Abstract
PURPOSE: To evaluate the association between industry payments to physicians related to poly (ADP-ribose) polymerase inhibitors (PARPis) and physicians' prescribing behaviors for PARPis. METHODS: This panel-data analysis used the publicly accessible Open Payments Database and Medicare Part D database between 2017 and 2021. All physicians who reported >10 claims for either olaparib, rucaparib, or niraparib were included in this study. Non-research payments for the PARPis to the physicians from the PARPi manufacturers were extracted from the Open Payments Database. Associations between the physicians' receipt of payments and likelihood of prescribing PARPis were assessed with logistic generalized estimating equations (GEEs). Dose-response associations between the number of payments and prescription volumes and Medicare expenditures were evaluated with linear GEEs. RESULTS: Of the 1686 eligible physician prescribers, 68.7% received one or more non-research payments related to any of the three PARPis from the manufacturers between 2017 and 2021. Median annual payments per physician were $57 for olaparib, $39 for rucaparib, and $62 for niraparib. Receipt of payments for each PARPi was associated with higher odds of prescribing olaparib (odds ratio [OR]: 1.30 [95% CI: 1.14-1.48], p < 0.001), rucaparib (OR: 2.07 [95% CI: 1.58-2.72], p < 0.001), and niraparib (OR: 1.49 [95% CI: 1.22-1.81], p < 0.001). Dose-response effects were observed between the number of annual payments and the number of prescriptions and/or Medicare expenditures for olaparib and rucaparib. CONCLUSION: Non-research payments to physician prescribers of PARP inhibitors from the manufacturers were significantly associated with increased prescriptions and Medicare expenditures for olaparib and rucaparib in the United States.