Is There A Patent Expiration Dividend for Medicare Part D: Generic Second-Generation Antipsychotics, Drug Spending, and Patient Adherence

Background: Spending on antipsychotic drugs has grown rapidly, largely driven by use of second-generation antipsychotics (SGAs). Medicare Part D drug benefits include formulary protections for antipsychotics intended to provide access to all drugs in the class; these protections, however, do not address patient cost-sharing, which can be considerable and has been associated with worse adherence and clinical outcomes among beneficiaries using antipsychotics. The introduction of generic SGAs, starting with risperidone in July 2008, could reduce antipsychotic drug spending and cost-related use barriers. Little is known about the patterns of generic take-up or the effects of generic use on spending or adherence.

Objective: To examine patterns of generic risperidone use, antipsychotic spending and adherence during the initial 18-months after introduction among Medicare Advantage (MA) Part D plans.

Methods: The study included 15,037 beneficiaries receiving ongoing SGA treatment (≥1 SGA fill in 2007 and Jan-Jun 2008). We differentiated between bioequivalent substitution (brand to generic risperidone) and substitution within the therapeutic class (non-risperidone SGA to generic risperidone). We used Cox proportional hazards models to examine plan and patient characteristics associated with time to generic risperidone use and linear models to examine differences in antipsychotic spending and adherence (proportion of days covered, PDC) in 2009 for those with versus without generic use, adjusting for patient and plan characteristics and prior year antipsychotic use (PDC). We stratified analyses for beneficiaries receiving low-income subsidies (LIS), which substantially reduce Part D cost-sharing, and non-LIS beneficiaries.

Results: Overall, the mean age was 60.3 years, 39.0% had schizophrenia or bipolar disorder diagnoses, and 48.7% received the LIS. Prior to July 2008, 24.0% of subjects last used branded risperidone and 76.0% used other SGAs. Among those with any antipsychotic use between July 2008-December 2009, 24.9% of subjects had ≥1 fill of generic risperidone: 85.5% of those previously using branded risperidone and 6.1% of those previously using other SGAs. Among LIS and non-LIS beneficiaries, enrollment in private fee-for-service (PFFS) versus HMO plans was associated with lower rates of generic use (HR=0.81, 95%CI: [0.70, 0.95], non-LIS). Among non-LIS beneficiaries, plan generosity (i.e., enhanced vs. basic benefits) and formulary tiering (i.e., tier copayment differentials) were not significantly associated with generic use. Compared with beneficiaries who continued using other SGAs, those who switched from non-risperidone SGAs to generic risperidone had lower out-of-pocket (OOP) spending (-$162 [-$240, -$84], non-LIS) and higher levels of adherence (10.8 percentage points [7.6, 13.9]) in 2009, as did those who switched from brand to generic risperidone (OOP: -$444 [-$489, -$400], PDC: 8.9pp [7.1, 10.7], non-LIS).

Conclusions: Generic risperidone use was associated with lower costs and better adherence, but levels of use were modest and concentrated among patients previously using branded risperidone. HMO plans appeared to be more effective at encouraging generic use than PFFS plans, which until 2011 were not required to have provider networks. Patient financial incentives, however, had limited influence on drug choice among those on existing treatment. Considerable opportunity for cost savings remains for both Medicare and beneficiaries, especially with the recent introduction of additional generic SGAs.