Using Medicaid Claims and Dosing Guidelines to Assess the Prescribing and Outcomes of Conventional Antipsychotics

Abstract

Background: While newer atypical antipsychotics are considered the treatment of choice for schizophrenia, it is important to study dosing guidelines for conventional antipsychotics as they continue to be used alone or in combination with newer agents for patients who are refractory to treatment with newer agents. Few studies have used Medicaid claims for this purpose, even though they provide a readily available and objective means of assessing dosing in relation to guidelines and outcomes of prescribing. Objective: To examine a new multi-method approach for studying dosing guidelines and outcomes of prescribing in Medicaid recipients receiving conventional antipsychotics. Methods: Data were collected from medical records, Medicaid prescription and hospital claims, and case managers in 8 county-based delivery systems in Wisconsin during a 12-month period in 1989–1990 as a part of a larger naturalistic study. Eligible patients had a primary diagnosis of schizophrenia, a prescription for oral antipsychotics, and no hospitalizations during baseline (first quarter). Results: Of 159 patients, 46.5%, 32.1%, and 21.4% had baseline antipsychotic doses below, above, and within the guidelines, respectively. Patients dosed above the guidelines were 4 times more likely to be hospitalized after baseline (p < 0.05) than those within the guidelines after controlling for adherence, baseline functioning, and past hospitalizations. Further analysis showed that the high-dose/hospitalization relationship was mediated or explained in part by adverse effects. Conclusions: The use of Medicaid claims revealed a substantial amount of prescribing outside of guidelines and an increased risk of hospitalization during the study period. Future trials are needed to compare this methodologic approach with other approaches and to determine whether current dosing patterns and outcomes have changed. This study was made possible through Rickles' National Institute of Mental Health (NIMH) Fellowship in Mental Health Services Research at the University of Wisconsin — Madison. The study obtained data from a larger project supported by NIMH grant MH-43555 to the late Dr. James Greenley and his colleagues at the University of Wisconsin — Madison. Data from this study were presented in March 2000 at the 147th Annual Meeting and Exposition of the American Pharmaceutical Association.