Staying the Course or Charting a New Path? An Examination of Therapist Options When Working with Low Engaging Patients

Thumbnail Image



Journal Title

Journal ISSN

Volume Title


Research Projects

Organizational Units

Journal Issue


Introduction: Although cognitive behavioral therapy (CBT) is an effective treatment for depression, a substantial portion of CBT patients still do not achieve positive therapeutic outcomes. Those who engage less with homework are especially at risk (Kazantzis & Shinkfield, 2007). When early engagement is low, a therapist might persist in using the same basic approach or change to a new approach. There is limited evidence to guide therapists in selecting between these options. To address this issue, we examine use of persistence versus changing approach strategies in differentiating sessions followed by high or low homework engagement among patients who had shown more limited engagement in early sessions. Methods: Drawing data from a clinical trial of two variants of CBT for depression (N = 150), we focus on a relatively low-engaging subset of 80 participants (53.3%) who reported homework engagement below the 75th percentile across sessions 2, 3, and 4. For each of these participants, observers rated two subsequent sessions, selecting for being followed by relatively high or low engagement for each patient. Using a 6-item scale designed for this study, six raters coded the extent to which therapists continued using previously used strategies or tried new strategies in these sessions. Three scores were derived from the 6-item scale, an overall score and two subscale scores. Results: Observer-raters' judgements of the therapist approach to homework assignment exhibited poor reliability (Intraclass Correlation Coefficient (ICC) < 0.5). Considerable caution is warranted for any findings obtained with these ratings. A paired t-test of sessions followed by low and high engagement revealed a significant difference in the overall score, with sessions preceding high engagement being characterized by both greater persistence and less change of approach. A similar analysis examining overall scores as a predictor in a model of session type was not significant. Additional analyses failed to support either specific subscale in differentiating or predicting session type. Discussion: Although caution is warranted, one finding suggests that the therapist approach of flexibly persisting with the assignment of homework may promote greater homework engagement. Given the challenges with low reliability of the observer ratings, we take this finding as leaving open the possibility that flexible persistence (or not changing approach) may predict homework more robustly in future research. Future research is needed to inform how therapeutic approach can best promote homework engagement when working with low engaging patients.



homework engagement, cognitive behavioral therapy, depression, psychology