Determinants of the effectiveness of HIV prevention through sport
Wim Delva, Marleen Temmerman
International Centre for Reproductive Health
Ghent University
Belgium
Sport and play are increasingly being used for youth wellness and development purposes, including prevention of HIV. This paper outlines a conceptual framework in which the impact of sport on HIV can be assessed. We argue that the effectiveness of HIV prevention through sport can be seen as the result of a cascade process from the introduction of the intervention to the intervention’s impact on HIV incidence. We describe how socio-cultural, socio-economic and epidemiological factors mediate the success of sport for HIV prevention at different stages of the intervention-impact cascade. The elements and processes of the conceptual framework are exemplified by facts and figures from Sub-Saharan Africa. Recommendations for the design of sport programmes and their evaluations are formulated and data gaps requiring additional research are identified.
Introduction
While the HIV prevalence in South African youth is high and unsafe sexual practices seem to be popular among African youth and young adults, so are sport and play. Sport is known to help build confidence and resilience: social competence, autonomy, and optimism can be enhanced through physical activity, games, and sport. (1,2) Essential life skills such as teamwork, cooperation, problem solving through communication and relationship building are inherent to sport. Coaches and team members can become important role models and values such as gender equality and respect for one another can be adopted through social learning. Thus, sport facilitates the development of life skills that are needed to translate knowledge, attitudes and behavioural intentions into actual behaviour. In addition, stigma and discrimination of people infected or affected by HIV can be tackled by the non-discriminative, impartial nature of sport and play. (3) Furthermore, sport and play have the potential to reach out to hard-to-reach and vulnerable groups such as learners who dropped out of school and unemployed young adults, migrant populations, women and girls, and children affected by HIV/AIDS. Sport and play can be introduced in these groups through grass root community-based initiatives with a low threshold for participation and at a low cost. (4) Sport offers the opportunity for repetitive contact with coaches and peers, thus creating a natural forum for interactive discussions, life skill building and peer education. For all of the above reasons, sport is thought to be a feasible, accessible, affordable and efficacious vehicle for HIV prevention. The national HIV/AIDS campaign LoveLife and the organization Sports Coaches’ Outreach (SCORE) both use sport and play as a tool for HIV awareness and community development in South Africa. (5,6)
A societal framework of effectiveness
Having the potential to effectively curb the HIV epidemic does not however guarantee success. Whether significant numbers of new HIV infections will be averted through sport programmes depends on the socio-cultural and political context of the intervention, the socio-economic context of the targeted sexual behaviour and the epidemiological context in which individuals acquire and transmit HIV infection. (7) Figure 1 illustrates how these contextual factors potentially attenuate the effectiveness of HIV prevention through sport at different stages of the intervention-impact cascade.
Figure 1. Stages and determinants of the intervention-impact cascade. (adapted from Grassly et al., 2002)
1. Determinants at the intervention level:
Is there breathing spaceroom for youth and sport in the community?
The socio-cultural and political contexts in which the youth and sport culture are embedded are likely to affect the effectiveness and sustainability of HIV prevention through sport. A myriad of frameworks rooted in social science theory have been proposed to identify structural facilitators and barriers of HIV prevention interventions and to understand how these factors interfere with the implementation of interventions. (8-11) Although social frameworks provide insights into the complex nature of sexual risk behaviour and health-seeking behaviour, they typically do not allow for an objective, quantitative evaluation of prevention programmes. (12) Nevertheless, social science is imperative to determine the socio-cultural and political appropriateness of a given intervention. Questions related to feasibility, accessibility and affordability of sport programmes include: Is sport being encouraged by parents, schools and governments? Do kids have the opportunities, ample leisure time and space to play and practice sports in the community? Are local, regional and national authorities making an effort to lower the (financial) threshold to participate in sport and play? Are human capacity, infrastructure and financial resources in place to train and guide the trainers and coaches, to supervise the programmes and to monitor and assess their effectiveness? Additionally, it is essential to investigate potential sport-related sources for stigmatisation, discrimination or formation of unfavourable stereotypes before promoting a particular sport activity as the vehicle for life skills building and behavioural change. Social science should however not be limited to Policy Advisory Research. When impeding and jeopardizing factors related to the intervention and/or the target population are known, informed actions can be undertaken and the intervention can be adapted/modified to maximize its accessibility. In spite of the importance of establishing an acceptable, feasible and affordable intervention, the current evidence base regarding the socio-cultural and political appropriateness of sport programmes for HIV prevention is sparse and social science research in this regard is urgently needed.
