"And while that may be a small fraction of the hospital's costs, in this day—when we're looking to get the most value from our healthcare system—we're looking at cost-benefits and every penny counts," says Frank, lead author of the paper with colleagues at Johns Hopkins and Brigham and Women's Hospital, in this week's issue of the journal Anesthesiology.
Frank's project used electronic records to examine blood utilization data for 53,526 patients who underwent 1,632 procedures at Johns Hopkins between January, 2010 and March, 2012. The algorithm has not yet been tested in clinical practice yet, however.
The team first looked at those cases that had a type and screen and/or a type and cross for donated units, and then calculated how many times the patients those units were prepared for actually required a transfusion.
Take, for example, thyroid surgeries. According to the "unique" algorithm Frank and his colleagues developed, only two in 1,000 such procedures at Johns Hopkins required the use of blood. Even in those few cases, no patient would die on the table for lack of blood because so called "emergency release" type O blood can be used instead.