BACKGROUND : There has been continuing prog- ress in measures to reduce the risk of transfusion- transmitted infection, including introduction of serologic tests of increased sensitivity and the recent implementation of investigational NAT in small pools of samples.

STUDY DESIGN AND METHODS : Data relating to all blood donations to the American Red Cross have been consolidated into a single database. The prevalence of confirmed-positive test results for HBsAg, HCV, HIV, and HTLV were evaluated for each year for first-time donors from 1995 through 2001. Incidence rates for these infections were evaluated among repeat donors having at least two donations in a 2-year period. The frequencies of HIV-1 RNA- and HCV RNA-positive, seronegative donations were assessed for first-time and repeat donations. The relationship risk = (window period) × (incidence) was used to assess residual risk among repeat donations and to evaluate the incidence of HCV and HIV infection among first-time donors.

RESULTS : During the study period, prevalence rates for all markers declined significantly over time: in 2001, the rates per 100,000 were 75.6 for HBsAg, 299 for HCV, 9.7 for HIV, and 9.6 for HTLV; the corresponding incidence rates (/100,000 person-years) were 1.267, 1.889, 1.554, and 0.239, respectively. Estimates of residual risk in donations from repeat donors (after NAT) for HCV and HIV were 1 per 1,935,000 and 1 per 2,135,000, respectively. However, incidence rates for these agents are approximately two times greater among first-time donors. For both HCV and HIV, NAT yield was concordant with that predicted by current window-period models.

CONCLUSION : These data cover about half of all the whole blood collected in the United States. They suggest increasing improvement in transfusion safety and clearly define the benefit of pooled NAT.