Symptom networks in glioma patients: understanding the multidimensionality of symptoms and quality of life.

PURPOSE: To comprehend the complex relationship between symptoms and health-related quality of life (HRQoL) in patients with diffuse glioma, we applied symptom network analysis to identify patterns of associations between depression, cognition, brain tumor-related symptoms, and HRQoL. Additionally, we aimed to compare global strength between symptom networks to understand if symptoms are more tightly connected in different subgroups of patients. METHODS: We included 256 patients and stratified the sample based on disease status (preoperative vs. postoperative), tumor grade (grade II vs. III/IV), and fatigue status (non-fatigued vs. fatigued). For each subgroup of patients, we constructed a symptom network. In these six networks, each node represented a validated subscale of a questionnaire and an edge represented a partial correlation between two nodes. We statistically compared global strength between networks. RESULTS: Across the six networks, nodes were highly correlated: fatigue severity, depression, and social functioning in particular. We found no differences in GS between the networks based on disease characteristics. However, global strength was lower in the non-fatigued network compared to the fatigued network (5.51 vs. 7.49, p < 0.001). CONCLUSIONS: Symptoms and HRQoL are highly interrelated in patients with glioma. Interestingly, nodes in the network of fatigued patients were more tightly connected compared to non-fatigued patients. IMPLICATIONS FOR CANCER SURVIVORS: We introduce symptom networks as a method to understand the multidimensionality of symptoms in glioma. We find a clear association between multiple symptoms and HRQoL, which underlines the need for integrative symptom management targeting fatigue in particular.

Network Intervention Analyses of cognitive therapy and behavior therapy for insomnia: Symptom specific effects and process measures.

Cognitive therapy (CT) and behavior therapy (BT) are both effective for insomnia. In this study we applied Network Intervention Analysis to investigate specific effects of CT and BT on outcomes and process measures. The analysis was based on a randomized controlled trial comparing CT (n = 65), BT (n = 63) and cognitive behavioral therapy for insomnia (n = 60; not included in this study). In the first networks, the separate items of the Insomnia Severity Index and sleep efficiency were included. In the second networks, the pre-specified process measures for BT and CT, sleep efficiency, and the sum-score of the Insomnia Severity Index were included. At the different time points, we found CT-effects on worry, impaired quality of life, dysfunctional beliefs, and monitoring sleep-related threats, and BT-effects on sleep efficiency, difficulty maintaining sleep, early morning awakening, time in bed, sleep incompatible behaviors and bed- and rise time variability. These observed effects of CT and BT were consistent with their respective theoretical underpinnings. This study provided new information on the mechanisms of change in CT and BT. In the future, this may guide us to the most effective treatment modules or even subsets of interventions.