Since the beginning of the epidemic we have been faced with two competing philosophies about the relationship between acceptable levels of risk and sustainable degrees of behaviour change and public health measures.

One model, that of risk reduction, has sought to demonstrate that the risks of HIV transmission are identifiable and that proven methods of protection against infection exist. These include:

• Use of condoms in penetrative sex.
• Screening of the blood supply.
• Heat-treatment of Factor VIII.
• Avoidance of breastfeeding where feasible and necessary.
• Adoption of universal precautions in medical settings where invasive procedures take place.

Thus HIV prevention based on a risk reduction philosophy involves substituting less harmful activities for those which pose the greatest risk. For example, the promotion of needle-exchanges is a form of risk or harm reduction; it does not eliminate the potentially harmful activity of injecting drugs, but it does offer a means of reducing the risk of HIV infection.

Similarly, the promotion of condoms to gay men for sex is seen as preferable to a policy of discouraging gay men from having sex altogether. It offers gay men a choice of ways in which to reduce their level of risk, and offers substitutes which are proven to be of lower risk, e.g. oral sex instead of unprotected anal sex. Exactly the same strategy has been employed with workers in the sex industry in many countries.

Risk elimination

Risk elimination, on the other hand, depends on the belief that protecting public health requires the elimination of risk. Risk elimination approaches take two forms:

• The promotion of abstinence over condoms and needle exchanges. This approach questions the efficacy of safer sex, and suggests that injecting drug use is just as harmful as the sharing of injecting equipment.
• The highlighting of very low risks as unacceptable risks.

The highlighting of very low risks can have the paradoxical consequence of inducing fatalism about past sexual practices and reducing adherence to current tried and tested safer sex guidelines, as a 1992 Dutch study showed amongst gay men. Those who became worried that oral sex was risky were most likely to abandon the use of condoms in anal sex, believing themselves to be already exposed to the virus through oral sex.

Another example is the emphasis upon the testing of health care personnel in order to protect the public from the tiny risk of infection through an invasive procedure. This draws attention away from the need for universal precautions to be employed in all invasive procedures, and undermines confidence in such precautions.

Of course, we have to make allowance for what we do not know and err on the side of caution in predicting the risks of the developing epidemics. Worst case predictions are intended to be treated as such precisely in order to prevent such scenarios from coming about. However, when such scenarios interfere with demonstrably effective health education, they become almost as dangerous as denial of any risk whatsoever.

A further principle associated with risk reduction is that of minimal disruption. Changes in behaviour are thought to be more sustainable if they involve the least possible change in behaviour required to protect oneself. This is why for many people, cutting down the amount of fat one eats is likely to be a far more realistic, if minimal, form of risk reduction than giving it up altogether. Advice of this sort works with the grain of long established and pleasurable behaviour, rather than against it. Sexual habits are deeply rooted in everyone's lives, and require rather more than will–power to change.

Those who advocate risk elimination argue that the public have the right to be aware of all the risks associated with HIV infection, and to make up their own minds on the basis of such information. Such an argument presupposes:

• That everyone makes up their mind about potential risks on the basis of all the available facts.
• That perceptions of risk in the community are all formed by the same factors, leading to an even perception of risk. An example might be sexually transmitted diseases: some people regard these as more disastrous because they have less sex, or because they interfere with an existing relationship, or because the consequences of a sexually transmitted disease might be more serious for a woman than a man.
• That the facts are presented neutrally.
• That information about AIDS and HIV is not received by people in the light of previous prejudice, misinformation or blaming.
• That the information is presented in such a way as to be easily understood.
• That everyone is equally capable of acting upon that information to protect themselves.
• That there are unlimited resources to present information about even the tiniest theoretical risks. An example of this might be the choice between educating gay men about the dangers of unprotected fucking and the much smaller danger from oral sex.

Realistic risk reduction advice

Following on from the criticisms of risk elimination, it is important that risk reduction advice should be:

• Easily understood by those it is intended to reach.
• Implementable. In other words, people have to be willing and able to take the suggested. precautions. To suggest that all sex is risky is to invite denial of any risk attached to sexual activity.
• Persuasive, not punitive.
• Not more disruptive than absolutely necessary.

For further discussion of these issues see Safer sex and Drug use.