In asking whether the therapists theoretical school makes a difference in the outcome of therapy, Don Nathanson suggests that therapeutic systems are the armature around which we build the sculpture of our selves, but they have little to do with the way we work when grown up.
My sense is that this statement is much more true for some schools of therapy than others. Some therapeutic systems conceptualize the client and his or her problems in fairly abstract terms and say little about exactly what the therapist can say or do to facilitate the desired changes. When this is the case, the therapists behavior will be influenced to some extent by the theoretical system but may be influenced more strongly by the supervision the therapist has received, the therapists personal style, and his or her clinical experience.
Other therapeutic systems make more concrete assertions about what the therapist can say and do to facilitate change. In this case, previous supervision, personal style, and clinical experience are still relevant but the therapeutic system will have a stronger impact on the therapists behavior. Certainly, the therapeutic system from which I work (Cognitive Therapy) has a great deal to do with the way I work. My theoretical orientation has an impact both in general (for example, with all clients I try to identify the specific thoughts and feelings that occur in problem situations, use a Socratic approach, and use concrete homework assignments) and in specific situations (for example, I use exposure and response prevention with clients with Obsessive-Compulsive Disorder and use if infrequently with other clients).
Certainly, one could question whether a focus on specific intervention techniques is a good thing. Cognitive Therapy argues that emphasizing intervention techniques and ignoring the therapeutic relationship is a bad idea and we argue that applying intervention techniques in a cookbook fashion is also a bad idea.
We would assert that if you have to choose between a good therapeutic relationship and skill at specific intervention techniques, a good therapeutic relationship is preferrable since this is a precondition for effective intervention. However, we would also assert that adding effective intervention techniques to a good therapeutic relationship can increase the effectiveness of treatment and produce results more quickly.