Outcome-Based Pricing for New Pharmaceuticals via Rebates

Abstract

The price of new brand-name prescription drugs has been rising fast in the United States. For example, the Amgen cholesterol drug Repatha had an initial list price of $14,523 per year. Patients, even with insurance coverage, must pay out of pocket a significant portion of this price. The treatment might not be successful, and this possibility reduces risk-sensitive patients’ incentives to purchase the drug. The high price together with the chance of negative treatment outcomes may lead payers to deny coverage for the drug. Outcome-based pricing has been proposed as a way to reallocate the risks and improve both payer resource allocation and patient access to drugs. According to an outcome-based rebate contract between Amgen and Harvard Pilgrim Healthcare, if a patient on Repatha suffers a heart attack or a stroke, both patient and insurer are refunded the cost of the drug. We use a stylized model to analyze the effect of outcome-based pricing via rebates. Our model captures the interaction between heterogenous, price-sensitive, risk-sensitive patients who decide whether to purchase the drug; a payer deciding whether to provide coverage for the drug; and a price-setting pharmaceutical firm seeking to maximize expected profits. We find that, in many cases, a pharmaceutical firm and payer cannot simultaneously benefit from outcome-based pricing, and who will benefit is determined by the probability of treatment success. Outcome-based pricing thus appears unlikely to solve the issues of high drug prices and high payer expenditures. However, supplementing outcome-based pricing with a transfer payment from firm to payer can make payer and firm (but not necessarily the patients) better off than under uniform pricing when the drug has a low chance of success. This paper was accepted by Stefan Scholtes, healthcare management.

Improving Access to Rare Disease Treatments: Subsidy, Pricing, and Payment Schemes

“…Other research in this stream focuses on examining the performance of outcome-based payment in the context of healthcare services, such as referrals and outpatient medical services (Andritsos andTang 2018, Adida andBravo 2019). In a recent healthcare operations paper, Adida (2021) analyzes the effect of outcome-based pricing on patients' welfare, payer's payoff, and pharmaceutical firm's profit. She finds that the pharmaceutical firm and payer cannot simultaneously benefit from outcomebased pricing.…”

Developing operations management data analytics

“…In these studies, the input–output relationship is often modeled at a high level with time aggregation, without the detailed structures to enable decision optimization. In the second stream of healthcare research, analytical models are formulated and techniques including stochastic program, approximate dynamic program, Bayesian dynamic program, queuing analysis, and equilibrium analysis are used to derive the decisions (recent examples include Adida, 2021; Ahuja et al., 2021; Ata et al., 2020; Bavafa et al., 2022; Carew et al., 2021; Slaugh et al., 2018; Tian et al., 2022). In these studies, data are often not explicitly modeled as an input to the decision models, though some demonstrate how to apply the model to real data (e.g., Aswani et al., 2019).…”

Converting a drug from off‐label to on‐label use: Government subsidies and patient welfare

“…To model the patient's purchasing decision, we consider a utility function that is based on the Willingness‐To‐Pay model with uniform valuation (e.g., Adida, 2021; Yu et al., 2018). If the off‐label drug (or on‐label drug if there is drug conversion) is successful, patients live and enjoy an “extra” utility V .…”

“…To capture patient heterogeneity, we assume V is a random variable uniformly distributed in the range $$[0,\bar{v}]$$ (i.e., $$V \sim U [0,\bar{v}]$$). The realization of V may depend on several factors, such as illness severity, budget of patients, opportunity cost of being ill, demographics, and remaining quality‐adjusted life years (Adida, 2021). For any realized value of V , the patient decides to purchase the off‐label drug (or on‐label drug if there is drug conversion) by taking the effective rate $${e_{\text{f}}}$$ (or $${e_{\text{n}}}$$) and the selling price $${p_{\text{f}}}$$ (or $${p_{\text{n}}}$$) into consideration.…”