OBJECTIVES: Specialty pharmacies employ multiple strategies to improve adherence to expensive self-injectable or oral medications (e.g., help apply for financial assistance, educate patients about toxicities, send refill reminders). While commercial enrollees are often required to receive specialty drugs through a specialty pharmacy, due to Medicare Part D’s Any Willing Pharmacy (AWP) provision, no such requirement is placed on Medicare beneficiaries. We evaluated the extent of specialty pharmacy use and its association with medication adherence in Medicare beneficiaries using oral anticancer medications (OAMs) for breast cancer and prostate cancer.

METHODS: Using SEER-Medicare data (2007-2019), we studied Medicare fee-for-service beneficiaries with Part D coverage who newly initiated OAMs for breast (palbociclib, lapatinib) or prostate cancer (abiraterone, enzalutamide). Primary outcomes were adherence (proportion of days covered [PDC]≥0.80 over 6 months post-initiation of OAM) and discontinuation (consecutive 90-day gap in OAM supply until end of follow-up). Logistic and Cox regressions controlling for characteristics were used to evaluate the association of specialty pharmacy use with adherence and discontinuation, respectively.

RESULTS: Our final sample included 20,114 beneficiaries (6,434 breast cancer and 13,680 prostate cancer); about one-quarter (27.1%) used specialty pharmacies for filling their OAM. Compared to beneficiaries initiating OAMs from non-specialty pharmacies, those who initiated OAMs through specialty pharmacies had higher odds of adherence (breast: odds ratio 1.17 [1.05-1.30]; prostate: OR 1.18 [1.09-1.28]) and lower hazard of discontinuation (breast: hazard ratio [HR] 0.89 [0.84-0.94]; prostate: HR 0.85 [0.82-0.89]). Results were consistent in subsamples of each individual OAM user.

CONCLUSIONS: We find that specialty pharmacy use is low in Medicare beneficiaries using OAMs for breast and prostate cancer. Specialty pharmacy use was associated with higher adherence and lower discontinuation for OAMs. Future research is needed to confirm similar findings in other diseases in the Medicare population and inform whether the Part D AWP provision should be reconsidered.