Faculty Bibliography

OBJECTIVE: This study sought to identify and characterize rates of clinically questionable prescribing in New York State. METHODS: As part of a quality improvement initiative, 34 national psychopharmacology experts identified a set of questionable prescribing practices recognizable from pharmacy claims data. Indicators of such practices were applied to Medicaid claims data for 217,216 beneficiaries in New York State who had an active psychotropic prescription on April 1, 2008. RESULTS: A total of 156,103 (72%) of these beneficiaries had one or more continuing (>90 days) prescriptions for a psychotropic. About 10% of adults were prescribed four or more psychotropics concurrently, and 13% of children and 2% of older adults were prescribed three or more concurrently. Prescribing an antipsychotic with a moderate-to-high risk of causing metabolic abnormalities approached 50% (46%) among individuals who had existing cardiometabolic conditions. Among beneficiaries prescribed second-generation antipsychotics with a moderate-to-high risk of causing metabolic abnormalities, over half (60%) had not received a metabolic screening test in the past year. Among women of reproductive age prescribed mood stabilizers, over one-quarter (30%) were prescribed a valproic acid-based formulation despite its potential for teratogenicity. Only 2% of youths under age 18 were prescribed benzodiazepines; however, about half (48%) had trials over 90 days' duration. CONCLUSIONS: Examination of pharmacy claims from Medicaid beneficiaries in New York State indicated that prescribing practices deemed clinically questionable by pharmacology experts are common. Aggregated pharmacy claims data can identify such practices, and reviews of these data can be a core component of efforts to improve prescribing practices.

This paper describes the psychometric properties of two fidelity scales created as part of the Substance Abuse and Mental Health Services Administration (SAMHSA) sponsored medication management toolkit and their metric properties when used in 26 public mental health clinics with 50 prescribers. A 23-item scale, based on chart reviews, was developed to assess whether prescribers are following good medication practices, in conjunction with a 17-item scale to assess organizational support for and evaluation of prescriber adherence to recommended medication-related practices. Fundamental gaps in routine practice, including poor documentation of medication history and infrequent monitoring of symptoms and side effects were found.

State mental health authorities have a leadership role in implementing evidence-based practices (EBPs), but few instruments are available to assess the impact of this role. We describe the development of the State Mental Health Authority Yardstick (SHAY), a behaviorally anchored instrument designed to assess state-level facilitating conditions associated with successful implementation of EBPs in community mental health centers. The SHAY assesses the SMHA role in seven domains: Planning, Financing, Training, Leadership, Policies and Regulations, Quality Improvement, and Stakeholders. Preliminary evidence from the National Evidence-Based Practices Project partially supports the construct and criterion-oriented validity of this instrument for rating state-level activities supporting or blocking the implementation of evidence-based practices.

OBJECTIVE: This study identified barriers and facilitators to the high-fidelity implementation of assertive community treatment. METHODS: As part of a multistate implementation project for evidence-based practices, training and consultation were provided to 13 newly implemented assertive community treatment teams in two states. Model fidelity was assessed at baseline and at six, 12, 18, and 24 months. Key informant interviews, surveys, and monthly on-site visits were used to monitor implementation processes related to barriers and facilitators. RESULTS: Licensing processes of the state mental health authority provided critical structural supports for implementation. These supports included a dedicated Medicaid billing structure, start-up funds, ongoing fidelity monitoring, training in the model, and technical assistance. Higher-fidelity sites had effective administrative and program leadership, low staff turnover, sound personnel practices, and skilled staff, and they allocated sufficient resources in terms of staffing, office space, and cars. Lower-fidelity sites were associated with insufficient resources, prioritization of fiscal concerns in implementation, lack of change culture, poor morale, conflict among staff, and high staff turnover. In cross-state comparisons, the specific nature of fiscal policies, licensing processes, and technical assistance appeared to influence implementation. CONCLUSIONS: State mental health authorities can play a critical role in assertive community treatment implementation but should carefully design billing mechanisms, promote technical assistance centers, link program requirements to fidelity models, and limit bureaucratic requirements. Successful implementation at the organizational level requires committed leadership, allocation of sufficient resources, and careful hiring procedures.

