Treatment Outcomes, Common Factors, and Continued Neglect of Mechanisms of Change


Address correspondence to Alan E. Kazdin, Child Study Center, Yale University School of Medicine, 230 S. Frontage Road, New Haven, CT 06520-7900.


In this commentary, three key theses are advanced. First, the presence or absence of differences among two or more treatments in a therapy outcome study does not necessarily have implications for the role of common factors as an explanation of treatment effects. Common factors can vary widely among conditions, whether they are veridical treatment or control conditions, and might explain outcome differences or the absence of differences. Second, the therapeutic alliance literature, although vast, also cannot be used to support or refute a common-factors view. Fundamental issues include establishing the timeline of alliance and symptom changes; delineating alliance from other client, therapist, and other aspects of treatment; and identifying and testing the specific ways in which alliance produces change are far from resolved. Third, key lines of research continue to be neglected that could elaborate the role of common factors. Direct tests of common factors, especially experimental studies that manipulate factors to alter outcome, would be enormously valuable. In addition, attention to the mechanisms of therapeutic change where specific mechanisms have been proposed is needed.