To the Editor: Our study did not discredit randomized controlled trials of psychotherapy but rather pointed out their limitations. Brief therapies studied in randomized controlled trials have different brand names and manuals prescribing different therapist interventions. Nevertheless, randomized controlled trials did not reveal what actually occurred in these treatments. Randomized controlled trials can provide evidence of efficacy but not evidence to support a therapy’s purported theory of change. Our study demonstrated that treatments may promote change in different ways than their underlying theories of therapy claim.

Dr. Markowitz asks whether our methods are biased and alleges that the Psychotherapy Process Q-set (1) cannot discriminate between interpersonal therapy and cognitive behavior therapy. Dr. Markowitz incorrectly states that the Psychotherapy Process Q-set was designed to study psychoanalytic psychotherapy. The Psychotherapy Process Q-set is pantheoretical, has demonstrated excellent discriminate validity, and can differentiate effectively among any number of therapies (2). In fact, almost one-half of the 100 Q-set items significantly differentiated interpersonal therapy and cognitive behavior therapy in the data set from the NIMH Treatment of Depression Collaborative Research Program (3). The Q items do indeed define strategies and techniques (e.g., "Therapist presents an experience or event in a different perspective" refers to cognitive restructuring). Dr. Markowitz mistakenly seems to think we reported that the Q-set could not differentiate the two treatments. What we found was that interpersonal therapy, as conducted by the therapists in this study, conformed more closely to what experts considered an ideal (or prototype) of cognitive behavior therapy than it did to a distinct prototype of interpersonal therapy.

The method used to create the prototypes, the Q technique, is a statistical approach for studying points of view (4). Dr. Markowitz acknowledges that he failed to respond to our questionnaire. It is a shame that he chose not to register his opinion so that it could be considered in our analyses along with those of the other experts sampled. The large majority of interpersonal therapy and cognitive behavior therapy experts contacted did respond and reported that the method captured the important aspects of their respective treatment approaches. As stated in the article, the experts were very experienced and had trained therapists in their orientation. Most had published work concerning their approach to therapy, and many were involved in the development of their treatment modality.

Apparent differences among newer manualized therapies may lie mostly in terminology and the ways of conceptualizing psychological constructs and processes that are actually quite similar. As we pointed out, the content of the cognitive behavior therapist’s focus (dysfunctional attitudes and irrational beliefs) is often quite different from the content of the interpersonal therapist’s focus (e.g., disruptions in personal relationships). However, when we shift our attention from content to process (i.e., the interaction between the therapist and patient), the similarities are compelling. In both treatments, the therapist assumed an active, authoritative role, coached compliant patients to think or conduct themselves differently, and encouraged them to test these new ways of thinking and behaving in everyday life. Most brief therapies probably promote change through similar processes, and specific techniques are likely less important. That is why—Dr. Markowitz’s claims notwithstanding—it has been so difficult to demonstrate any large or consistent differences in outcome across types of brief therapies (5).