In the study of the Miami dental practice [1], only 20% of the patients were tested, allowing one to assess only whether a cluster of perhaps five or more nosocomial infections with human immunodeficiency virus (HIV) occurred. Moreover, of the four patients with no identified risk factors suggesting the possibility of nosocomial infection, only two were available for molecular studies.

The authors assumed that "given that intense publicity. most dental patients with HIV infection would have been reported to HRS (Florida Department of Health and Rehabilitative Services)." Approximately 10 persons who suspected that they were infected (most with hepatitis B or C virus, a few with HIV) by dental or medical devices have contacted me. In each case, the onset of symptoms was within expected incubation periods. Their physicians acknowledged that the patients had no identifiable risk factors, but none reported the cases to their state health department or to HRS. Only one patient sought legal advice and tried to report his case to the Florida state health department, but he was told that his physician would have to report it. The other patients would not discuss their suspicions with their dentists or physicians; one would not even inform her spouse and children that she had become infected.

A dentist in Denmark once remarked to me: "I've seen blood come out of my equipment." When nonsterile equipment is widely used in invasive procedures and is contaminated with visible amounts of bloody saliva, HIV will be transmitted occasionally and other pathogens (for example, hepatitis B and C virus and cytomegalovirus), much more frequently. It is probably safe to assume that HIV infections in dentistry are rare, usually involve only a single transmission in a practice, are usually transmitted from patient to patient (one is more likely to be exposed to blood from an infectious patient through equipment than directly from an infected dentist), usually occur in practices with noninfected dentists (relatively few dentists are infected), and are extremely unlikely to be documented (only practices with infected health care workers are investigated, unless a sizable cluster of cases with no identifiable risk factors were to develop, be recognized, and be reported to a health agency willing to investigate the practice).