their potential advantages and risks through the informed consent process. If a new treatment in general health care is considered experimental, review by an institutional review board is required. Psychotherapies are unique in this regard in that a given therapist may offer a new therapeutic approach without its undergoing a safety or effectiveness review and without having to inform the patient about the extent to which its safety and effectiveness have been established.
The committee concludes that a more comprehensive, systematic, and coordinated approach is needed to describe, assess, and classify M/SU treatments and practices according to the level of evidence that supports their use. Better coordination of current national and international review activities, as well as coordination of those efforts with the evidence review activities that underlie the guideline development process of many organizations, could prevent redundancy and waste, produce more evidence reviews on a timelier basis, and avoid conflicting interpretations of the data for clinicians and consumers. The organizations engaged in these activities are natural partners for building a more comprehensive, coordinated, and systematic review network. Many of these same organizations are also involved with the dissemination of their review findings in the form of practice guidelines and other clinical decision-support tools.
The production of evidence will be less fruitful if it is not accompanied by accurate diagnosis and comprehensive longitudinal assessment. Because having a mental illness or alcohol- or other substance-use diagnosis is a leading risk factor for suicide (Maris, 2002), failure to diagnose these conditions can be lethal. An inaccurate diagnosis also can lead to ineffective treatment and even harmful outcomes. Yet individuals with the same symptoms presenting to different mental health clinicians can receive different diagnoses. For example, variations have been documented in the extent to which depression is diagnosed in individuals with similar symptoms by both psychiatrists (Kramer et al., 2000) and primary care providers (Mojtabai, 2002) and in the extent to which ADHD is diagnosed within different communities (Lefever et al., 2003). Recently, the diagnosis of bipolar illness in children, especially preschoolers, has been the subject of considerable controversy among psychiatrists (McClellan, 2005). For many conditions, significant discrepancies have been observed among diagnoses generated from structured interviews for research purposes and those resulting from clinician judgments (Lewczyk et al., 2003) and diagnostic tools developed for clinical purposes (Eaton et al., 2000).
In children, diagnoses may have an even greater range of variability because diagnostic manifestations change over the course of development.