The goal of this article is to delineate training implications regarding harmful effects associated with psychotherapy. The authors strongly recommend that trainees be made aware of (and encouraged to examine carefully) the potentially harmful treatments that have been recently identified (Lilienfeld, 2007). Consistent with a broad perspective on evidence-based practice, it is also argued that additional guidelines for the prevention and repair of harmful impacts can be derived from psychotherapy research on process (technique and relationship) and participant (client and therapist) variables. For example, rigid adherence to the application of psychotherapy techniques can be a potentially harmful therapist behavior that necessitates careful training on the nature and flexible use of interventions. Furthermore, the authors suggest that trainers and supervisors tentatively consider training implications linked to clinical observations and theoretical assertions, such as the premature use of clinical interpretations, with the assumption that more confidence in such therapeutic guidelines can be gained when they are supported by multiple knowledge sources (empirical, clinical, conceptual). Finally, training implications related to the monitoring of harmful effects in terms of treatment outcome and process are demarcated. (Contains 1 table and 3 footnotes.)