A training program in universal precautions was developed and implemented in 1991-92 for second-year students at the George Washington University School of Medicine and Health Sciences. The students were required to participate in a three-hour session that consisted of lecture, demonstration, and practice components focused on the risks of bloodborne-disease exposure and the techniques of phlebotomy and intravenous insertion using universal precautions. All 135 second-year students participated in the lecture component, but only 120 students, who were unfamiliar with the procedures, were required to participate in the demonstration and practice components. Each of these students was asked to answer pre- and postsession knowledge questions and to rate his or her preparedness on a five-point Likert scale, ranging from 1, "not prepared at all," to 5, "well prepared." Paired t-tests were used to compare the pre- and postsession knowledge scores and self-assessed preparedness scores. At the end of the training program, the students were offered the opportunity to volunteer for additional, individualized training with the hospital phlebotomy service. Unpaired t-tests were used to compare differences between the postsession knowledge scores of the volunteers and nonvolunteers. A total of 103 students completed both pre- and posttests. The students' knowledge scores increased from means of 64.7% to 88.5% (p = .001). Their self-assessed preparedness scores also increased, ranging from a low of means of 1.6 presession and 3.4 postsession for intravenous insertion to a high of means of 3.19 presession and 4.26 postsession for addressing personal concerns about possible exposure. The 43 students who volunteered for additional training scored significantly better on the postsession knowledge questions than did the nonvolunteers, suggesting that those who may have needed it most failed to sign up for additional training. The training session significantly improved the students' knowledge and sense of their own competency.