The authors discuss the implications of the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) for mental health services researchers, in particular the need to monitor and change prescriber behavior to encourage informed medication selection. Given the complexity and variability of response to antipsychotic medications in CATIE, use of restricted formularies is not the answer. Rather, services researchers should collaborate with service systems to develop interventions to identify questionable prescriber practices and develop interventions to change them. The CATIE results also suggest that some quality measures for antipsychotic treatment, such as the proportion of a population taking second-generation antipsychotics, need to be revisited by researchers. Also, because the CATIE findings highlighted the prevalence of cardiac and metabolic disorders among treatment populations and the potential impact of antipsychotics on these conditions, services researchers should use secondary data to monitor whether prescribers are providing appropriate screening and treatment. Given the health risks of some antipsychotics, services researchers should develop ways to identify individuals at risk, encourage behavior change among prescribers, and support informed and shared decision making about medications. Mental health services researchers can build relationships with multiple stakeholders, including service system administrators, service providers, and consumers, to help translate results from trials such as CATIE into policy and practice.

BACKGROUND: A panel of academic psychiatrists and pharmacists, clinicians from the Texas public mental health system, advocates, and consumers met in June 2006 in Dallas, Tex., to review recent evidence in the pharmacologic treatment of schizophrenia. The goal of the consensus conference was to update and revise the Texas Medication Algorithm Project (TMAP) algorithm for schizophrenia used in the Texas Implementation of Medication Algorithms, a statewide quality assurance program for treatment of major psychiatric illness. METHOD: Four questions were identified via premeeting teleconferences. (1) Should antipsychotic treatment of first-episode schizophrenia be different from that of multiepisode schizophrenia? (2) In which algorithm stages should first-generation antipsychotics (FGAs) be an option? (3) How many antipsychotic trials should precede a clozapine trial? (4) What is the status of augmentation strategies for clozapine? Subgroups reviewed the evidence in each area and presented their findings at the conference. RESULTS: The algorithm was updated to incorporate the following recommendations. (1) Persons with first-episode schizophrenia typically require lower antipsychotic doses and are more sensitive to side effects such as weight gain and extrapyramidal symptoms (group consensus). Second-generation antipsychotics (SGAs) are preferred for treatment of first-episode schizophrenia (majority opinion). (2) FGAs should be included in algorithm stages after first episode that include SGAs other than clozapine as options (group consensus). (3) The recommended number of trials of other antipsychotics that should precede a clozapine trial is 2, but earlier use of clozapine should be considered in the presence of persistent problems such as suicidality, comorbid violence, and substance abuse (group consensus). (4) Augmentation is reasonable for persons with inadequate response to clozapine, but published results on augmenting agents have not identified replicable positive results (group consensus). CONCLUSIONS: These recommendations are meant to provide a framework for clinical decision making, not to replace clinical judgment. As with any algorithm, treatment practices will evolve beyond the recommendations of this consensus conference as new evidence and additional medications become available.

OBJECTIVE: This article presents fidelity outcomes for five evidence-based practices that were implemented in routine public mental health settings in the National Implementing Evidence-Based Practices Project. METHODS: Over a two-year period 53 community mental health centers across eight states implemented one of five evidence-based practices: supported employment, assertive community treatment, integrated dual disorders treatment, family psychoeducation, and illness management and recovery. An intervention model of practice dissemination guided the implementation. Each site used both human resources (consultant-trainers) and material resource (toolkits) to aid practice implementation and to facilitate organizational changes. External assessors rated fidelity to the evidence-based practice model every six months from baseline to two years. RESULTS: More than half of the sites (29 of 53, or 55%) showed high-fidelity implementation at the end of two years. Significant differences in fidelity emerged by evidence-based practice. Supported employment and assertive community treatment had higher fidelity scores at baseline and across time. Illness management and recovery and integrated dual disorders treatment had lower scores on average throughout. In general, evidence-based practices showed an increase in fidelity from baseline to 12 months, with scores leveling off between 12 and 24 months. CONCLUSIONS: Most mental health centers implemented these evidence-based practices with moderate to high fidelity. The critical time period for implementation was approximately 12 months, after which few gains were made, although sites sustained their attained levels of evidence-based practice fidelity for another year.

The New York State Office of Mental Health has implemented the browser based Psychiatric Clinical Knowledge Enhancement System (PSYCKES) medication application throughout its inpatient system of care. PSYCKES provides detailed current medication regimens and histories, as well as medication best practices reports at the patient, psychiatrist, ward, and facility levels. South Beach Psychiatric Center (SBPC) has made specific use of a best practices report which details proportions of caseloads and number of patients on two or more concurrent antipsychotic medications. Psychiatrists received extensive application and desktop support and individual and group training was conducted. PSYCKES current and historical medication regimens were reviewed for individual cases and best practices reports were used in psychiatry supervision from January 2004 to March 2005. SBPC psychiatrists achieved marked reductions in the proportions of their caseloads on two or more concurrent antipsychotics. Although we cannot impute causality from this pre-post implementation design, the results of this evaluation suggest that the use of PSYCKES was effective in reducing such regimens for inpatients.