Playing the AIDS Odds

(21 Oct 93)
from the FAQ
Robert S. Walker, Ph.D.            Phone: (210)224-9172
Emeritus professor                 Internet: [email protected]
Trinity University, Pol.Sci.
715 Stadium Drive                  office: 128 Main Plaza, No.310
San Antonio, TX 78212                      San Antonio, TX, 78205

Everyone worries about the degree of transmission-risk involved in various activities. Can you get infected from mutual masturbation? From fisting? From using poppers? From this and from that? The real question is, "Is it possible to provide answers with sufficient precision to allow an individual confidently to assess risk and modify behavior in specific situations?" The answer is "No." No one knows enough about either sexual or drug behaviors, and their relation to HIV sero-conversion, to speak with assurance. But this doesn't mean that meaningful recommendations are out of the question.

Those interested in risk assessment might read two articles representing different approaches. First: Michael Shernoff, "Integrating Safer Sex Counseling into Social Work Practice, Social Casework: The Journal of Contemporary Social Work, vol. 69 (1988), pp. 334-339. The author offers a scaled list of 30 sexual behaviors from abstinence through fisting to condomless, receptive anal intercourse. The list is graded from "least likely" to transmit virus to "most likely." Some of the relative rankings are arguable, but the biggest problem is that the intervals of the "risk" scale are not equal. For example, #29 is "vaginal intercourse to orgasm without condoms," #30 is "anal intercourse to orgasm without condoms;" these two are separated by the same scaler distance as abstinence (no.1) and solitary masturbation (no.2). But everyone agrees that, anal intercourse is many times more dangerous than vaginal for the receptive partner, not just "one interval" more dangerous. Such lists are not too useful; I doubt that any subscriber to this list needs to be told that solitary masturbation is safer than receptive anal intercourse. Further, until a lot more is known about the relationships between specific behaviors and sero-conversion, the intervals cannot be meaningfully quantified.

The second article is Norman Hearst and Stephen B. Hulley, "Heterosexual AIDS," Journal of the American Medical Association, April 22, 1988. The authors calculate probabilities for HIV transmission for different parameters (such as: the area's seroprevalence rate, the infectiousness of a partner, the condom/spermicide failure rate, and the number of sexual encounters). The "odds" of transmission with different parameters (such as: 500 encounters, .01 condoms failure rate, area seroprevalence of .0001, and so forth) are then projected. The resulting odds range from a "low" of 1 chance in 5 billion to a "high" of 1 transmission in 500 encounters. In the lowest risk example, there is 1 in 5 billion chance that HIV will be transmitted when: (1) your partner tests negative; (2) he/she has no history of high-risk behavior; (3) condoms are used in intercourse, and the condom failure rate is .01; (4) the area seroprevalence rate is 0.000001, (5) the infectivity value is 0.002; and (6) there is only one sexual encounter.

As behavioral guides, neither approach is very helpful. When the possible sex or drug scenarios become as disparate as they are in real-life situations, and when the odds resemble your chances of winning a major lottery, then stating intervals or odds does not provide much more than a illusion of knowledge and resulting security.

I suggest a different approach to thinking about risk. First, do not worry about practices for which there is no documentation of transmission (as distinct from speculation about it). If there is any risk in kissing, masturbation, skinny-dipping or whatever, it is probably much less than the chance of being hit by lightning - and few people worry about that. Focus on those activities, like intercourse and/or injecting drugs, which common sense tells you are risky, if for no other reason than that they have a long history of transmitting other diseases (like syphilis or hepatitis). Such behaviors would clearly include injecting drug use within a group, condomless anal and/or vaginal intercourse, and less clearly oral sex, fisting, or any S&M; practice that involved a possible blood exchange.

Second, take into account the overall setting within sexual or drug activity is taking place. While it seems that we are all biologically at equal risk, we do not face equal environmental risks. While HIV theoretically can spread uniformly from the North to the South pole, it has not in fact done so. It is one thing to pick up someone at a bar in Brahma, Oklahoma and another in San Francisco, California. The risk involved in employing a prostitute in Des Moines is much less than in Newark, NJ or Washington D.C. where the seroprevalence rate among prostitutes is very high. Similarly, patronizing a Newark shooting gallery or crack house is like asking for AIDS, but the risk of transmission within the West Coast drug scene is much less. For area comparisons see the Centers for Disease Control's quarterly HIV/AIDS Surveillance Report, and/or Jonathan Mann et al, AIDS in the World, Harvard U. Press, 1993.

What I am suggesting is that some information plus common sense is a better guide than current statistical or quasi-statistical statements about relative risk. This will remain the case until a great deal more empiric data is amassed about some of our most private behaviors. If you are a person who does not feel comfortable without precise, reliable, quantified guidelines, then your only course is to abstain from activities wherein there is a possibility of transmission. There are many mood-altering substances that do not require injection, and a lot of sexual behavior that does not involve penetration and fluid exchange.

With respect to non-sex or drug modes of transmission, all one can say is that there have been no documented cases of transmission through insect bites, shared utensils, shared occupational space or equipment, food handling, and so on. Theoretical risks for an infinite number of imagined scenarios can be computed, but in the actual world there are no data supporting transmission in these scenarios. An excellent survey of 14 principal articles searching for data on other routes of transmission can be found in: Robyn R.N Gershon et al, "The Risk of Transmission of HIV-1 Through Non-Percutaneous, Non-Sexual Modes: A Review," Department of Environmental Health Sciences and Department of Epidemiology, The Johns Hopkins University School of Hygiene and Public Health, distribut- ed by New York City's Gay Men's Health Crisis, AIDS Clinical Update, October 1, 1990. There have been cases of transmission through transfusions /transplants of contaminated whole blood, blood products, donor organs, and dental work. The only thing one can do is to be aware of the possibility, and make sure that those who treat you take all precautions.

Currently, the only way to load the dice in your favor is to use common sense in any situation wherein someone else's body fluids might be introduced into yours through sexual or drug behaviors. If one can foresee that there would be opportunity for fluid exchange - blood, semen, vaginal secretions - then a large measure of safety can be had from the use of condoms (see: Condom Faq) and/or your own works for injecting drugs. The only safer course - and it is an honorable and intelligent one - would be to abstain from such activities altogether.

What must be kept in mind is that the risk of HIV transmission is totally unlike the risk of losing at the races. Because you cannot recoup the loss represented by infection, you ought not think of the "odds" in the same way. In fact, it is better not to focus on the so- called "odds" at all. Given that (1) infection almost always leads to AIDS (estimates=95%), and (2) that AIDS almost always leads to death (estimates=99%), people must now think of sex or injecting drug use as an all-or-nothing game, . Each time you play, there are only two possible outcomes. If you win you have, perhaps, enjoyed a pleasant encounter; if you lose, you die. And each time you play without regard to common sense evaluation and personal protection, you enhance the possibility that you will lose. Its as simple as that.

[email protected] (Mon Jul 11 21:08:13 1994)