(introduction...)

Evidence from a variety of countries suggests that open communication about sex between family members and young people remains the exception rather than the rule. In cultures where traditional systems for helping young people learn the roles and responsibilities of adult life existed, changing social circumstances and family structures have affected these channels of communication. In the absence of open discussion about sex within the family or wider community, and in recognition of the needs of young people for information which might help them to protect their sexual health, a number of formal programs of HIV-related health promotion including sex education have been instituted in countries across the developing world.

Styles of HIV-related prevention work aimed at young people have changed over the years. Early in the epidemic, individualistic approaches based on theoretical frameworks such as the Health Belief Model and Social Learning Theory were quite common (Aggleton, 1996). These emphasised the importance of helping young people to acquire accurate information and skills relating to the prevention of HIV/AIDS. It was assumed then that if young people could only develop appropriate knowledge and skills, they would be able to change their behaviour in order to enhance their sexual health. However, such approaches are now recognised as being over-simplistic and are criticised for failing to take account of contextual, environmental and structural factors influencing young people's 'choices', actions and behaviours. These include economic constraints, the effect of migration and war, power relations between women and men, inequalities between young and old, and relationships between dominant versus minority ethnic groups. In the most extreme circumstances, young people living in stressful situations may, for example, engage in 'survival sex' in order to meet their need for shelter, food and adult protection (e.g. Rotheram-Borus, Mahler & Rosario, 1995). In such precarious circumstances, young people are not well placed to make rational decisions on the basis of new information or to practice newly acquired skills, but are often constrained by the circumstances they find themselves in.

The middle years of the epidemic were characterized by the increasing development of HIV-prevention programs aimed at the level of community (Aggleton, 1996). These programs shared a common acknowledgement that decisions about behavior, including sexual decision- making, are made in the context of shared social experiences. In particular, peer education programs have attempted to address the social processes which influence the gender and sexual norms of young people. Several studies have demonstrated that peers are important in shaping gender identity and roles and attitudes towards sexual behavior among young people (Svenson, Hanson & Johnsson, 1995). Programs which attempt to work at the level of community, go some way towards a recognition of the social construction of gender roles and sexual attitudes and behavior.

More recently though, researchers and practitioners working with young people for the prevention of HIV/AIDS have shown greater interest in bringing about structural and environmental change. A burgeoning research literature has demonstrated that young people are constrained in their behaviours by social, economic, legislative and other factors which are beyond their personal control. Gender inequality, for example, means that many young women across the world are not able to participate as equal partners in sexual decision-making, and so cannot easily control their sexual health. There is now widespread acknowledgement that HIV prevention programs need to address public policy concerns so as to enable young people to protect their sexual health, while persuading them to take action that helps to protect them from becoming infected with HIV (Tawil, O'Reilly and Vester, 1995).

A broad variety of prevention programs have now been undertaken in developing countries with the aim of reducing the risks of HIV infection among young people. While some have been formally evaluated to determine whether or not young people's behaviour has been influenced, a good number are yet to be systematically evaluated. Broadly speaking, these programs can be divided into four main types: (i) programs designed to help adults improve their skills and increase effective communication about sex with young people, (ii) work with young people in schools, (iii) work with young people out of schools and (iv) work with young people at heightened risk. Here we offer some examples of recent programs which fall within each of these categories, and discuss their major strengths and weaknesses.