The Effect of a Motivational Interviewing Style in Cognitive Therapy for Depression

Abstract

While cognitive therapy (CT) is one of the most well-validated and widely used treatments for patients with Major Depressive Disorder (MDD), many individuals remain symptomatic at the end of treatment or drop out prematurely (Cuijpers, van Straten, Anderson, & van Oppen, 2008; Vittengl, Clark, Dunn, & Jarrett, 2007). Evidence suggests that certain types of therapist relational styles, such as one characterized by empathy and support, are facilitative of better therapeutic outcomes (Keijsers, Schaap, & Hoogduin, 1997) and motivational interviewing (MI; Miller & Rollnick, 2002) is a therapeutic approach which emphasizes this type of relational stance. The present study examined whether cognitive therapists exhibit a relational stance that is emphasized in motivational interviewing and whether this relational stance is associated with greater symptomatic improvement in cognitive therapy for depression. The Motivational Interviewing Treatment Integrity Skills Code (MITI; Moyers, Martin, Manuel, Miller, & Ernst, 2007), an observational coding system originated for assessing fidelity to MI, was used to assess three aspect of MI relational stance (MI Spirit, MI Adherent behaviors, and MI Nonadherent behaviors) among cognitive therapists in a randomized-controlled of CT for individuals with moderate to severe depression. Multilevel modeling was employed to examine the effect of MI relational stance on overall symptom trajectories throughout treatment and subsequent symptom reduction immediately after the use of MI relational stance. In order to rule out early symptom reduction as a potential confound, shared variance between MI relational stance and early symptom reduction was removed. The hypothesis that MI relational stance would be associated with more rapid symptom reduction was confirmed for MI Adherent behaviors but not for MI Spirit or MI Nonadherent behaviors. The prediction that initial depression severity would moderate the effect of MI relational stance on symptomatic improvement was not confirmed; however, a three-way interaction between initial depression severity, MI Adherence, and session number revealed that patients with high initial depression severity did not significantly improve through the course of therapy unless they received high MI Adherence. The hypothesis that MI relational stance in a given session would be associated with a reduction in depressive symptoms in the following sessions across the first four sessions was not confirmed. As predicted, early clinical improvement was not associated with MI relational stance in a later session, suggesting that MI relational stance was not merely an artifact of early clinical improvement. There was no support for the prediction that MI relational stance would be associated with subsequent retention in therapy or the therapeutic alliance. Overall, these findings suggest that a specific type of MI relational stance, MI Adherent behaviors, contribute to more symptomatic improvement. Implications of the role of MI relational stance in cognitive therapy are discussed.