Towards a better understanding of persistence of anxiety disorders: a network perspective

Background: : This study aims to gain more insight into participants with anxiety disorders by conducting network analysis with experience sampling methodology data (i.e. data measured multiple times a day via smartphone), comparing group networks of participants diagnosed with an anxiety disorder and healthy controls. We expect the networks of participants with an anxiety disorder to be more connected and to reveal relationships between symptoms that may perpetuate problems. Methods: 224 participants, 141 diagnosed with an anxiety disorder (Group A) at baseline (T0) and 83 healthy controls (Group B), from the Netherlands Study of Depression and Anxiety participated in ESM monitoring at 9-year follow-up (T1). Anxiety disorder status was assessed at T1 as well and used to group participants into group A-1, anxiety disorder at T0 and T1, and A-2, anxiety disorder at T0 but not at T1. 6 ESM items were used to conduct network analysis. The permutation test mnet was used to inspect the networks for significant differences. Results: The autoregression coefficients of ESM-items ‘worrying’ and ‘nervousness’ were larger in Group A compared to group B, as well as the temporal influence of worrying on tiredness. The autoregression coefficients of worrying and tiredness were larger in Group A-2 than in Group A-1, those of enthusiasm were larger in Group A-1. The influence of tiredness on apathy was greater in Group A-2. Conclusions: It seems likely that participants with chronic mood- and anxiety disorders are not distinct from each other on a level of symptomatic interconnectedness, even if they differ in diagnostic categories 9 years after baseline. These findings confirm the existing information regarding the chronic course that anxiety disorders often take.