Frequent use of the term "difficult patient" suggests the existence of a single, well-described group. However, in a literature review of 94 articles, Bauke Koekkoek, R.N., C.N.S., and colleagues found evidence to support the existence of three subgroups of such patients—"the unwilling care avoider," "the ambivalent care seeker," and the "demanding care claimer." The authors also found that past studies have offered four patient-level theoretical explanations for the perceived difficulty: chronicity, dependency, character pathology, and lack of reflective capabilities. However, some commentators have located the difficulty in the therapeutic relationship and have focused on the emotional struggles of transference and countertransference. Other observers have called attention to systemic and sociological factors—prejudice, labeling, and exclusion—as the major explanation for the difficulty of treating these patients. Mr. Koekkoek and his colleagues review interventions and approaches found to be helpful with this patient group (page 795). In a related Taking Issue commentary, David A. Adler, M.D., offers guidance and support to clinicians who work with difficult patients in systems that are ill prepared to provide the extensive long-term services that they need (page 767).

Each year about one million persons with severe mental illness spend time in local jails, and most rely on Medicaid to cover needed services upon release. However, more than three-quarters of states have policies mandating termination of Medicaid benefits at jail entry. Two articles by Joseph P. Morrissey, Ph.D., and colleagues focus on the role of Medicaid in the lives of detainees with mental illness. Because of the critical importance of Medicaid coverage in linking detainees with services upon release from jail, the authors were concerned about delays in restoration of terminated benefits. However, their examination of nearly 7,400 detentions of persons with mental illness in two large metropolitan jail systems indicated that despite state policies on termination of benefits, only a few detainees were disenrolled from Medicaid, largely because most cycled rapidly in and out of jail (page 803). In a second analysis the authors found that individuals who had Medicaid benefits when they were released from jail were significantly more likely than those who did not to use services, to obtain them more quickly, and to receive more services (page 809). In a brief report, Amy Blank Wilson, L.S.W., and Jeffrey Draine, Ph.D., M.S.W., present results from a national survey that identified emerging practices in 50 state initiatives to help persons with mental illness reenter the community after incarceration (page 875).

Participation in competitive employment plays an important role in community integration of persons with psychiatric disabilities. However, many fear that they will lose public disability benefits if they enter the workforce or remain in it. To counter this fear, the state of Vermont began providing specialized benefits counseling in 1999, assigning counselors to existing vocational and supported employment programs across the state. The counseling included education about Social Security Administration programs and work incentives, individualized counseling on participants' benefit packages, and assistance in managing their benefits while making the transition to work. Timothy Tremblay, M.S., and his colleagues examined data for 364 counseling recipients and two matched control groups and found significant increases in earnings for those who had worked with a counselor. The authors conclude that benefits counseling should be a fundamental component of evidence-based employment services (page 816).

Both clinicians and researchers benefit when self-report measures are shown to reasonably reflect findings obtained with more time-intensive clinician-rated measures—and it is also welcome news when shorter instruments are shown to be nearly as accurate as longer ones. A. John Rush, M.D., and colleagues evaluated the concordance between the self-report and clinician-rated versions of the Inventory of Depressive Symptomatology (IDS-30) and between the two versions of the briefer Quick Inventory of Depressive Symptomatology (QIDS-16). Baseline and 12-month scores for 544 adult outpatients with psychotic or nonpsychotic depression were analyzed. Findings suggest that the self-report and clinician-rated versions of the IDS-30 and the QIDS-16 yielded highly similar overall ratings of symptom severity. There was also substantial agreement between versions about patients' response to treatment and remission of depression, regardless of whether a patient's depression had psychotic features (page 829).