Active Ingredients: How and Why Evidence-Based Alcohol Behavioral Treatment Interventions Work Richard Longabaugh, Dennis M. Donovan, Mitchell P. Karno, Barbara S. McCrady, Jon Morgenstern, and J. Scott Tonigan This article summarizes the proceedings of a symposium that was organized and chaired by Richard Longabaugh and presented at the 2004 Research Society on Alcoholism meeting in Vancouver, British Columbia, Canada. The aim of the presentation was to focus on evidence for the active ingredients of behavioral therapies for patients with alcohol use disorders. Dennis M. Donovan, PhD, reviewed evidence for the active ingredients of cognitive behavioral therapy. Barbara S. McCrady, PhD, presented a concep- tual model for mechanisms of change in alcohol behavior couples therapy and reviewed evidence for this model. J. Scott Tonigan, PhD, presented data testing three hypothesized mechanisms of change in twelve- step facilitation treatment. Mitchell P. Karno, PhD, presented therapy process data that tested whether matching therapist behaviors to client attribute across three therapies affected drinking outcomes. Jon Morgenstern served as discussant. Key Words: Active Ingredients of Treatment, Mechanisms of Change, Matching. BACKGROUND AND CONTEXT FOR A FOCUS ON MECHANISMS OF CHANGE IN BEHAVIORAL TREATMENTS Richard Longabaugh More than 20 years ago, Moos and Finney (1983) pub- lished a seminal paper on the black box of treatment in the American Psychologist, “Expanding the Scope of Evalua- tion.” Their central point was that we do not know what treatment involves or how treatment works. Despite their efforts and those of others, it has been only in the last 5 years or so that this has become a high priority for the alcohol treatment field. First, there was the issue of establishing that alcohol treatments could be effective and that some were more effective than others. Miller and colleagues have conducted and updated comprehensive reviews of treatment effective- ness studies for the past 15 years. We now have confidence that many alcohol treatments are effective, some more than others (Miller and Wilbourne, 2002), but despite the suc- cessful efforts of alcohol treatment researchers to conduct rigorous clinical trials and spell out in treatment manuals what the treatments consist of, little research has actually investigated what the active ingredients of these treatments are. The prototypic publication testing the effectiveness of a treatment would involve a rationale for the therapy de- livered, followed by a test of the treatments, and for those treatments that tested as effective, a discussion of why the therapy worked. Missing was an actual test of the theory underlying the treatment’s effectiveness. We assumed that because the treatment was effective, we knew how and why it worked. Project MATCH attempted to raise the bar by studying patient-treatment matching. Selecting three effective treat- ments, MATCH sought to show that each of these treat- ments was more effective for some types of patients be- cause of specific active ingredients that each emphasized. For example, motivational enhancement was expected to be especially effective for less motivated patients because of its focus on increasing and sustaining motivation, but this hypothesis and most others were not supported (Long- abaugh and Wirtz, 2001a; Project MATCH Research Group, 1997a,b). In a decade-long examination of the causal chains hy- pothesized to underlie these matching hypotheses, several reasons were identified for their lack of support. One im- portant reason was that the three treatments often did not differ from one another in their emphasis on hypothesized active ingredients. Thus, although patients of a subtype Bio-Med Psychiatry (RL, MPK), Center for Alcohol and Addiction Stud- ies, Brown University, Providence, Rhode Island; Alcohol and Drug Abuse Institute and Department of Psychiatry and Behavioral Sciences (DMD), University of Washington, Seattle, Washington; Center of Alcohol Studies and the Graduate School of Applied and Professional Psychology (BSM), Rutgers, The State University, Piscataway, New Jersey; Department of Psychiatry and Health Policy (JM), Mount Sinai School of Medicine, New York, New York; and Psychology Department (JST), University of New Mexico, Albuquerque, New Mexico Received for publication September 3, 2004; accepted October 3, 2004. This symposium and the work presented were sponsored by National Institute on Alcohol Abuse and Alcoholism Grants U10-AA11799 (DMD), R01-AA12155 (MPK), R37-AA07070 (BSM), R01-AA10268-07 (JM), and K02-AA00326-01 (JST). Reprint requests: Richard Longabaugh, Bio-Med Psychiatry, CAAS, Box G-BH, Brown University, Providence, RI 02912; Fax: 401-444-1850; e-mail: Richard_Longabaugh@brown.edu. Copyright © 2005 by the Research Society on Alcoholism. DOI: 10.1097/01.ALC.0000153541.78005.1F 0145-6008/05/2902-0235$03.00/0 ALCOHOLISM:CLINICAL AND EXPERIMENTAL RESEARCH Vol. 29, No. 2 February 2005 Alcohol Clin Exp Res, Vol 29, No 2, 2005: pp 235–247 235