State Mental Health Policy: Pharmacy Costs: Finding a Role for Quality

The Pennsylvania Medication Algorithm Project (PennMAP) was implemented at one of Pennsylvania's state hospitals (personal communication, Fiorello SJ, 2003) after reports of the increased use of combination antipsychotic therapy in the Pennsylvania state mental health system with little documented justification or rationale. This finding led to the implementation of PennMAP, which was based on the Texas Medication Algorithm Project, a best-practice consensus guideline. The goals of PennMAP included a guideline for clinical decision making, consistent treatment, improved patient outcomes, accurate documentation of care, implementation of new technologies, and management of patient costs. After the implementation of PennMAP, the percentage of patients taking two second-generation medications decreased from approximately 48 percent in September 2002 to 28 percent by August 2003 without any adverse clinical sequelae (personal communication, Fiorello SJ, 2003).

In Philadelphia, the Southeastern Pennsylvania Regional Pharmacy and Therapeutics Committee was established to safeguard patients from potentially adverse pharmacy practices and ensure quality-driven prescribing practices (3). The committee is composed of representatives from local and state offices of mental health and substance abuse, medical directors of the three behavioral health managed care organizations, medical directors and pharmacy directors of the three physical health maintenance organizations (HMOs), and six psychiatrists from the practicing community. The committee reviews formulary changes and acts as a forum to discuss HMO initiatives related to the authorization process for psychotropic medications. Thus far the committee has been able to address the problems of removing sertraline from one HMO's formulary, the occurrence of therapeutic duplication of second-generation antipsychotics, the uses and authorization of generic versus brand-name clozapine, the establishment of a policy for authorizing prescriptions for gabapentin by psychiatrists, and the development of a single preauthorization policy and form for the HMOs. These discussions foster the development of a best-practices model for the region, enabling the system to provide high-quality clinical care while containing costs.

In California, Kaiser-Permanente's approach to addressing the place of selective serotonin reuptake inhibitors (SSRIs) in its formulary (4) was to conduct an internal study of medication substitution of SSRIs. The four SSRIs most commonly prescribed at the time—citalopram, sertraline, fluoxetine, and paroxetine—were found to have an equal rate of substitution for another SSRI. Given this information and evidence showing that these four were comparable in efficacy, it was decided that it was appropriate for the cost of an SSRI to be a determinant in establishing a preferred medication list. When a study comparing fluoxetine, paroxetine, and sertraline did not show any significant difference in effectiveness, Kaiser implemented its "fluoxetine first" program, in which patients would be prescribed fluoxetine before other SSRIs. If fluoxetine was not effective, patients would be given a prescription for another SSRI. In one year, first-time SSRI prescriptions for fluoxetine rose from 30 to 80 percent. A thoughtful and systematic review of the literature and the development of an internal study to preserve the quality of care for patients make this a notable effort.

Massachusetts approached its Medicaid pharmacy benefits by using a psychopharmacology work group composed of psychiatrists and pharmacists to establish policy (5). If physicians were prescribing more than one SSRI or second-generation antipsychotic for 60 days or if five or more psychotropic medications were prescribed, an educational intervention was made, which included informing physicians of their practice and offering physicians a conference to educate them on issues relating to polypharmacy. The authors of the study projected substantial savings six months from initiation; however, no financial data were provided. Prior authorization procedures were streamlined, and an overall atmosphere of cooperative change was encouraged. It was noted that "stakeholder participation has improved the state's limit-setting policies and has thus far resulted in a relatively cooperative, litigation-free implementation process." It was the composition of the work group through which policy was implemented that was significant in this process.

In 2001 the VA treated 200,000 veterans with psychoses and filled 1.5 million prescriptions for antipsychotic medication valued at $158 million, 80 percent of which were for second-generation antipsychotics. This finding prompted the VA to initiate a utilization strategy for these medications (6). A task force of two VA psychiatrists and two VA pharmacists developed a guideline on the basis of a review of the literature. The VA task force decided to recommend a utilization strategy that used education to influence prescribing practices, allowing recommendations to be made that were both clinically appropriate and cost-effective. The General Accounting Office supported this approach, stating that the evidence base suggesting one second-generation antipsychotic might be more efficacious than another was equivocal; therefore, all second-generation antipsychotics may be regarded as clinically equivalent. It recommends that less expensive medications be tried first, unless there is a specific reason that a patient might need one of the more costly medications. Finally, the General Accounting Office emphasized that excessive pressure to contain cost is not an acceptable way to prescribe.

Missouri's plan to control pharmacy costs began in 2002, in partnership with Comprehensive NeuroScience, Inc., a company that promotes best clinical practices in the use of psychotropic medications in the public sector with special focus on Medicaid (personal communication, Parks JJ, 2003). The plan involves a quality management strategy in which pharmacy claims for behavioral medications are analyzed monthly and compared with quality indicators that are based on best-practice guidelines. These indicators include prescription of two or more second-generation antipsychotics to a patient concurrently; prescription of dosages of medication above or below recognized therapeutic dosages; prescription of three or more psychotropic drugs to a child concurrently; receipt of antipsychotic, sedative, or anxiolytic prescriptions from multiple prescribers concurrently; failure of patients to fill antipsychotic drug prescriptions in a timely fashion; and use of two or more psychotropic drugs from the same therapeutic class.

Prescribers of multiple psychotropic medications were made aware of their deviation from best-practice standards and given a packet of information that included the patient's pharmacy history, a Medication Best Practice Briefing monograph, and a report that compared their prescribing practices with those of their peers. If necessary, peer-to-peer calls to outlier prescribers within the community were made. Initial outcomes of the project from the first three quarters of 2003 (January to September) showed substantial changes in prescriber behavior after this intervention. Of the 391 physicians who prescribed outside the practice guidelines, more than half changed their prescribing patterns in the subsequent quarter. The study found that fiscal savings have exceeded the cost of the project; a detailed report describing the projected cost savings is in preparation.

Restricting formularies may not be a useful strategy for controlling the costs of psychotropic medication (7). Biological heterogeneity within mental disorders, the use of restrictive formularies to exclude patients with chronic mental illness, reduced pharmaceutical incentives to develop new drugs, and the characteristics of specific mental health institutions can affect how useful a restrictive formulary will be (7). The implementation of Medicare Part D of the Medicare Modernization Act is certain to test the effectiveness of restricted formularies in controlling costs, as dually eligible clients (those covered by both Medicaid and Medicare) who have previously had their pharmacy benefits covered by Medicaid are shifted to Medicare coverage. This shift has caused some Medicaid directors to wonder whether these individuals will actually lose some of their medication coverage (8).

Among the innovative strategies discussed, approaches differ in objective and locus of control. The initiatives of the Pennsylvania state system, Missouri, and the VA have as their primary goal improving or maintaining the quality of care, with cost saving as an anticipated secondary outcome. In contrast, Kaiser-Permanente's primary goal is to manage cost without decreasing quality of care. Philadelphia and Massachusetts negotiate pharmacy benefits and cost and patient care through a multidisciplinary work group or committee. Although these programs have not yet documented the actual cost savings, it is encouraging that states and health care organizations can promote thoughtful and quality driven prescribing practices.

Dr. Ning is affiliated with the department of psychiatry and behavioral health at Temple University Hospital-Episcopal Campus, 100 East Lehigh Avenue, Medical Arts Building, Suite 305, Philadelphia, Pennsylvania 19125 (e-mail, [email protected]). Dr. Dubin is with the Temple University School of Medicine in Philadelphia. Dr. Parks is with the Missouri Department of Mental Health in Jefferson City. Fred C. Osher, M.D., is editor of this column.