Antiparkinson Drug Use and Adherence, and the Impact on Medicare Part D Enrollees with Parkinson's Disease

Study objectives. To examine the prevalence of and factors associated with Antiparkinson drug (APD) use and adherence, and the association of APD use and adherence with healthcare utilization and expenditures in Medicare Part D enrollees with Parkinson's disease (PD). Methods. A retrospective observational study was conducted using the 2006--2007 Medicare Chronic Care Condition Warehouse data which represents a 5 percent sample of Medicare beneficiaries. The PD sample was selected with (1) at least 1 medical claim with an ICD-9 code 332.0 in two consecutive years; and (2) continuous enrollment in Medicare Parts A, B, and D from 6/1/2006 through 12/31/2007 or date of death. The total study interval was 579 days. APD use measures included any use, duration, and adherence (Medication Possession Ratio [MPR]). Factors associated with APD use measures were examined using modified-Poisson regressions with Generalized Estimating Equations. The association of APD use/adherence with utilization and expenditure outcomes was evaluated with negative binomial and gamma General Linear Models, respectively. Results. 12% of PD patients (n=8,758) did not use any APDs, and a fourth of APD users had a duration of therapy for 435 days or fewer and an MPR of less than 0.80. Patients with cognitive impairment and with 11 or more comorbidities were less likely to use APDs; and if using, they were less likely to possess APDs persistently and regularly. Other factors associated with not using APDs included low-income-subsidy eligibility and having no neurologist visits. Factors associated with poor adherence included older age (65 or older), non-white ethnicity, and having changes in APD therapy. Longer duration and higher adherence were associated with a reduced rate of all-cause utilization for acute (hospital and emergency room [ER]), chronic (Part A skilled nursing facility [SNF] and home health agency), and physician care (only for adherence). Similar patterns were found with PD-only and PD-related-comorbidities hospital, ER, and Part A SNF care. Also, significantly reduced total, Part A, and Part B, and increased Part D expenditures were observed in longer-duration users and in higher adherers. Conclusion. Significant reduction in healthcare utilization and expenditures could be achieved by improved duration of use and adherence to APDs.