Texas Medication Algorithm Project

Abstract: Background This article describes the development of consensus medication algorithms for the treatment of patients with major depressive disorder in the Texas public mental health system. To the best of our knowledge, the Texas Medication Algorithm Project (TMAP) is the first attempt to develop and prospectively evaluate consensus-based medication algorithms for the treatment of individuals with severe and persistent mental illnesses. The goals of the algorithm project are to increase the consistency of appropriate treatment of major depressive disorder and to improve clinical outcomes of patients with the disorder. Method A consensus conference composed of academic clinicians and researchers, practicing clinicians, administrators, consumers, and families was convened to develop evidence-based consensus algorithms for the pharmacotherapy of major depressive disorder in the Texas mental health system. After a series of presentations and panel discussions, the consensus panel met and drafted the algorithms. Results The panel consensually agreed on algorithms developed for both nonpsychotic and psychotic depression. The algorithms consist of systematic strategies to define appropriate treatment interventions and tactics to assure optimal implementation of the strategies. Subsequent to the consensus process, the algorithms were further modified and expanded iteratively to facilitate implementation on a local basis. Conclusion These algorithms serve as the initial foundation for the development and implementation of medication treatment algorithms for patients treated in public mental health systems. Specific issues related to adaptation, implementation, feasibility testing, and evaluation of outcomes with the pharmacotherapeutic algorithms will be described in future articles.

Abstract: Background In the Texas Medication Algorithm Project (TMAP), detailed guidelines for medication management of schizophrenia and related disorders, bipolar disorders, and major depressive disorders have been developed and implemented. Discussion This article describes the algorithms developed for medication treatment of schizophrenia and related disorders. The guidelines recommend a sequence of medications and discuss dosing, duration, and switch-over tactics. They also specify response criteria at each stage of the algorithm for both positive and negative symptoms. The rationale and evidence for each aspect of the algorithms are presented.

Abstract: Background Medication treatment algorithms may improve clinical outcomes, uniformity of treatment, quality of care, and efficiency. However, such benefits have never been evaluated for patients with severe, persistent mental illnesses. This study compared clinical and economic outcomes of an algorithm-driven disease management program (ALGO) with treatment-as-usual (TAU) for adults with DSM-IV schizophrenia (SCZ), bipolar disorder (BD), and major depressive disorder (MDD) treated in public mental health outpatient clinics in Texas. Discussion The disorder-specific intervention ALGO included a consensually derived and feasibility-tested medication algorithm, a patient/family educational program, ongoing physician training and consultation, a uniform medical documentation system with routine assessment of symptoms and side effects at each clinic visit to guide ALGO implementation, and prompting by on-site clinical coordinators. A total of 19 clinics from 7 local authorities were matched by authority and urban status, such that 4 clinics each offered ALGO for only 1 disorder (SCZ, BD, or MDD). The remaining 7 TAU clinics offered no ALGO and thus served as controls (TAUnonALGO). To determine if ALGO for one disorder impacted care for another disorder within the same clinic ("culture effect"), additional TAU subjects were selected from 4 of the ALGO clinics offering ALGO for another disorder (TAUinALGO). Patient entry occurred over 13 months, beginning March 1998 and concluding with the final active patient visit in April 2000. Research outcomes assessed at baseline and periodically for at least 1 year included (1) symptoms, (2) functioning, (3) cognitive functioning (for SCZ), (4) medication side effects, (5) patient satisfaction, (6) physician satisfaction, (7) quality of life, (8) frequency of contacts with criminal justice and state welfare system, (9) mental health and medical service utilization and cost, and (10) alcohol and substance abuse and supplemental substance use information. Analyses were based on hierarchical linear models designed to test for initial changes and growth in differences between ALGO and TAU patients over time in this matched clinic design.

Abstract: This article provides an overview of the issues involved in developing, using, and evaluating specific medication guidelines for patients with psychiatric disorders. The potential advantages and disadvantages, as well as the essential elements in the structure of algorithms, are illustrated by experience to date with the Texas Medication Algorithm Project, a public-academic collaboration. Phase 1 entailed assembling research findings on the efficacy of medications for schizophrenic, bipolar, and major depressive disorders. This knowledge was evaluated for its quality and relevance, integrated with expert clinical judgment as well as input by practicing clinicians, family advocates, and patients. Phase 1 (the design and development of the algorithms) was followed by a feasibility test (Phase 2). Phase 3 is an ongoing evaluation comparing the clinical and economic effects of using specific medication guidelines (algorithms) versus treatment as usual in public sector patients with severe and persistent mental illnesses.