Public-Academic Partnerships: A Program to Improve the Quality of Antipsychotic Prescribing in a Community Mental Health System

In 2009, the medical director and the lead administrator of the Bureau of Behavioral Health (ER) initiated an antipsychotic quality improvement program. The program aimed to reduce use of antipsychotics with high likelihood of cardiometabolic side effects, reduce antipsychotic polypharmacy and use of antipsychotics for indications other than psychosis, and increase adherence to side effect-monitoring guidelines. The long-term goal of the program was to improve the physical health of patients with severe mental illnesses who are served in the state public mental health system.

Partnerships were established between the Bureau of Behavioral Health and leaders of the state psychiatric hospital (AD), the Medicaid program (DL), and the academic services research group (SJB) by meeting to discuss the public health problem and a strategy to address it. These leaders recognized the seriousness of the problem and readily partnered with the bureau. The previously established relationships between these leaders facilitated rapid buy-in.

The first strategy to address the problem involved educating community mental health leaders, other community stakeholders, and, later, all community mental health prescribers over several years. Second, a quality improvement team delivered in-person education to prescribers at each mental health center. The rationale for and details of the antipsychotic prescribing quality improvement program are described below.

To improve the quality of prescribing practices, the leaders considered using preferred-drug programs, prior authorization, academic detailing, and audit and feedback.

Medicaid and private insurance companies in most states use preferred-drug programs. These programs require clinicians to use medications considered preferred before using other medications within the same class, thereby shaping prescribing practices toward use of less expensive, but therapeutically similar or equivalent, agents. These programs are designed to contain costs while maintaining access to many or all medications (4).

Preferred-drug programs are used in conjunction with prior authorization, which requires formal approval to access a nonpreferred medication. When used to manage some classes of medications, such as nonsteroidal antiinflammatory medications, this strategy can significantly reduce costs (5), but the impact of such programs on antipsychotic prescribing is unclear. In fact, very few studies that assessed outcomes of such programs have been published. One study of such a program demonstrated that use of nonpreferred antipsychotics was reduced by up to 13.9% (6) but was associated with an unintended increase in antipsychotic treatment discontinuities among Medicaid recipients and did not change total pharmacy reimbursements (7,8). Prior authorization with step therapy—the required use of one medication before trying another—achieved the intended goal of reducing use of second-generation agents but was associated with increased outpatient treatment (9). The cost of increased outpatient treatment equaled the savings achieved by reducing use of second-generation medications.

Academic detailing, or educational outreach, entails visits by physicians or respected peers to counsel prescribers about the evidence for risks, benefits, and efficacy of medication alternatives (10). Studies have shown that this strategy can change prescribing practices to conform to evidence-based standards while maintaining the capacity for individualized treatment (11). Changes in prescribing have been found for about 5% of patients of general medical practices that have received academic detailing, although some studies demonstrated changes in prescribing practices that affected as many as 50% of patients (11). When delivered with additional intensive training, academic detailing reduced use of antipsychotics among elderly nursing home residents (12). In contrast, a recent study found that four in-person educational visits did not reduce antipsychotic polypharmacy in a Danish municipality, but the meetings were poorly attended (13).

Audit and feedback involves showing clinicians data on their prescribing patterns and comparing them with recommended practices. Multiple studies have demonstrated that this method reduced psychotropic medication polypharmacy (14,15). Education may also be effective, especially if the outcomes are perceived to be serious (16).

After reviewing these strategies, the team chose to implement a combination of approaches. Preferred-drug and prior authorization programs were already in place and were designed to leverage lower prices for medications while maintaining access. The team chose to avoid increasing the stringency of prior authorization or adding step therapy because these strategies could lead to interruptions in treatment or increases in service use and were universally unpopular with stakeholders. Instead, they developed a program that combined academic detailing with audit and feedback because these strategies were supported by the most promising evidence of efficacy in improving antipsychotic prescribing and also provided clinicians and patients with more flexibility and choice. The team planned to monitor the project's efficacy by analyzing Medicaid claims data and collecting data on the rates of office and laboratory monitoring of cardiometabolic side effects during annual reviews of community mental health center records.

The project was implemented on two levels. The first level of the initiative was an effort to achieve broad stakeholder engagement and education. As part of this process, the quality improvement team engages and educates state leaders, mental health center leaders, and consumer and family groups. They provide community mental health center leaders with access to lectures by experts and current data summaries of antipsychotic prescribing practices that allow them to compare practices at their own center with those of others. All community mental health prescribers are sent a quarterly letter describing the quality initiative and its goals and recommendations as well as articles relevant to the initiative. Additionally, the quality improvement group has partnered with the Medicaid program to send educational letters to medical professionals who have prescribed an antipsychotic without obtaining the recommended laboratory tests of cardiometabolic side effects.

The second level of the program consists of 45-minute, in-person educational visits to all of the prescribers within each community mental health center. A series of three visits in ten months incorporates academic detailing and audit and feedback strategies. The visits include a presentation describing current research evidence for antipsychotic treatment and providing information through summary tables from state Medicaid claims data about the prescribers' recent prescribing patterns as well as feedback about and reinforcement for changes made over time in their antipsychotic prescribing. Data tables about prescribing practices at the center allow prescribers to compare patterns in antipsychotic prescribing and monitoring among centers. Tables summarizing prescribing practices by individuals allow them to compare their practices with those of peers within their center. A part-time data analyst provides key technical support for this component of the educational intervention. The presenters give prescribers easy-to-use reference materials and encourage discussion about the site-specific prescribing data.

State mental health authorities and academic researchers may have differing goals. Mental health authorities often focus on limiting cost and adhering to regulations. Although they seek quality, improving quality in response to new information may be deferred or even defunded, particularly in challenging fiscal and political environments. Academic partners, on the other hand, often focus on developing and testing treatments.

Improving the quality of care is an area where mental health authorities and researchers may have mutual interests. Academic partners can bring energy, resources, and expertise to address new public health problems. By employing academic partners within DHHS, a public mental health authority may preserve its capacity for quality improvement and also maintain long-standing relationships that enable departments to collaborate to improve care.

In this case, the state Medicaid program had already implemented a prescription management strategy, through a preferred-drug list, that focused primarily on cost and control. In the absence of a public-academic partnership, it could have sought further restrictions to address the quality goals. Instead, the team chose to enhance quality by using a research-based strategy to increase prescriber knowledge and motivation while maintaining flexibility and choice for prescribers and patients and meeting the needs of the public mental health system to improve care.

State mental health authorities can use public-academic partnerships to create leadership roles in which leaders can track trends, identify new problems, and carry out quality improvement projects to address key issues. In this case, members of the partnership responded to growing concerns about early general medical morbidity and mortality among people with severe mental illnesses by initiating a program to improve antipsychotic prescribing.

This project was supported by the New Hampshire DHHS and by grant R01MH089811 to Dr. Bartels from the National Institute of Mental Health.

The authors report no competing interests.