(PS7-80) The Impacts of Group Cohesion on the Effectiveness of DBT Group Therapy

Skills training in the context of group therapy is an important part of dialectical behavior therapy (DBT; Linehan, 1993, 2015). In fact, DBT group is one of the core ingredients that contributes to the efficacy of this treatment (Linehan et al., 2015). Although use of skills has been pinpointed as one mechanism of change in the link between DBT and improved outcomes (Neacsiu et al., 2010), there are other aspects of DBT skills training groups that may contribute to improvements in treatment. One such element is group cohesion, or the degree to which the group members have a positive bond with each other (Burlingame et al., 2018). Given that patients with borderline personality disorder (BPD) undergoing DBT treatment tend to have at least some problems with interpersonal functioning (Lazarus et al., 2014), group cohesion may be particularly important for this population, as group therapy offers an excellent space to practice interacting with other people. Thus, group cohesion could potentially facilitate BPD symptom improvement. It is also possible that DBT, which focuses on skills acquisition, may rely less on group cohesion than other interventions (e.g., interpersonal psychotherapy).

The present study will examine two questions: (1) Does interpersonal psychotherapy vs. DBT skills groups predict greater group cohesion? and (2) Does more group cohesion predict improvements in BPD over 6 weeks of group therapy? Participants (N = 84, M_age = 25.61, SD = 9.61, 88.1% female) with 4+ BPD criteria on the Structured Clinical Interview for DSM Disorders (SCID-IV; First et al., 1994)  underwent 6 weeks of virtual group therapy (either DBT emotion regulation [ER] skills training, DBT interpersonal effectiveness [IE] skills training, or an interpersonal psychotherapy [IP] group). Group cohesion was assessed using the group questionnaire (Burlingame et al., 2017; Burlingame et al., 2010) and BPD symptoms were assessed with the borderline personality symptom list (BSL23; Bohus & Limberger, 2009). Preliminary analyses revealed that there was significantly higher group cohesion in the IP group (B = 7.95, SE = 3.94, p = .046) compared to DBT-ER, and no significant differences in group cohesion between the two DBT conditions. These analyses also indicated that all the three groups experienced symptom reduction (BSL23).

It is worth mentioning that previous research has shown that virtual compared to in-person DBT groups foster lower group cohesion (Lopez et al., 2020). Results will aide our understanding of whether DBT groups that reduce negative symptoms for patients focus more on skills acquisition compared to group cohesion. Further research should investigate which particular skills are bolstered by increased group cohesion. Further research should investigate which specific DBT skills group cohesion may have more influence over.