In collaboration with the Tanzania National Institute for Medical Research (NIMR) and London School of Hygiene and Tropical Medicine (LSHTM)

Evaluation of the long-term impact of an adolescent sexual health intervention programme (MEMA kwa vijana) in Mwanza, Tanzania: a community randomised controlled trial

Principal Investigators: David Ross, John Changalucha, Richard Hayes, Sheena McCormack, Deborah Watson-Jones

Project Coordinators: Aoife Doyle (LSHTM), Kaballa Maganja (MITU)

Funding: UK Department for International Development (DFID), and Irish Aid, Ireland

Staff learning to use PDAs in the field.

Background: Young people are particularly at risk from sexually transmitted infections (STIs), including HIV. In addition, they are also more likely than older adults to experience other reproductive health problems such as complications related to pregnancy and birth. In the absence of a vaccine or cure, sexual and reproductive health programmes, including behavioural interventions, have been advocated as the most effective HIV control strategy among adolescents. In order to assess the impact of an adolescent sexual and reproductive health programme, a community-randomised trial (MEMA Kwa Vijana) was conducted in 20 communities of four districts of Mwanza Region. The MEMA Kwa Vijana intervention was developed as a collaborative venture between AMREF Tanzania, the MRC in Glasgow, the LSHTM and the NIMR Mwanza Research Centre. The MEMA Kwa Vijana intervention had four components: (1) teacher-led, peer-assisted classroom sessions; (2) the provision of youth-friendly reproductive health services at local government clinics; (3) youth condom promoters and distributors; and (4) supportive community-wide activities.

The first phase of the trial commenced in 1998 and was completed in 2002. Process evaluation during the trial found high quality and coverage of the intervention. Results of the evaluation in 2001-2 (3 years after the start of the intervention) showed that the intervention had led to significant improvements in knowledge, reported attitudes and some reported sexual behaviours but did not lead to a significant reduction in the incidence of HIV infection and prevalence of HSV-2 and other STIs. However, during the 2001-2 survey, the trial’s power to detect a true difference in HIV incidence if one existed was very limited. In 2007-8, NIMR/MITU, in collaboration with LSHTM and other partners, did a survey to evaluate the long-term impact (8 years after the start of the intervention) of the MEMA Kwa Vijana intervention.

Investigator meeting with team leaders.

Objectives and study design: The primary objective of the long-term evaluation of MEMA Kwa Vijana was to investigate whether, in the longer-term, the intervention could have an impact on the prevalence of biological outcomes (HIV and HSV-2). Secondary outcomes included knowledge, reported attitudes, reported sexual behaviours, reported pregnancies, and the prevalence of other STIs. The survey was conducted in the 20 trial communities where the interventions were implemented. All young people who attended at least one year of primary education within school years (Standards) 5, 6 or 7 between 1999 and 2002 within a trial community and who were willing to provide informed consent were eligible to join this survey. 13,814 eligible young people were interviewed between July 07 and July 08. The median age of participants was 22 years for males and 21 years for females. Participants in the intervention communities had, on average, last been exposed to the in-school component of the intervention 5.4 years prior to the survey.

Census Team ready to depart for the field.

Main study findings: Correct knowledge and desirable reported attitudes related to sexual risk were higher in intervention communities than comparison communities. These differences were statistically significant or borderline significant except for the ‘attitudes to sex’ score in females.

There was no consistent or statistically significant impact on reported sexual behaviour outcomes or on the secondary biological outcomes, though condom use at last sex with a non-regular partner was reported significantly more often by females in intervention communities (adjusted relative risk 1.34 (95% CI 1.07, 1.69)).

The HIV and HSV-2 prevalences were similar in intervention and comparison communities and there was no evidence that the intervention led to a reduction or increase in these primary outcomes. The aRR for HIV was 0.91 (95% CI 0.50, 1.65) in males and 1.07 (95% CI 0.68,1.67) in females. Similarly, the aRR for HSV-2 was 0.94 (95% CI 0.77, 1.15) in males and 0.96 (95% CI 0.87, 1.06) in females.

Conclusions: Despite the failure of the MEMA Kwa Vijana intervention to reduce the prevalence of HIV and other STIs, the increase in and retention of knowledge is encouraging and important.

However, it was clear that these interventions on their own had not been sufficient to reduce HIV and other sexual risks among the young people within the trial. It is likely that wider societal norms related to adolescent sexual risk behaviours will need to be changed to permit this to occur. Local-level strategies to do this are currently being designed and pilot tested within the MEMA Kwa Jamii Project, while strategies to scale-up the existing core MEMA Kwa Vijana interventions in schools and health facilities have been tested within the MEMA Kwa Vijana Phase 2 (MkV2) Project.