Abstract
Objectives: Most cost-effectiveness analyses (CEAs) assume a drug's price remains unchanged following launch. Because prices typically change, this assumption can distort cost-effectiveness projections. Limited data on how prices change over a drug's life cycle have limited the inclusion of dynamic pricing assumptions. This study characterized changes in drug prices following launch and prior to loss of market exclusivity.
Methods: We analyzed inflation-adjusted price data for 32 brand-name drugs that contribute substantially to U.S. healthcare spending. We developed two regression models: (1) an ordinary least squares (OLS) regression model to estimate average annual price changes following launch, and (2) a linear mixed-effects regression to project how prices change over time. We identified independent factors potentially influencing price changes based on a literature review. We selected factors for model inclusion based on Akaike Information Criterion (AIC) improvements. Net price data came from SSR Health.
Results: The average inflation-adjusted mean annual drug price change was -4.7 percent (median: -2.4%). The OLS model predicted negative mean annual price changes for all combinations of drug characteristics except for drugs with Medicare-protected class designations (p-value <0.01), all else equal. Both models identified drug characteristics associated with smaller price declines or with price increases; the mixed-effects model indicated that price change rates tend to moderate with more time since launch.
Conclusions: For large-market branded drugs, inflation-adjusted prices often decline following launch and before loss of market exclusivity. Empirical modeling helps to refine projections based on observable characteristics, thus facilitating incorporation of dynamic pricing into CEAs.