Steady-state calculation of the risk of HIV infection from transfusion of screened blood from repeat donors (Q1361486)

Although all blood donations in the United States are tested for antibodies to human immunodeficiency virus (HIV), the virus that causes AIDS, there remains a small risk of HIV transmission through blood transfusion. The primary source of this transmission is recently infected donors who are infectious but have not yet developed the antibodies that the screening tests detect. We describe the methods that we recently used [\textit{E. M. Lackritz}, the author et al., N. Engl. J. Med. 333, 1721-1725 (1995)] to estimate the probability of transfusion-related HIV transmission. The goal of this analysis is to develop methods to estimate the risk of HIV transmission with the use of surveillance data available from the American Red Cross (ARC). By using simple steady-state arguments and by accepting upper-bounded limits on some probabilities required for our calculation, we were able to estimate the number of potentially infectious units of blood donated without using longitudinal data on individual blood donors. This ability is important because blood donors have a very low incidence of HIV: an estimated 4.1 new HIV infections per 100,000 person years of observation. Thus, a very large cohort of blood donors would be required to obtain an adequate sample of seroconverting blood donors. We have focused our analysis on repeat blood donors for two reasons. First, because approximately 75\% of blood donations are made by repeat donors, they account for much of the risk of HIV transmission. Second, blood bank data cannot be used to estimate HIV incidence among first-time donors. In Section 2, we develop a probability model to calculate the expected number of HIV transmissions from a single repeat blood donor who becomes HIV positive. In Section 3, we develop estimators of the quantities required in this model that use surveillance data on blood donations. In Section 4, we compare the risk of HIV transmission from first-time and repeat donors, and in Section 5 we discuss our results.