A number of conclusions can be drawn from the evidence discussed in this chapter and in other effectiveness reviews.

Community mobilisation methods have been proven to work

Research evidence clearly demonstrates that community–based education efforts which rely on diffusion techniques and the development of skills lead to behaviour change. There are also sound theoretical reasons (see Social diffusion above) for assuming that well–planned and carefully managed community mobilisation interventions which are sensitive to the values and the diversity of a community will be successful in reducing the incidence of HIV infection, although no study has yet been designed to test this assumption.

There is a doseresponse in HIV prevention activities

A variety of intervention studies provide evidence that greater exposure to HIV prevention messages, skills building and community norms which promote safer behaviour will be more effective than one–off or occasional interventions. This requires the development of a sequence of activities which are co–ordinated to achieve an adequate level of exposure amongst members of the target group, not one–off interventions.

Intervention techniques are adaptable to different communities and risk situations

The medium is not the message. Many of the interventions described in this chapter can be adapted to different communities and to communicate different types of messages. There is no reason why techniques already successfully deployed amongst gay men, African–American youth and injecting drug users should not work with other groups, provided that they are culturally sensitive. Nor is there any reason to abandon tried and tested methods simply because HIV prevention messages are becoming more complex.

Interventions with a sound basis in research on risk and behavioural change

Much HIV prevention work taking place now is based on a mish–mash of theories about behaviour change and HIV risk factors. All the interventions discussed in this chapter have one thing in common: a clearly thought–out rationale based on a set of research findings about HIV risk and the way that human behaviour can be influenced.

A variety of tools exist for measuring the success of prevention activities: use them

Using a combination of outcome measures may be the best way of evaluating the effectiveness of a programme of interventions. Unless very sound evidence exists of the relationship between one variable and HIV incidence, the value of a project should not be assessed on the basis of changes in one variable alone. HIV prevention needs to develop performance indicators which measure quality hand in hand with quantity of activity.

Interventions should challenge reasoning, not just risk perception

Many of the interventions discussed contained cognitive components designed to make people think about the ways they rationalise risky activities. This is not proof in itself that reasoning needs to be foregrounded in future activities, but it does suggest a potentially useful avenue for research and project development.

How can we get best value for money in HIV prevention now?

It depends which population is being targeted. Amongst gay men, community mobilisation coupled with more interventionist counselling techniques in STD clinics might yield the best returns, together with a continued emphasis on condom distribution. Face-to-face outreach work does not appear cost–effective in comparison with media work and free condom distribution, particularly in large population centres. Research which can identify the reasons for seroconversions, and which can clarify the future role of GUM clinics in HIV prevention is urgently needed, and would be a good investment.

Amongst African communities, diffusion of safer sex messages is likely to be best accomplished by community organisations and community organisers, but the measurement of outcomes is hampered by fact that it will be difficult to judge whether HIV infections have been avoided as a consequence of HIV prevention activities in the UK or in African countries. However, recent evidence gathered by a variety of research projects suggests that there are big variations between African communities in levels of knowledge about HIV, and an investment is needed in action research which can point the way towards more extensive activities.