• Steven Kleinman, Claire Cameron, Brian Custer, Michael Busch, Louis Katz, Boris Kralj, Ian Matheson, Ken Murphy, Jutta Preiksaitis, and Dana Devine.
• Department of Pathology & Laboratory Medicine, University of British Columbia, Vancouver, British Columbia, Canada.
• Transfusion. 2010 Dec 1; 50 (12): 2592-606.

BackgroundAs part of its risk management process, Canadian Blood Services (CBS) constructed mathematical models of how newly emerging pathogens might affect blood transfusion recipients.Study Design And MethodsCBS convened an expert panel including medical, health economics, analytical, risk management, and insurance professionals to examine multiple data sources. The model for emerging pathogen risk included separate modules to calculate the frequency and severity of infections from transfusion-transmitted agents that could cause either acute transient or chronic persistent infection. Important model input variables were annual number of components transfused, the presumed incidence and prevalence of a new agent, the time interval of recipient risk, recipient age and sex, projected recipient survival, rate of secondary infection, pathogen-induced morbidity, and the associated medical costs of such morbidity.ResultsIn the 5-year time frame considered in the model, it was estimated that approximately 3500 recipient infections (two-SD range of 0 to 11,370 infections) could occur from an emerging pathogen that establishes a chronic infection in donors, with 60% of these due to red blood cell transfusion. The medical costs associated with recipient outcomes due to a catastrophic emerging pathogen could be lowered by 20% if an effective pathogen reduction method for either platelets or plasma were in place.ConclusionThis modeling exercise offers a framework for other blood services to construct similar models. It also provides a useful way to model the implementation of new blood safety interventions (e.g., pathogen reduction) on emerging pathogen risk.© 2010 American Association of Blood Banks.

41 keywords...

hide…