fects of a clinical intervention (Chambless and Hollon, 1998) and in determining that an intervention can work. However, demonstration of efficacy in an RCT does not guarantee that the treatment will be effective in actual practice settings. For example, some reviews suggest that behavioral interventions in psychotherapy are generally beneficial (Matt and Navarro, 1997), others suggest that interventions are less effective in clinical settings than in the laboratory (Weisz et al., 1992), and others find particular interventions equally effective in experimental and clinical settings (Shadish et al., 1997).

The Division of Clinical Psychology of the American Psychological Association recently established criteria for “empirically supported” psychological treatments (Chambless and Hollon, 1998). In an effort to establish a level of excellence in validating the efficacy of psychological interventions the criteria are relatively stringent. A treatment is considered empirically supported if it is found to be more effective than either an alternative form of treatment or a credible control condition in at least two RCTs. The effects must be replicated by at least two independent laboratories or investigative teams to ensure that the effects are not attributable to special characteristics of a specific investigator or setting. Several health-related behavior change interventions meeting those criteria have been identified, including interventions for management of chronic pain, smoking cessation, adaptation to cancer, and treatment of eating disorders (Compas et al., 1998).

An intervention that has failed to meet the criteria still has potential value and might represent important or even landmark progress in the field of health-related behavior change. As in many fields of health care, there historically has been little effort to set standards for psychological treatments for health-related problems or disease. Recently, however, managed-care and health maintenance organizations have begun to monitor and regulate both the type and the duration of psychological treatments that are reimbursed. A common set of criteria for making coverage decisions has not been articulated, so decisions are made in the absence of appropriate scientific data to support them. It is in the best interest of the public and those involved in the development and delivery of health-related behavior change interventions to establish criteria that are based on the best available scientific evidence. Criteria for empirically supported treatments are an important part of that effort.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement