A community-driven approach to reducing teenage pregnancy in Sierra Leone: midline evaluation brief
Preliminary evidence from a 2009 global evidence review suggested that community-based child protection mechanisms are likely to be more effective and sustainable if they are linked with formal aspects of the child protection system. To test the value of nonformal-formal linkages, this action research uses a quasi-experimental design to test the effectiveness of a community owned and driven intervention that seeks to reduce teenage pregnancy.
In each of Moyamba and Bombali Districts, there were two clusters of three communities in different but comparable chiefdoms. One cluster was an intervention cluster, whereas the other was a comparison cluster. In the intervention clusters, community members from three villages worked collaboratively to develop an intervention that addressed a child protection concern of their choosing. In both intervention clusters, the communities elected to focus on teen pregnancy, an issue that had been documented as a key concern in previous ethnographic work. The intervention, which was developed by the community, included components on family planning, sexual and reproductive health education, and life skills and was implemented in partnership with NGOs and District Ministry of Health partners.
- exposure to the intervention was found to increase with age, and was particularly high among individuals who had a partner but were unmarried. The latter individuals had 1.7 times the odds of attending a presentation when compared to those who were married
- between baseline (T1) and T2, the percentage of teenagers aged 15-17 in intervention areas who were willing to ask their partners to use a condom increased by 17.1 percentage points, whereas the control villages showed a decrease of 6.2 percentage points
- girls and adolescents under 15 years in intervention areas showed a significant increase in their intention to use condoms regularly, while the opposite was true in control areas
- similarly, adolescents who were exposed to the interventions programs were nearly twice as likely (1.74 times the odds) to report intending to say no to unwanted sex
- sexual activity was also found to have increased between T1 and T2 in control areas, while no increase occurred in intervention areas, suggesting possible signs of impact of the program
It is evident that certain groups, particularly those over 18 years, currently have a higher likelihood of accessing the presentation component of the intervention. While this is a positive development, particularly if high access among unmarried adolescents with a partner indicates a desire to prevent pregnancy before marriage, it may also indicate a need for additional strategies for engaging younger populations. Collectively, and given the short window of time in which it has been active, the findings presented in this brief suggest positive trends and signal that the intervention is moving in the right direction with the potential for a much larger impact in the coming years.