Assessing treatment integrity in cognitive-behavioral therapy: comparing session segments with entire sessions.

The evaluation of treatment integrity (therapist adherence and competence) is a necessary condition to ensure the internal and external validity of psychotherapy research. However, the evaluation process is associated with high costs, because therapy sessions must be rated by experienced clinicians. It is debatable whether rating session segments is an adequate alternative to rating entire sessions. Four judges evaluated treatment integrity (i.e., therapist adherence and competence) in 84 randomly selected videotapes of cognitive-behavioral therapy for major depressive disorder, social anxiety disorder, and hypochondriasis (from three different treatment outcome studies). In each case, two judges provided ratings based on entire therapy sessions and two on session segments only (i.e., the middle third of the entire sessions). Interrater reliability of adherence and competence evaluations proved satisfactory for ratings based on segments and the level of reliability did not differ from ratings based on entire sessions. Ratings of treatment integrity that were based on entire sessions and session segments were strongly correlated (r=.62 for adherence and r=.73 for competence). The relationship between treatment integrity and outcome was comparable for ratings based on session segments and those based on entire sessions. However, significant relationships between therapist competence and therapy outcome were only found in the treatment of social anxiety disorder. Ratings based on segments proved to be adequate for the evaluation of treatment integrity. The findings demonstrate that session segments are an adequate and cost-effective alternative to entire sessions for the evaluation of therapist adherence and competence.