We studied the students' acceptance and utilization of virtual patients (VPs) authored by faculty using the Web-SP system over two consecutive years. We also studied factors of importance for the utilization of VPs for self-assessment. Both year-groups studied found the Web-SP system easy to use and their overall opinion of Web-SP was positive (Median: 5, p25-p75: 4-5). They found the VPs engaging, realistic, fun to use, instructive and relevant to their course. Students used, on average, 9.68 VPs per course, which constitutes 43 percent of the available VPs. The number of VPs available seemed to be sufficient for the target course, even if some of the students preferred a higher number of VPs. Of the VPs encountered, 71% (CI: 68-75%) were VPs with feedback, and correspondingly 29% of the VPs chosen were without feedback. The difference in utilization between both types of VPs was significant, at p < 0.001. Thus, the students clearly favoured VPs with feedback compared to VPs without feedback. There were three modes of engagement in which the VP was utilized. Mode 1 was the preferred mode for VPs without feedback, while mode 3 was dominant for VPs with feedback.. Whether or not a VP was selected for review during a teacher led seminar or not, did not affect student behaviour, at least on the surface. Teacher led seminars may still be of importance to provide credibility to the VPs by integrating them into the curriculum.