In one of his first messages as 1998-1999 president of the American Psychiatric Association, Dr. Rodrigo Munoz called for a plan to return psychiatrists to "a time when we will once again have a direct relationship with our patients, independently from those who want to limit care for financial gain" (1). Who can disagree with the need for more direct relationships with patients, provided all mental health professions are included in working toward this goal? But at this late hour for clinicians' independence from managed care's oversight, how can psychiatry facilitate a strategy to reach this goal?

It is naive to think that our profession will succeed in regaining our independence as a result of political will alone. We need to operationalize clinical practice to respond to payers' expectations. Had we been able to clearly demonstrate the value of our treatments early in the managed care era, we would never have been subjected to the current level of intrusion from forces purchased by payers. Ultimately, to be successful, we will have to demonstrate that we are delivering best practices.

As a psychiatrist and as the editor of the Best Practices column in this journal, I am concerned about the dearth of publications describing how to measure the variability that exists in psychiatric practice and how to use that variability to identify benchmarks or best approaches to clinical problems (2). Several obstacles to developing a best-practices mentality exist, and many mental health care professionals seem to lack clarity about how and where the best-practices process should be initiated.

One obstacle is psychiatry's historical tendency to view issues of practice from the perspective of the individual clinician. Until recently, psychiatry was a cottage industry, with individual practitioners and individual clinics or hospitals functioning independently. However, the best-practices perspective requires an organizational focus that takes into account variability in practice at the level of the individual clinician, groups of clinicians, clinics, hospitals, agencies, states, and regions.

Because of the changing health care climate in the U.S., the majority of psychiatric services now come from organized entities. They include behavioral health carve-out companies that handle commercial as well as public-sector benefits; state systems, such as state departments of mental health, and federal systems, such as the health services system of the Department of Veterans Affairs, that treat seriously mentally ill populations; and organized nongovernmental delivery systems, some owned and operated by physicians. Several models for organizing physician-owned and -operated systems have been developed, including physician-hospital organizations, physician-sponsored organizations, individual practice associations, foundations, medical service organizations, integrated delivery systems, and group practices that assume financial risk for treatment.

These organized approaches encompass components necessary to establish best practices—data collection systems, quality assurance functions, and provider networks. The driving forces of the best-practice effort are found in the systems' quality assurance and regulatory functions. Although no individual practitioner can establish best practices in the field of psychiatry alone, individuals can support and participate in the process.

A second obstacle to establishing a best-practices mentality is the general fear that the exercise will result in criticism and loss of business for someone. Unfortunately, although research has demonstrated the efficacy of many psychiatric treatments, the results have not convinced payers that this efficacy generalizes to all practitioners and care systems. The behavioral health carve-outs are actively implementing procedures to delineate best psychiatric practices. To be competitive, independent delivery systems organized by physicians must take risks to demonstrate their effectiveness.

Another obstacle is the belief that nonacademic delivery systems do not have the expertise to measure and interpret variability in psychiatric practice accurately. However, as in other areas of medicine, the data for establishing best practices come mainly from claims or encounter reports and clinical measures (3). National agencies such as the Joint Commission for Accreditation of Healthcare Organizations, the Agency for Health Care Policy and Research, and the National Committee for Quality Assurance, which is refining the Health Plan Employer Data and Information Set, have used these data sources to develop a wide range of quality indicators. Examples of these indicators include hospital readmissions within 30 days of discharge, posthospital continuation of depression treatment,

availability of medication management and psychotherapy for patients with schizophrenia, appropriate use of medications, and patient satisfaction with mental health care. Psychiatry needs to generate meaningful information that allows practitioners to improve these indicators.

The purpose of the Best Practices column is to report experiences with the process of identifying or developing sources of information about the effectiveness of psychiatric treatments in actual clinical situations and using this information to improve clinical care. Yes, human nature and psychiatric practice is wonderfully complex, but we can identify differences in how we practice, and we can correlate those differences with successes as well as failures in treatment of patients. If you have material that may be suitable for this column, contact me at my Web site (www.glazmedsol.com). Your experience may be extremely valuable to the field.

Dr. Glazer, the editor of this column, is associate clinical professor of psychiatry at Harvard Medical School and Massachusetts General Hospital. Address correspondence to him at 100 Beach Plum Lane, Menemsha, Massachusetts 02552 (e-mail, glazmedsol@aol.com).