventions should prioritize this symptom to reduce its negative impact on patients’ overall health status.