Followup: did that dentist really infect his patients with AIDS?


SHARE Followup: did that dentist really infect his patients with AIDS?

Dear Straight Dope: Regarding your recent article about intentionally infecting people with AIDS, I read a pretty persuasive article (or heard an argument on the radio, I can’t recall which) maintaining it was far more likely this dentist’s patients contracted AIDS in the more traditional ways. Kimberly Bergalis, for instance, who became the poster girl for mandatory testing of health workers, was alleged to have been quite promiscuous. But she looked so damn tragic testifying to Congress in her wheelchair, barely able to talk, that no one brought this up. And it was alleged that no one has YET come up with a plausible way for the patients to get AIDS from the dentist’s instruments. Do you know anything about this? JOHNVARLEY

SDStaff Jillgat replies:

It is far more likely that health care workers would be at risk of contracting HIV from their patients than the other way around. That’s one reason this case was so unusual. As far as I know, no states have laws requiring mandatory HIV testing of health workers. I have also seen no epidemiological studies that report sexual risk factors for Kimberly Bergalis. Four of the other five infected patients also had no obvious risk factors for HIV, and most had blood samples available from earlier medical procedures or blood donations that tested negative, establishing that they became infected after having dental procedures done by this doctor. These patients also tested negative for other sexually transmitted diseases and hepatitis.

Viral DNA sequencing of HIV found in lymphocyte samples obtained from the dentist and five of his patients strongly suggest that these patients became infected with HIV while receiving care from the infected dentist. None of the five patients had other confirmed exposures to HIV, all had invasive procedures performed by the dentist, and all were infected with HIV strains that were closely related to each other and to the strain infecting the dentist, but distinct from viruses obtained from control patients living in the same geographic area as the dental practice. One other infected patient had a strain of HIV that was not closely related genetically to that of the dentist and the other patients, and he also had behavioral risk factors for HIV.

There is no way to know exactly how this dentist may have infected his patients. At least one of his friends said that he had been angry that the media and government didn’t give enough attention to the AIDS crisis, and that maybe they would if others – besides homosexual men – were becoming infected. Some people think he actually injected his blood into his patients. There is no way to prove this, but there is ample evidence, to my mind, that he did indeed infect them.

If you would like the specific articles about investigations of this case, email me at, and I will fax them to you.

Dear Straight Dope:

Thanks for debunking the Urban Myth of The Intentional AIDS Infector. A comment, though. By your own reckoning, predicting the chances of getting AIDS from one unprotected encounter is iffy at best, yet you have provided a “rule of thumb.” Is this 1 in 500 substantiated anywhere? Is it for both vaginal and anal intercourse? Does it include oral sex and semen consumption? In any event, it’s not clear what this statistic means, if anything, or from where it comes.

My real comment, though, is this: Some people will take your rule of thumb and think, “Hey, those are pretty good odds in my favor,” which, as you probably agree, is not the best perspective to take with this sort of thing. Because the 1 in 500 odds you provided is so potentially variable, it strikes me as irresponsible to provide it at all when it’s likely that people will construe it as one argument for not participating in safer sex.

Admittedly, I’m all for people believing that they can get AIDS from looking at someone the wrong way; some people would be better off IMHO with that kind of caution in their repertoire. And I agree with you that people should have all the information available, and what they do with it is typically their own business. However, I think that if you’re going to suggest that it’s not all that easy to contract HIV and get AIDS and then back it up with a statistic, you need to provide your readers with more information about where that stat came from so that it can be put into proper perspective. Thanks for hearing (reading?) me out.


SDSTAFF Jillgat replies:

I don’t agree that we should convince people that HIV is easy to get (especially not from looking at someone the wrong way!). It’s important to be honest in HIV/AIDS education, and sometimes it’s difficult to speak accurately without either exaggerating or discounting the risks.

You are right, however, that a statistic like “1 chance in 500” is far too simplistic to be a guide. (Little Ed insisted I stick that number in; it came from an article in the Journal of the American Medical Association that Cecil cited in a column once. The column isn’t in the Straight Dope’s AOL archive yet, but supposedly it’ll get posted next week.) One has to be very careful about passing statistics like this one onto a lay audience (no pun intended). Statisticians love to throw such numbers around, but they don’t really generalize to the general human population. The HIV epidemic, like any other, occurs within a complex social environment. At least some people will hear this statistic and conclude that they must only use condoms for every 500th sexual encounter.

There is tremendous variability in the risk factors for contracting HIV. How common is HIV in the population from which you draw your sex partners? If it is very rare, you are at low risk no matter what kind of sex you engage in. How many sex parters do you have? The risk of contracting HIV varies quite a bit between different sexual activities, even when all other variables are equal (which they never are). Most people know that receptive anal intercourse is far riskier than oral sex, and that the virus tends to spread more easily, for a variety of reasons, from male to female than from female to male. Homosexual men who have receptive anal sex with multiple partners are at high risk, heterosexuals having vaginal intercourse are at somewhat less risk, and lesbians are at practically no risk, if they stick to female partners and don’t shoot drugs. I have seen different studies attaching numbers to the risks of these activities, and I apologise that I can’t put my hands on the sources of this information. I also have given positive HIV tests to two men who I believe from their sexual history only had engaged in receptive (the mouth part) oral sex, so I believe such transmissions occur. Several studies have concluded this as well.

An infected person’s viral load at the time of the exposure will influence the likelihood of them spreading the virus to others. A very early or very late stage of infection is a strong predictor of infectiousness according to both epidemiological and biological data. New combination therapies which lower blood viral load may affect this, but I haven’t seen any recent studies that are conclusive about this. When the blood viral load goes down to “undetectable”, I’m not sure that seminal viral load does also.

Epidemiologic data suggest that at least a few people are actually immune, or lack susceptability to HIV infection. These people remain uninfected despite repeatedly having unprotected sexual intercourse with HIV infected partners (studies have found this in some sex workers and homosexual men).

HIV subtypes have distinct geographic distributions, with pattern II countries (SE Asia, Africa) having more heterosexual spread of HIV than pattern I countries (US and Europe), where homosexual men and injection drug users still account for the majority of cases. Some studies suggest that these different strains favor different transmission routes.

Some other factors that can confound such a simplistic formula are the presence of genital sores or abrasions or other sexually transmitted diseases (higher risk), male circumcision (prevalence of HIV infection is 1.7 – 8.2 times a high in men with foreskins as in circumcised men), cervical ectopy (relative risks ranging from 1.7 – 5.0), sex during menstruation (men who have sex with HIV-infected women during menstruation are 3.4 times as likely to have HIV infection as those who do not), choice of contraception, mucosal PH, duration and frequency of sexual intercourse, and the number of partners.

The probability of female to male transmission per contact (presumably referring to vaginal intercourse) in the US and Europe is 0.0003 to 0.0014. It’s higher in Thailand. The article I looked at, SEXUAL TRANSMISSION OF HIV (NEJM, April 10, 1997, Volume 336, Number 15) has a chart showing the infectivity (per-contact) transmission on a log scale. It’s a faxed copy, and isn’t clearly readable enough for me to see the numbers. You might want to do a search and get a copy of this article, because it also lists 97 other reference articles on this topic.

SDStaff Jillgat, Straight Dope Science Advisory Board

Send questions to Cecil via