A network analysis of how obsessive-compulsive symptoms change during exposure and response prevention treatment

Although exposure and response prevention (EX/RP) is recommended as a first-line treatment for obsessive-compulsive disorder (OCD), responses vary among patients. This study was the first to use network analysis to examine how OCD symptom networks change with EX/RP and vary across different progress trajectories.

Data from four clinical trials with 334 adults with OCD who received manualized EX/RP were pooled. The Yale-Brown Obsessive-Compulsive Scale (Y-BOCS) was administered at baseline, midpoint, and post-treatment. OCD symptom networks were constructed using individual Y-BOCS items at these three time points, both for the entire sample and for three different progress trajectories (dramatic, moderate, and little-to-no progress) previously identified using growth mixture modeling. Network measures, including global efficiency, modularity, and weighted degree centrality, were computed to quantitatively assess network properties across treatment.

Network analysis revealed two distinct modules at baseline: resistance/control and interference/distress. In the full sample, these two modules became integrated over time, as indicated by significant increases in global efficiency and weighted degree centrality and decreases in modularity; at post-treatment, the network shifted toward a fully connected network, and the strength of associations between nodes increased. These changes were most pronounced in the dramatic progress class.

Our findings indicated that effective EX/RP treatment was associated with more integrated OCD symptom networks, which may serve as an indicator of treatment response. Future research should examine how these shifts in network connectivity correspond to changes in underlying brain circuitry and/or to early identification of treatment responders.