This presentation is part of: Poster Session - Utility Theory; Health Economics; Patient & Physician Preferences; Simulation; Technology Assessment
IMPACT OF GENERIC ENTRY ON COST-EFFECTIVENESS ANALYSES
Ya-Chen Tina Shih, PhD1, Shu Han, MS2, and Scott B. Cantor, PhD1. (1) The University of Texas M. D. Anderson Cancer Center, Biostatistics & Applied Mathematics, Section of HSR, Houston, TX, (2) Rice University, Department of Statistics, Houston, TX
Purpose: The market share of generic drugs has grown substantially since the passage of the Waxman-Hatch Act, increasing from 19% in 1984 to 50% in 2001. At entry, generics typically are priced lower than brand name drugs; thus, incorporating the impact of generic entry introduces an additional source of uncertainty in economic models as the timing of entry and the level of generic pricing are subject to variation. Our study explored the impact of generic entry on cost-effectiveness analyses. Methods: We developed a mathematical model that incorporated the possibility of generic entry. We constructed two examples using simulated data on costs and effectiveness. In the first example, we compared a new treatment with an existing treatment for a severe acute condition to examine the short-term effect of generic entry. In the second example, we compared a new drug with an old drug for a chronic condition by introducing variations in drug prices and the timing of generic entry to analyze the long-term effect of generic entry. To better assess the modeling uncertainty, we employed both a deterministic analysis and a Bayesian probabilistic approach and presented the results using cost-effectiveness acceptability curves. Results: The first example showed that the incremental cost-effectiveness ratio (ICER) increased as the price of the generic drug decreased. Corresponding to a societal willingness-to-pay of $50,000 per quality-adjusted life year (QALY), the probability that the new treatment was cost-effective was approximately 0.50 in the base case (i.e., no generic entry) and decreased as the annual cost of generics reduced, ranging from 0.45 at a price of $18,000 to 0.28 at $10,000. In the second example, the probability of an ICER less than $50,000 per QALY was 0.70 in the base case, but changed to 0.10 if the entry took place in year 3, 0.44 in year 5, and 0.64 in year 8 when the monthly drug cost was $600. If the monthly cost became $700, the probability was 0.18, 0.50, and 0.68 if entry occurred in year 3, 5, and 8, respectively. Conclusions: Failure to incorporate the impact of generic entry would underestimate the ICER, and thus, overstate the economic benefit of the new drug. Incorporating generic entry into pharmacoeconomic models would yield more accurate projections of the ICER and enhance decision making.
