Drug Cost Containment Changes and Quality of Care for Mentally Ill Dual Enrollees

点击翻译按钮获取中文摘要

DESCRIPTION (provided by the applicant): Disabled, low income Americans who receive both Medicaid and Medicare insurance coverage (i.e., dual enrollees), represent one of the fastest growing segments of the Medicare population. About one-third of these beneficiaries (est. 2.5 million) have expensive, often debilitating mental illness, including schizophrenia and bipolar disorder. Poor drug treatment adherence among these dual enrollees leads to worse health and expensive downstream clinical events including hospitalizations for severe mental illness (SMI). The Medicare Modernization Act of 2003 (MMA) transferred the responsibility for outpatient prescription drug coverage for dual enrollees from individual state Medicaid programs to private Part D plans funded by the Medicare program on January 1, 2006. The Centers for Medicare and Medicaid Services (CMS) randomly assigns dual beneficiaries to Part D Prescription Drug Plans (PDPs) with relatively low premiums. These standalone PDPs can vary in the numbers and types of prescription drugs included in the plan formulary and can employ utilization management approaches such as prior authorization (PA) for any drug prescriptions. Because each state determined its own Medicaid coverage policies, dual beneficiaries faced a range of drug coverage benefits prior to their transition to Part D coverage. For example, in 15 states, dual enrollees had caps on the number of prescriptions. Thus, the transition to Part D in 2006 expanded their drug coverage and removed a potential risk factor for costly adverse health events. Most states, however, excluded antipsychotic and anticonvulsant therapy from Medicaid prior authorization (PA), while PDPs often use PA as the primary approach for managing drug costs. Thus, as drug coverage responsibility shifts from states to private plans, dually enrolled beneficiaries face fewer state-determined cost barriers to outpatient prescription drug access, but potentially more plan-determined administrative barriers to psychotropic drug access. We will analyze the impact of Part D separately in four large and geographically diverse states: two that placed caps on the number of prescriptions and had relatively higher copayments between 2004 and 2007 (South Carolina: limit of 4 prescriptions per month; California: limit of 6 per month), and two that had no caps and relatively lower copayments during the same period (Missouri, New Jersey). In Aim 1, we will use a strong quasi-experimental design, Multiple Interrupted Time Series, to examine the population level impacts of this transition on: (1) the prevalence and persistence of psychiatric medication use; (2) use of non-drug psychiatric services; and (3) costs. In Aim 2 we use a randomized design at the patient level to estimate the effect of coverage restrictions on use of psychiatric medications (including discontinuation and switching of medications), psychiatric outpatient and ER visits, and costs. In Aim 3 we will use both designs to examine the impact of the policy changes among at-risk subgroups (defined by somatic comorbidities and minority status) of patients with schizophrenia and bipolar disorder.