Modeling Potential Demand for Supply-Constrained Drugs: A New Hemophilia Drug at Bayer Biological Products

This paper describes the evolution and application of a novel approach for forecasting drug demand in markets where supply limitations have significantly curtailed sales volumes and thus reduced the usefulness of conventional sales-based forecasting methods. This occurs frequently with biological (biotech) drugs. We use methods from decision analysis to explicitly model the variability in epidemiological data together with the variability in treatment modalities to estimate latent therapeutic demand (LTD)—the underlying demand that captures how physicians would prescribe treatment and how patients would comply if ample supplies of drugs were available and affordable. Our approach evolved from efforts to help Bayer Biological Products with strategic decisions regarding its drug for treating hemophilia A, the future of which had been clouded for several years, primarily due to a lack of confidence in demand estimates. Use of the LTD model resulted in a better understanding of the therapeutic needs of the global hemophilia community and helped Bayer make good decisions. We believe this approach is widely applicable to forecasting potential demand for supply-constrained as well as brand-new drugs, and thus can be very useful in helping both drug manufacturers and health-care agencies worldwide to ensure adequate supplies of critical drugs.