The Therapy Process Observational Coding System: Group cohesion scale in youth anxiety treatment: Psychometric properties

Background There are no well‐established measures of group cohesion, defined as the collaborative bond between group members, in group cognitive behavioral therapy (GCBT) with youth. We therefore examined the Therapy Process Observational Coding System for Child Psychotherapy—Group Cohesion Scale (TPOCS‐GC), which has previously only been used with adult samples, in a youth sample. Methods Observers coded 32 sessions from 16 groups with 83 youth aged 8 to 15 years (90.7% European White). Youth had anxiety disorders and received manualized GCBT in community clinics. We examined psychometric properties of the TPOCS‐GC and its' construct validity in terms of relations with pretreatment variables, alliance and fidelity during treatment, and post‐treatment variables. Group cohesion was measured twice during treatment (early and late). Results The TPOCS‐GC was internally consistent (α = 0.72) and was reliably coded (M ICC = 0.61). Higher clinical severity at pretreatment predicted lower early group cohesion. Higher youth age, higher clinical severity at pretreatment, and higher youth‐rated early alliance predicted lower late group cohesion. Higher therapist‐rated early alliance predicted higher early group cohesion. Higher therapist‐rated late alliance predicted higher late group cohesion. Higher late group cohesion predicted lower clinical severity and higher client treatment satisfaction at post‐treatment. Early group cohesion did not predict any post‐treatment variables. Conclusions A four‐item version of the TPOCS‐GC can be reliably used in youth GCBT. The TPOCS‐GC is distinct from, but associated with, multiple clinical variables.