Lacking evidence should however not prevent us from pursuing a favourable, supportive environment in which we can unlock the potential of sport for HIV prevention. Acknowledging the socio-cultural and political determinants for successful implementation of HIV prevention programmes for youth, collaborations with schools and governments as well as efforts to involve parents are crucial. In this way, the process of adopting positive values and building social skills through sport continues after the training and play. Anecdotal evidence from South Africa indicates that some schools have not or incompletely implemented “curriculum 2005” (a school-based approach to HIV prevention) because teachers and/or parents believed that explicit information on sexuality and safe sex would increase sexual exploration in learners. (13) Another issue demanding attention is the plight of Children Affected by AIDS (CABA), who are particularly vulnerable to lose the opportunities to sport and play as their work burden in the household and during income generating activities is often extremely high. (14) Reaching these children will prove to be a growing challenge in the light of a continuously expanding population of orphans and vulnerable children. Lastly, governmental support and investments in long-term capacity building throughin trainers and coaches and their umbrella organisations are key to ensure the lasting legacy of current development efforts. (3)
2. Determinants at the level of sexual behaviour:
Does the intervention result in sustained behavioural change?
No one's perceptions and attitudes are developed independently of one's cultural background and social networks. Subsequently, (sexual) behaviour is never just a matter of making personal choices. It always involves issues such as peer pressure and role models, self-efficacy and perceived benefits versus risks. Great lessons can be learnt from the evaluations of school-based programmes aimed at behavioural change. A recent systematic literature review from South Africa points out that although school-based programmes are usually associated with improved awareness, knowledge and attitudes, very few result in actual changes in risk behaviour. (15) Only programmes dedicating ample time to communication skills, gender equality, self-esteem and self-efficacy training and role-plays showed to impact on sexual behaviour. (16,17) In contrast, programmes based upon the belief that behaviour was the result of an informed choice failed to act beyond raising awareness and more positive attitudes and intentions. (18-20) Additionally, people’s behaviour is very often rooted in economic and developmental realities: unemployment, poverty, migrant work and gender inequality have been identified as the most important driving forces of the HIV epidemic in Sub-Saharan Africa. (21-25)
In conclusion, planning of HIV prevention programmes through sport should include a comprehensive situational analysis of barriers that could hamper sustained behavioural changes. In addition, experience and study findings from past operations research can help develop best practice guidelines in addressing and overcoming anticipated sources of disempowerment.
3. Determinants at the level of the epidemic:
Will behavioural change result in a lower HIV incidence and prevalence?
