Evidence-based practice in psychology has been defined as the integration of the best available research, clinical expertise, and the individual client’s characteristics, values, and preferences (APA, 2006). This definition suggests that psychotherapists should be able, and willing, to integrate techniques from different theoretical orientations based on the context. Although integration is important, it is also important that therapists do not haphazardly utilize a multitude of techniques in their clinical work. Perhaps the need to integrate should also be based on the context!
In one recent study, Goldman, Hilsenroth, Gold, Owen, and Levy (2016) examined the context for when integration might be most helpful for clients. In this naturalistic study, the authors surveyed outcomes for 75 clients who received treatment at a university-based community outpatient clinic. The therapists in this study (29 advanced doctoral students) primarily used a short-term psychodynamic psychotherapy approach, but were allowed to flexibly incorporate cognitive-behavioral techniques with their clients as they deemed appropriate. Goldman and colleagues reviewed video recordings of the 3rd or 4th session for all clients in order to code the use of both psychodynamic-interpersonal (PI) and cognitive-behavioral (CB) techniques. These codings were then compared to ratings of the therapeutic alliance and treatment outcomes.
Use of PI techniques or CB techniques alone did not predict ratings of the alliance or outcome. Interesting, integrating both sets of techniques also did not predict ratings of alliance or outcome. However, Goldman and colleagues (2016) did find an important moderation effect. When there was a lower level of agreement between the client and therapist on the goals and tasks, including both PI and CB techniques was slightly damaging; however, if there was a higher level of agreement on the goals and tasks, clients did much better when their therapist integrated compared to using one set of techniques alone. Similarly, less integration was better when there was a lower level of confident collaboration, but more integration was better when there was a high level of confident collaboration. In summary, integration was beneficial when there was a strong therapeutic alliance, but perhaps somewhat damaging when the alliance was weak!
The results of this study have some important implications for clinical practice. If there is a strong therapeutic relationship, integration is likely better than non-integration. This may be because the client feels more comfortable following the therapist as different techniques are applied. However, if there is a weak therapeutic relationship, therapists may want to work on building that relationship so that agreement on the therapeutic tasks and goals can be obtained, before attempting to integrate techniques from different orientations.
Graph 1. Relation between Technique Integration and Reliable Change in Global Symptom Index of the Brief Symptom Inventory (BSI-GSI-RCI) as a Function of Goals & Task Agreement
Graph 2. Relation between Technique Integration and Reliable Change in Global Symptom Index of the Brief Symptom Inventory (BSI-GSI-RCI) as a Function of Confident Collaboration
Figures from Goldman et al. (2016). Used with permission.