Estimating prevalence and financial impact of “bagging” oncology drugs: Evidence from privately-insured patients with cancer.

25 Background: “Bagging” is a drug delivery model that distributes physician-administered drugs through pharmacies, most specialty pharmacies. Under the traditional buy-in-bill model oncology drugs are covered in medical benefit, whereas drugs distributed through bagging practices are reimbursed under pharmacy benefit. Several professional associations have urged policymakers to limit or prohibit bagging practices, citing patient safety as the main concern. This study estimated the prevalence of bagging practice and the associated financial impact among cancer patients with private insurance. Methods: We selected 10 cancer drugs with highest total spending from the 2020 Medicare Part B Drug Spending Dashboard. Using 2019-2020 MarketScan data, we identified these drugs from medical and/or pharmacy claims and estimated the prevalence of “bagging” practice as the proportion of patients who had their cancer drugs covered by pharmacy benefit. We compared the per patient per month (PPPM) insurance payment and out-of-pocket expense (OOPE) between medical and pharmacy claims and tested the difference in PPPM costs using non-parametric Wilcoxon test. Results: The top 10 drugs selected for this study included four immunotherapy (pembrolizumab, nivolumab, durvalumab, atezolizumab), five targeted therapy drugs (rituximab, bevacizumab, daratumumab, trastuzumab, pemetrexed), and one supportive care product (pegfilgrastim). The study cohort consisted of 39,757 cancer patients, including 10,599, 19,276, and 17,882 patients (not mutually exclusive) receiving immunotherapy, targeted therapy agents, and supportive care drug, respectively. On average, “bagging” practice was observed in 4.4% of patients, ranging from less than 1% for immunotherapy and 2.4% for targeted therapy to 6.9% for supportive care drugs. The prevalence of bagging practice was higher in comprehensive plans (11.3%) and PPO (4.3%) versus other insurance plan types (3.8%). By census region, the prevalence was highest in Northeast (7.2%) and lowest in North Central (3.4%). Insurance payment PPPM was lower for drugs distributed through bagging practices than the conventional buy-and-bill model (mean $11,075 vs. $13,622; median $9,275 vs. $10,728, P < 0.001). OOPE PPPM exhibited the opposite pattern, with mean and median OPPE $437 and $53 for bagging practices and $186 and $0 for buy-and-bill. Conclusions: We observed wide variation in bagging practices by drug classes, plan types, and geographic regions. Evidence from the privately insured suggested that although bagging oncology drugs appeared to have reduce oncology drug costs for payors, this practice was associated with higher OOP for cancer patients with private insurance. Future research should examine the impact of bagging practices on clinical outcomes, such as treatment delays or disruptions and treatment-related side effects.