A large randomised controlled trial of sexual and reproductive health interventions in schools, health services and communities in Tanzania has found that, while the interventions increased adolescent knowledge about sexual health, they had no significant impact on HIV transmission or other sexually transmitted infections, nor on pregnancy rates.
The findings were presented last week at the Fifteenth International Conference on AIDS and Sexually Transmitted Infections in Africa, held in Dakar, Senegal.
HIV prevention among young people is an urgent priority, since according to UNAIDS 45% of HIV transmission worldwide takes place in young people aged 15 to 24 years. However there is still a lack of evidence about which approaches work best in preventing new infections, the gold standard for any HIV prevention study.
In order to accumulate more evidence about what works, researchers at the Tanzanian National Institute for Medical Research have been conducting research in the Mwanza region of Tanzania for nearly ten years, in a trial which randomised twenty communities either to receive existing services or to receive a comprehensive package of interventions aimed at adolescents. The study is called Mema kwa Vijana (`Good things for young people`), and was developed in partnership with the London School of Hygiene and Tropical Medicine, the Liverpool School of Tropical Medicine and the UK’s Medical Research Council.
The intervention package consisted of:
- In-school sexual and reproductive health education targeting adolescents aged 14 years and above
- Youth-friendly reproductive health services
- Community-based condom promotion (between 2000 and 2002 only)
- Community interventions to create a supportive environment for the interventions.
The intervention package was implemented by the African Medical Research Foundation (AMREF).
The first evaluation, of the initial cohort of 9645 adolescents, looked at outcomes after three years, and found substantial improvements in knowledge about sexual health, but no significant difference in biological outcomes such as HIV incidence, HSV-2 infection, sexually transmitted infections and pregnancy between the intervention communities and the control group.
Study investigators questioned whether the intervention had carried on for long enough, and a further evaluation was carried out in 2007/08, covering 13,814 young people who had been exposed to the intervention between 1999 and 2002 (together with the control population).
The second evaluation found a sustained difference in knowledge of sexual health, although not as pronounced as in 2002, and a trend towards delayed sexual debut in the intervention group. Condom use was significantly higher only among young women with a non-regular partner in the intervention group. There was no difference in the frequency with which concurrent sexual partnerships were reported, nor in use of health services by those who experienced symptoms of a sexually transmitted infection in the past twelve months.
There was no significant difference in sexually transmitted infections, HIV prevalence or HSV-2 prevalence between intervention and control communities, despite a much larger sample size and greater statistical power to detect a difference in the second evaluation, and despite the fact that two-thirds of the young people surveyed had been exposed to the in-school intervention for at least three years.
A second study, in Zimbabwe, using a similar package of interventions, also showed an increase in knowledge in the intervention group, but failed to show a significant impact on new infections.
The researchers warn that further research is needed to better define the package of interventions that can reduce new HIV infections and other sexually transmitted infections among young people. They note that, while the 2001 UN Declaration of Commitment on HIV/AIDS called for 95% of young people to have access to education about sexual health by 2010, two well-designed randomised studies now show that changes in knowledge and attitudes as a result of interventions do not reliably predict changes in the most important outcome, new HIV infections.
In other words, improving knowledge and access to services among young people is not enough to stem HIV infections in this age group. Other factors are clearly preventing young people from acting on the knowledge they have acquired.
The Mema kwa Vijana study could not measure population-wide attitudes towards sexual behaviour that continue to pressurise young people - especially young women - to engage in unprotected sex, and the researchers say that more work is needed to determine how population norms were changed in countries where HIV incidence has declined, such as Uganda and Zimbabwe.
These norms include gender- and age-related power imbalances, reluctance to use condoms, and transactional sex, especially between young women and older males.
While education about sexual health will continue to be necessary in order to equip young people with knowledge, additional interventions will be necessary. Identifying effective interventions to change population norms related to sexual risk behaviour should be a priority, the researchers conclude, together with the investigation of alternative interventions.
Further information
The study has a website that provides comprehensive information about the interventions carried out during the study.
Briefing papers providing more detailed information on the trial results and their policy implications are available by following this link.