Context: An understanding of characteristics of blood donors donating in times of crisis may help predict blood supply safety and donor return patterns. Objectives: To characterize the volume of donations and prevalence of infectious disease markers in blood donated by US donors responding to the September 11, 2001, terrorist attacks, and to evaluate return rates in those who donated for the first time. Design Cross-sectional survey data from the National Heart, Lung, and Blood Institute Retrovirus Epidemiology Donor Study for 4 weeks before and 4 weeks starting with September 11, 2001, and the corresponding 8-week period in 2000. Setting and Participants: A total of 327065 volunteer blood donors making 373 628 allogeneic donations at 5 large regional US blood centers. Main Outcome Measures: Changes in number of donations overall and by first-time and repeat status, prevalence of infectious disease markers, estimated risks of transfusion-transmitted viral infections, and first-time donor return rates. Results: About 20000 allogeneic donations were collected weekly in the 4 weeks preceding September 11, whereas ≈49000 (2.5-fold increase) and ≈26000 to 28000 (1.3-fold to 1.4-fold increases) donations were made per week in the first and in the second through fourth weeks starting with September 11, respectively. All demographic groups donated more than usual after the attacks, and after adjusting for seasonal and annual variation there was a 5.2-fold (95% confidence interval, 5.0-5.4) increase in the number of first-time donations vs a 1.5-fold (1.4-1.5) increase in the number of repeat donations made in the first week starting on September 11 vs the 4 weeks before. The weekly proportion of repeat donors returning after not donating for 10 or more years increased from 2% before September 11 to 6% in the first week starting with September 11. Donations confirmed positive for human immunodeficiency virus (HIV), hepatitis C virus (HCV), and hepatitis B surface antigen nearly tripled between 1 week before September 11 (0.1%) and 1 week after the attacks (0.3%), largely explained by the increase in first-time and lapsed repeat donors. Estimated viral residual risks increased slightly after the attacks (HIV, 1/1.5 million vs 1/1.8 million donations; HCV, 1/1.3 million vs 1/1.6 million; hepatitis B virus, 1/140000 vs 1/170000). First-time donor 12-month return rates for 2000 and 2001 were similar, ≈28% (P=.37) for donors in the first week starting with September 11 (or September 12, 2000) and 30% (P=.69) for the second to fourth weeks. Conclusions: The September 11 events resulted in an influx of first-time donors without substantial increase in absolute risk of transfusion-transmissible viral infections. First-time donor return rates were equally relatively low before and after the attacks, suggesting that those donating in times of crisis have return behaviors similar to those of other first-time donors. Their relatively low return rates reinforce the need for education about the importance of donating regularly.