Have all won and must all have prizes? Revisiting Luborsky et al.'s verdict.

Although most reviews of comparative psychotherapy literature have failed to find significant differences among treatments, it is premature to give up the search for differential effects. There are a large number of patient, therapist, and treatment variables that may mediate the effects of treatments. Given the enormity of the task of exploring potential interactions among the many patient, therapist, and psychotherapy types, a guiding model is needed by which to narrow our search for variables that mediate between treatment type and outcome. However, theoretical constructs that represent both patient and therapy variations frequently are poorly defined. Tests of treatment selection models that cut across narrow theoretical differences among psychotherapies and that operationalize definitions of patient types hold promise for revealing meaningful Patient x Therapist interaction effects in psychotherapy.