Does Centrality in a Cross-Sectional Network Suggest Intervention Targets for Social Anxiety Disorder?

Abstract

Objective: Network analysis allows us to identify the most interconnected (i.e., central) symptoms, and multiple authors have suggested that these symptoms might be important treatment targets. This is because change in central symptoms (relative to others) should have greater impact on change in all other symptoms. It has been argued that networks derived from cross-sectional data may help identify such important symptoms. We tested this hypothesis in social anxiety disorder. Method: We first estimated a state-of-the-art regularized partial correlation network based on participants with social anxiety disorder (N = 910) to determine which symptoms were more central. Next, we tested whether change in these central symptoms were indeed more related to overall symptom change in a separate dataset of participants with social anxiety disorder who underwent a variety of treatments (N = 244). We also tested whether relatively superficial item properties (infrequency of endorsement and variance of items) might account for any effects shown for central symptoms. Results: Centrality indices successfully predicted how strongly changes in items correlated with change in the remainder of the items. Findings were limited to the measure used in the network and did not generalize to three other measures related to social anxiety severity. In contrast, infrequency of endorsement showed associations across all measures. Conclusions: The transfer of recently published results from cross-sectional network analyses to treatment data is unlikely to be straightforward.

Bridging eating disorder symptoms and trait anxiety in patients with eating disorders: A network approach

“…Identifying bridge nodes may enhance our understanding of comorbidity processes (c.f. Rodebaugh et al, ). Moreover, simulations that model the spread of one psychiatric disorder to others suggest that when treating comorbidities, bridge nodes may represent the most influential intervention targets (Jones et al, ).…”

“…In addition, debate exists about whether central nodes identified in cross‐sectional networks are meaningful intervention targets. Emerging evidence supports that central nodes in cross‐sectional networks can predict treatment outcomes (Olatunji, Levinson, & Calebs, ; Rodebaugh et al, ), but prediction may depend on whether the same measures are used to estimate the network and assess outcomes (Rodebaugh et al, ).…”

“…We combined theoretical and data‐driven approaches to select EDE‐Q and STAI‐T items to include as nodes in the network (Fried & Cramer, ; Levinson et al, ; Rodebaugh et al, ). Theoretical decisions about node selection were made first.…”

Bridging the gap: A network approach to Dark Triad, Mental Toughness, the Big Five, and perceived stress

“…The informative value of centrality indices in psychological networks is currently debated. Some studies have shown that central elements of psychopathology networks have high clinical relevance (Boschloo, van Borkulo, Borsboom, & Schoevers, ; Christensen, Kenett, Aste, Silvia, & Kwapil, ; Rodebaugh et al, ). However, centrality indices have been originally developed for social networks and might not be adequate to capture the type of relationships encoded by psychological networks (Bringmann & Eronen, ).…”

What do centrality measures measure in psychological networks?

“…However, a study on social anxiety disorder (Rodebaugh et al 2018) found that although degree centrality (not closeness or betweenness centrality) seemed to have some utility in predicting change processes and social anxiety severity, it was simply the number of times that patients endorsed (i.e., reported) a symptom that had the most predictive power in indicating which items were the most important ones. The authors conclude that clinicians could use highly central symptoms of cross-sectional networks, but simply treating the most commonly reported symptoms would probably work better (Rodebaugh et al, 2018). Furthermore, there remains a lack of studies on the predictive value of centrality indices in temporal networks.…”