According to the epidemiological framework outlined by Grassly et al., the impact of programmes altering sexual behaviour and promoting condom usage on HIV incidence depends on the epidemiological context, indicated by the HIV prevalence in the target population, the prevalence of cofactors of HIV transmission (e.g. STI prevalence), mixing patterns between the target population and untargeted populations, and the sexual behaviour of the untargeted populations. In other words, merely proving that sport can reduce unsafe sexual practices would not be enough to ensure significant consequences in terms of averted HIV infections. Indeed, besides the effects of sport on sexual behaviour (delayed sexual initiation, secondary abstinence, increased condom use, reduced number of partners, reduced number of casual and transactional sexual contacts), the indicators for the epidemiological context as mentioned above need to be measured as well. Obviously, the effectiveness of a programme merely targeting secondary school boys may be seriously curtailed if most of the learners' partners have dropped out of school already and are therefore not reached. Even worse results may be expected if these female partners tend to have concurrent sexual partnerships with older men because women in such relationships often lack the power to negotiate safe sex and older men are more likely than younger men to be HIV positive. Conversely, a programme's effectiveness may be enhanced when the epidemiological context is known. For instance, adding promotion of STI screening and treatment through sport-based peer education may be required when STIs are rife and disappointingly little changes in HIV incidence are observed even after significant increases in condom usage. Garnett and Anderson showed that dramatic increases in condom distribution may have very little impact on HIV spread until use during sexual intercourse is close to 100% in high-risk partnerships. (26)
The epidemiological context in South African youth
HIV prevalence and incidence in the target population
The HIV prevalence in South African youth is worrisome and, in girls, has continued to rise since the 2002 national household survey. (27) Figure 2 shows how the HIV prevalence peaks at a higher level and in younger age groups for women than for men. Based on weighted data to correct for stratified, disproportionate sampling and account for non-response to HIV testing, the HIV prevalence in the 10 to 14 year old age group is estimated to be 1.64% for boys and 1.75% for girls. In 15 to 19 year olds a differential infection rate becomes apparent as 3.23% of these male adolescents is infected compared to 9.40% of female adolescents. This trend is reinforced in the 20 to 24 year olds with 6.03% of men being infected in contrast to 23.85% of women. (28) When applying these prevalence rates to the 2005 mid-year population estimates for South Africa, more than one million young South Africans between 10 and 25 years old are estimated to be HIV positive, representing nearly one quarter of all people living with HIV in South Africa. (29)
Figure 2. HIV prevalence in South Africa for 2005, by age and gender.
The relatively lower prevalence rates in youth compared to the adult population may generate a false sense of comfort and control. In fact, they hide/disguise shockingly high HIV incidence rates: using advanced testing essays, the HIV incidence among youth aged 15–24 years was estimated at 3.3% in 2005. Highly alarming is the fact that females in this age group have an eight-times higher HIV incidence than males (6.5% compared to 0.8%). These findings are consistent with data on sexual behaviour indicating that youth have a high turn over of sexual partners and that a sizable proportion becomes sexually active early in adolescence. Reasons for the increased susceptibility of girls include biological factors (cervical ectopy, incomplete vaginal lining, larger surface of mucosal membrane) and difference in mixing patterns: a high proportion of girls tend to sustain sexual relationships with men who are older (and therefore more likely to be HIV positive). Additionally, forced sex and sexual violence may also contribute to their vulnerability to HIV infection.
The prevalence of cofactors of HIV transmission
Additionally, STI prevalence rates are high in South African youth – especially in females – justifying intensified efforts to improve STI screening and treatment in these groups. In a South African community-based study, the prevalence of Chlamydia trachomatis was 3.5% for males aged 15-19 and 9.1% for females of the same age. Neisseria gonorrhoeae was prevalent in 1.1% of 15 to 19 year old males and in 3.5% of their female counterparts. (30)
Mixing patterns between the target population and untargeted populations
While boys and young men tend to have sexual partners of their own age groups, this is not true for many of their female counterparts. Through high-risk – and often transactional – sex, these women are at high risk for HIV acquisition, thus introducing the virus into the sexual networks of younger age groups. These mixing patterns have important implications for the design and implementation of HIV prevention interventions: An isolated intervention only focussing on youth is unlikely to have a significant impact on HIV incidence and HIV prevalence rates.
Sexual behaviour of the untargeted populations
Addressing the sexual behaviour of the adult population is equally important if HIV preventions for youth are to be effective. Modelling exercises indicate that targeted interventions for individuals engaging in high-risk sex, such as commercial sex workers and migrant workers effectively avert HIV infections. In Family Health International’s AVERT simulation model, social marketing of condoms in combination with presumptive treatment of STIs were estimated to lead to a 39% decrease in HIV incidence was for women using the STI services while a 48% decrease in HIV incidence was estimated for miners. (31) Recently, South Africa was the first country to present experimental evidence on the effectiveness of male circumcision for the prevention of HIV infection from a randomised controlled trial. After a mean follow-up period of 18.1 months, the annual risk of HIV-1 transmission in the intervention group was 60% lower than that in the control group. (32) Whether mass media interventions such as Soul City or LoveLife attenuate the HIV incidence in South Africa is far more unclear. Although associations have been described between exposure and HIV prevalence, no causal relationship nor an association with changing sexual behaviour could be shown. (2,33)