Retrospective assessment of Medicaid step-therapy prior authorization policy for atypical antipsychotic medications

Background: Antipsychotic medications account for more prescription expenditures in Medicaid than any other therapeutic category. This has made them an attractive target for states hoping to curtail rising expenditures. Objective: The objective of this study was to document the effects of a step-therapy prior authorization (PA) policy for atypical antipsychotic medications on: (1) Medicaid prescription expenditures among all Medicaid beneficiaries and (2) prescription and health service expenditures among patients with schizophrenia. Methods: Prescription, inpatient, outpatient, and long-term care State Medicaid Research Files from Georgia and Mississippi from January 1, 1996, to December 31, 1997, were used to model an interrupted time-series analysis. We compared a step-therapy PA policy implemented in Georgia to a nonequivalent/no-treatment control group (Mississippi) over 10-month prepolicy, 11-month policy, and 3-month postpolicy periods. Segmented regression was used to estimate antipsychotic prescription expenditures among all eligible Medicaid beneficiaries. We used generalized estimating equations to model prescription and other health service expenditures with difference-indifference regressions among a cohort of patients with schizophrenia. Results: Compared with Mississippi, Georgia saved ~$7 million in atypical antipsychotic expenditures over the 11-month policy period. Among patients with schizophrenia, the PA policy was associated with a $19.62 per member per month (PMPM) decrease in atypical antipsychotic expenditures and a $2.20 PMPM increase in typical antipsychotic expenditures (both, P < 0.001). Among the same patients with schizophrenia however, the reduction in atypical antipsychotic expenditures was accompanied by a $31.59 PMPM increase in expenditures for outpatient services (P < 0.001). Conclusion: Although PA of atypical antipsychotics was associated with significant prescription savings to the Georgia Medicaid program, among a vulnerable cohort of patients with schizophrenia, an increase in outpatient expenditures was associated with overall savings.