Treatment Processes and Mediators of Substance Use Disorders Treatment Effects: The Benefits of Side Road Excursions


  • Conflict of Interest: The author states no conflict of interest.

Reprint requests: John W. Finney, PhD, VA Palo Alto Health Care System, Menlo Park Division (152), 795 Willow Rd, Menlo Park, CA 94025; Fax: (650)-617-2736; E-mail:


Background:  This article provides some commentary on six brief presentations on treatment processes and mediators of substance use disorder (SUD) treatment effects that were made at the 2005 Research Society on Alcoholism Satellite Preconference on “The Search for Mechanisms of Change in Evidence-Based Behavioral Treatments.”

Methods:  Six presentations are discussed.

Results:  The foci of the presentations ranged from the linkages of intermediate outcomes of cognitive-behavioral and 12-step treatment (e.g., spirituality) to posttreatment substance use to the relationship between treatment completion of brief motivational interviewing and treatment outcome. None of the six presentations met all or even most of the criteria for analyses of mediation of intervention effects outlined by Kazdin and Nock. In fact, some did not examine mediation at all. However, the presentations were valuable in illustrating that important new knowledge can be gained from addressing SUD treatment processes and mechanisms of treatment change.

Conclusions:  Given the paucity of extant research on mediation of SUD treatment effects, correlational explorations of treatment processes and treatment mediators are useful at this point, even though most mediational analyses do not meet many of the criteria listed by Kazdin and Nock. More attention should be devoted to assessing mediators before the assessment of outcome variables (e.g., during treatment) in order to address temporal precedence. Only after reasonable evidence has accrued regarding statistical mediation (expected associations, temporal precedence, reduction in treatment effects when putative mediator is controlled) does it make sense to mount expensive studies with experimental designs to examine causal mediation effects.