Abstract
Background: Between 1987–1997, the demand for allogeneic blood (whole blood and packed RBC) in the US decreased due to concerns about safety of the blood supply1. Paradoxically, these concerns also led to a decrease in the available supply. During this same period, collection of allogeneic blood declined from 13.6 million to 11.9 million units, resulting in a 48% reduction in the margin between available supply and demand.1 Although both supply and demand for blood has since increased, the margin has further declined to only 6.1% in 2004 (allogeneic collection: 14.8 million units; transfusion: 13.9 million units)2. This situation is further exacerbated by procedures used for qualifying fully screened units. In 2004, 240,000 units were rejected after screening, leaving a margin of only 648,000 units available (4.5% of the supply). Given the introduction and adoption of treatments that reduce transfusions, such as erythropoiesis-stimulating agents (ESAs) for patients (pts) with chemotherapy-induced anemia (CIA), in the past 15 years, periodic shortages in the blood supply due to this limited surplus have been moderated. Purpose: To estimate the impact of limiting the use of ESAs for CIA on the US blood supply.
Methods: A modeling simulation was employed using a top-down approach to compare the number of RBC units transfused in ESA-treated pts to the number of RBC units that would be transfused if ESAs were discontinued or limited in the same population. The excess number of RBC units that would be required if ESA treatment in CIA pts was limited was contrasted with the available marginal blood supply from 2004 (latest data available). Model inputs included incident cases of CIA pts treated with ESA, transfusion rates from clinical trials, and volume of RBC units required for ESA-treated and untreated pts. Data were obtained from published literature or expert opinion where published evidence was unavailable. Estimates were developed for multiple ESA reduction scenarios, and sensitivity analyses were conducted using a range of +/–10% for each input parameter.
Results: Under the base case scenario, it was estimated that 492,002 incident CIA pts received a total of 372,809 RBC units despite ESA treatment. The model predicted that up to a third of the marginal US blood supply would be required to cover the incremental demand for blood that would arise from a 25% decrease in ESA use (incremental RBC units transfused: 118,602 units, sensitivity range: 63,030–210,110 units). For ESA use reductions of 50% and 75%, the model predicted 37% (237,203 units) and 55% (355,805 units) of the marginal US blood supply would be required, respectively. In the case of total cessation of ESA use for CIA, the available US blood supply could be exceeded (incremental demand: 474,407 units, sensitivity range: 252,119–840,441 units).
Conclusions: This current model of blood demand showed limiting ESA use in CIA pts imposes considerable pressure on the available US blood supply given the small margin between usable blood and transfusion demand. This added pressure on the blood supply does not consider additional exacerbations due to regional and seasonal variation in the number of available units as well as donation frequency variations.
Modeling the Population Health Impact of ENDS in the U.S.
