As many have noted, jails and prisons have become the mental hospitals of last resort. In November 2000 Congress authorized funds for the creation of up to 100 mental health courts to divert people with severe mental illness from the criminal justice system. In this month's issue Amy Watson, M.A., and her colleagues describe mental health courts that have been operating in four jurisdictions, their basis in jurisprudence, and the continuum of approaches for meeting the needs of local populations. The authors present critical issues that must be considered in establishing mental health courts (see page 477). As Henry J. Steadman, Ph.D., and his colleagues point out, mental health courts are based on the drug court model, first introduced in 1989. These authors advise caution in the rush to duplicate the relative success of this therapeutic approach to persons who may not belong in the criminal justice system. They speculate about whether mental health courts are a "shell game" involving the neediest population (see page 457).

In 2001 a special focus of Psychiatric Services is on evidence-based practices in the treatment of persons with severe mental illnesses. In this month's contribution, Robert E. Drake, M.D., Ph.D., and his colleagues from a national panel describe the evolution of dual diagnosis programs since the problem was recognized in the late 1980s. They discuss components of dual diagnosis services that have been proved effective by research, limitations of the research, barriers to implementing such programs in routine treatment settings, and strategies for overcoming those barriers (see page 469).

For the next five months, the Datapoints column will highlight research on current psychiatric practice and on how psychiatrists are faring in today's health care environment. This month, Roland Sturm, Ph.D., looks at the role of computer use in day-to-day practice in psychiatry, other specialties, and primary care. His research shows that psychiatrists are lagging behind their peers in the use of computers to obtain or record clinical data and to obtain treatment information. In particular, he notes a gender gap in the use of computers in psychiatry (see page 443). Future Datapoints columns will examine psychiatrists' satisfaction with their medical careers compared with other physicians as well as differential demographic characteristics, practice patterns, and professional work activities.

Is the course of severe mental illness inevitable deterioration? What are we to make of first-person accounts by consumers of their recovery, or the accounts of clinicians engaged in helping people recover from severe mental illness? Nora Jacobson, Ph.D., and Dianne Greenley, M.S.W., J.D., describe a model developed in Wisconsin to help state programs move toward a "recovery-oriented" mental health system. The model incorporates experiences of those who describe themselves as being in recovery—hope, healing, empowerment, and connection—as well as external conditions that make those experiences possible—implementation of the principle of human rights, a positive culture of healing, and recovery-oriented services (see page 482). In a related commentary, Herbert Peyser, M.D., characterizes Dr. Jacobson and Ms. Greenley's article as a "hopeful exposition and a forceful advocacy that does not claim to be objective." He wonders whether clinicians enamored of the recovery model have lost sight of the degree to which some patients lose their freedom to mental illnesses (see page 486).

The 1996 Mental Health Parity Act mandated limited parity for mental health coverage. Four actuarial studies conducted at the time estimated that health insurance premiums would increase from 3.2 percent to 8.7 percent. In the Economic Grand Rounds column, Merrile Sing, Ph.D., and Steven C. Hill, Ph.D., describe the assumptions behind those and other cost estimates for limited and full parity. They then use their own model to estimate premium increases for behavioral health parity options for different diagnostic groups and types of plans. Under this model, premium increases to fund full parity range from .6 percent for health maintenance organizations to 5.1 percent for preferred-provider organizations (see page 437).

The proportion of Medicaid beneficiaries enrolled in managed care plans has increased dramatically—from 9.5 percent in 1991 to 55.6 percent in 1999. Kristina W. Hanson and Haiden A. Huskamp, Ph.D., describe the wide variation in the ways states deliver behavioral health care to Medicaid populations. They conclude that although a state-based approach in which each state functions as a "laboratory" for policy experimentation has its advantages, the approach raises ethical concerns and the possibility of migration across state borders, which may lead to a "race to the bottom" among states (see page 447).