Balancing Quality, Intensity and Scalability: Results of a Multi-level Sexual and Reproductive Health Intervention for Very Young Adolescents in Kinshasa

"[T]hese findings underscore the potential of even a light-touch intervention to improve knowledge, assets, and gender-equitable attitudes. "

Growing consensus among researchers and public health practitioners recognises early adolescence as a critical time for early intervention to promote healthy development and foster more gender-equitable attitudes and norms among very young adolescents (VYA), aged 10-14, to improve life-long sexual and reproductive health (SRH) outcomes. In Kinshasa, Democratic Republic of the Congo (DRC), Growing Up GREAT! (GUG) used a scalable approach to engage in- and out-of-school VYAs, caregivers, schools, and communities. The purpose of this paper is to present the results of a mixed-methods study consisting of a quantitative impact evaluation and an implementation research study to evaluate how GUG contributed to changes in VYAs' knowledge, attitudes, and behaviours.

As detailed at Related Summaries, below, GUG was implemented by Save the Children during the 2017-2018 school year. The multi-level programme used a gender-transformative approach to engage VYAs, their families, and communities in critical reflection of gender norms. In recognition of the social and institutional factors influencing VYAs' gender attitudes, behaviours, and norms. school- and community-based clubs engaged 2,000 in-school (IS) and 350 out-of-school (OOS) VYAs, respectively. Clubs used a toolkit with interactive, age-appropriate materials to learn about SRH and explore related gender and social norms. Supplemental activities implemented by partner community-based organisations (CBOs) engaged family, school, community, and health system actors who interact with VYAs. For example, over 2,000 caregivers and 2,200 community members participated in six hour-long reflective group discussions, held monthly. Discussions centred around testimonial videos featuring local parents who adopted positive behaviours such as talking to children about SRH, prioritising girls' education, and dividing household responsibilities equally among male and female children.

The quantitative impact evaluation was conducted by the Global Early Adolescent Study (GEAS), a multi-country longitudinal study exploring gender socialisation in early adolescence and its implications for adolescent health and wellbeing. Focusing on two economically poor urban communes (Kimbanseke and Masina), GEAS used a quasi-experimental longitudinal design with intervention and control arms, each divided into two subgroups (IS/OOS). In total, 2,519 VYAs were included in this analysis (n = 704 OOS and n = 1,815 IS). The baseline survey (Wave One) was conducted between June and November 2017. Adolescents were followed up one year later (Wave Two), three months after the GUG intervention ended.

This paper focuses on GUG's four intermediate outcomes: SRH knowledge and communication, assets and agency, gender-equitable attitudes and norms, and gender-equitable and nonviolent behaviours. Difference-in-differences (DID) analysis was completed to compare trends in outcome indicators from baseline to follow-up between study groups. A qualitative study completed post-implementation included in-depth interviews with 34 teachers, health providers, and CBO staff, as well as six mystery client observations.

Results show significant improvement in SRH knowledge and assets such as caregiver connectedness, communication, and body satisfaction among the intervention group. The intervention was also associated with significant improvements in gender-equitable attitudes related to adolescents' household responsibilities and decreased teasing and bullying. Intervention effects on awareness of SRH services, body satisfaction, chore-sharing, and bullying were stronger for OOS and younger VYAs, suggesting the intervention's potential to increase positive outcomes among vulnerable adolescents. The intervention did not shift key gender norm perceptions assessed.

Though it is not realistic to expect such low-intensity interventions to shift norms in the short term, the researchers discuss several aspects of GUG design and implementation that may have contributed to the mixed results, and they make associated recommendations, as follows:

1. Facilitator competency may have been adequate to instill information but not to support the nuanced process of dialogue and consensus-seeking that leads to norm shifts. Compressed trainings (to support scalability) did not provide enough time to build capacity in this facilitation style, nor allow for deep reflection on teachers' own attitudes and norm-driven beliefs. Data also portrayed a challenging environment in schools, which faced both resource scarcity and logistical issues.
2. Inconsistent participation among VYAs and caregivers due to the challenging urban context undoubtedly influenced outcomes. Additional research and implementation-based evidence on successful and feasible approaches for engaging VYA caregivers to support SRH and norm change is needed. Identifying ways to strengthen male caregiver participation may require directly addressing prevalent norms attributing caregiving responsibilities to female family members, in addition to ensuring activities are accessible and attractive uses of limited free time.
3. Differing results across VYA sub-groups point to the need for improved segmentation and understanding of VYAs' needs by gender, age, school status, and other key socio-demographic markers.
4. GUG's experience raises considerations for balancing the scalability of multilevel norm-shifting efforts with VYAs with their potential impact, including tensions between adequate intensity and quality of interventions and the ease with which they can be sustained within existing systems.

In conclusion: "Results affirm the potential of early intervention to increase SRH knowledge, assets and gender-equitable behaviors. They also highlight the need for more evidence on effective program approaches and segmentation for shifting VYA and SRH norms."