BACKGROUND: In 2003, West Nile virus (WNV) nucleic acid amplification testing (NAT) was implemented to detect potentially infected donors. Of more than 5.3 million donations screened prospectively by the American Red Cross during the epidemic periods of 2003 and 2004, 974 were NAT-reactive and 519 confirmed-positive. A subset of both the confirmed-positive and the false-positive groups was assessed for demographic characteristics, symptoms, and symptom reporting relative to date of donation. STUDY DESIGN AND METHODS: All donors with initial WNV NAT-reactive results were invited to participate in a study that included a demographic, symptom, and date-of-symptom questionnaire. WNV confirmed-positive cases were compared to false-positive controls for comparison of frequency of symptom reporting before, on the day of, and after donation. RESULTS: Enrolled cases and controls were similar in all characteristics except cases were more likely to live in rural areas. Symptoms were reported by 61 percent of cases versus 20 percent of controls, with 74 percent of symptoms reported by cases within the 14 days after donation. The frequency of headache and fever reported together in the 7 days before donation was not significantly different between cases and controls; only the individual frequencies of headache, eye pain, and new rash during this time were significantly different. The most commonly reported symptoms, after adjustment for symptom reporting by controls, were headache, new rash, and generalized weakness; these symptoms were reported by 25 percent of cases. CONCLUSIONS: The demographic characteristics of infected donors reflected the rural nature of the 2003 to 2004 WNV epidemics. This study suggests that asking donors about predonation headache and fever had no detectable contribution to blood safety.