Adolescent Transition in West Africa (ATWA) Outcome Evaluation: 2022-2023 Endline Report

"ATWA led to many significant improvements in SRHR [sexual and reproductive health and rights]-related knowledge, attitudes and behaviour among in- and out-of-school adolescents of all age groups, including significant improvements in gender equitable attitudes."

This report shares the results of the Year 2 outcome evaluation of the Adolescent Transition in West Africa (ATWA) programme. Implemented in Mali, Burkina Faso, and Niger, the four-year (2019-2023) culturally-adapted programme was designed to influence and promote positive adolescent sexual and reproductive health and rights (ASRHR) outcomes among very young and older adolescents (10-14 and 15-19 years). The evaluation used a longitudinal cluster randomised trial with a comparison group to measure the effects of delivering ASRHR education curricula in schools by teachers and in small peer education groups on very young and older adolescents. ATWA was implemented by a consortium that included Save the Children-Netherlands along with Save the Children country offices in Mali, Burkina Faso, and Niger and research partners Center on Gender Equity and Health at the University of California San Diego (UCSD/GEH) and GRADE Africa. 

In each of the three countries, the programme worked with teachers, mentors, and peers to engage adolescent girls and boys on SRHR topics to improve their attitudes, behaviours, and norms towards gender and SRHR and to build critical life skills. The ATWA Life Skills Education (LSE) curricula were tailored and adapted for each country context by the Ministries of Education in Burkina Faso, Niger, and Mali in collaboration with various agencies and further adapted in collaboration with Save the Children. Alongside the provision of LSE, ATWA sought to increase access to, and uptake of, quality adolescent-responsive SRH services among adolescents by working with service providers to improve the quality and youth-friendliness of services (this component of the project was evaluated in a separate study not included here).

The overall objective of this evaluation study was to assess the ATWA programme's impact in improving knowledge, attitudes, behavioural intentions, and other precursors of behaviour change related to ASRHR and menstrual hygiene and management. Data collection through a questionnaire occurred in November-December of 2022 (baseline, n=2,113 adolescents) and in May-June of 2023 (endline, n=1,998). The evaluation had a high follow-up rate from baseline to endline (94.6%), with most, but not all, adolescents followed over time. Adolescent respondents answered questions about puberty; menstrual hygiene and management; HIV, sexually transmitted infections (STIs), and pregnancy knowledge; contraception; personal sexual history; health services; gender attitudes and norms; gender-based violence (GBV); self-efficacy; child marriage; and female genital cutting (FGC). The study also assessed the impact of peer education in the intervention group by comparing those who received peer education (PE) and those who did not.

The following are some of the results as highlighted in the Evaluation Results Brochure summary (see link below), with more results, including results on the impact of peer education, in the full report:

Puberty: Adolescents were asked a series of nine questions to assess knowledge about body changes during puberty, the menstrual cycle, fertility and conception, pregnancy risk, and prevention. There was a 10.1 percentage point increase in correct knowledge among control group adolescents (34.5% of adolescents answering all nine questions correctly at baseline compared to 44.6% at endline) and a 22.1 percentage point increase from baseline to endline among intervention group adolescents (27.9% to 50.0%). This difference in change between the two study arms over time was statistically significant, indicating that participation in the ATWA intervention was associated with adolescents being 2.6 times more likely to have accurate knowledge about puberty (odds ratio (OR)=2.63, p<0.001). 

Menstruation: Accurate knowledge and beliefs about menstruation were low among all adolescents (control and intervention) at baseline, with only around 15% of boys and 19% of girls correctly answering at least two of three true or false questions: Girls' menstrual cycles are typically 14 days long; menstruation is dirty/impure; and used sanitary napkins can be used to cast magic spells. Significant increases were seen among the intervention group, with 46% of ATWA adolescents (compared to 25% of control adolescents - a 21 percentage point increase) answering at least 2 of the 3 questions correctly - corresponding to a 2.6 times greater likelihood among ATWA participants in having accurate knowledge about menstruation (OR=2.62, p<0.001).

STI and HIV prevention and treatment: The findings showed a significant increase in knowledge among adolescents in ATWA, relative to those in the control group. Adolescents in the control group saw a 1.6 percentage point increase in correct knowledge, from 55.5% at baseline to 57.1% at endline. On the other hand, ATWA adolescents saw a 16.6 percentage point increase, from 53.9% at baseline to 70.5% at endline. This increase corresponded to ATWA adolescents being 2.6 times more likely to have correct knowledge of STI risk, prevention, and treatment than adolescents in the control group (OR=2.63, p<0.001).

Contraceptive methods: When asked whether they planned to use modern contraceptive method options in the future, the study saw a 12.6 percentage point increase from baseline to endline in the control group (from 45.3% at baseline to 57.9% at endline), compared to a 19.3 percentage point increase in the intervention group (51.4% to 70.7%). This difference in change between the study arms over time was statistically significant, indicating that participation in the ATWA intervention was associated with adolescents being 1.7 times more likely to intend to use modern contraception (OR=1.68, p=0.041). 

Adolescents' views on child marriage: At endline, a large majority (94.2%) of respondents strongly agreed that early marriage often has harmful consequences for a girls' health, psychological wellbeing, and education. More adolescents in the ATWA intervention strongly agreed that child marriage has harmful consequences than adolescents in the control group (96.8% versus 92.0% of controls (χ2 p=0.005)), indicating positive intervention effects on this indicator. In addition, at endline, a majority (71.9%) of all adolescents strongly agreed that they have the right to choose the conditions of their marriage (in other words, whether, when and with whom they want to marry), and more ATWA adolescents than control group adolescents held this view (78.6% versus 65.5% of controls) (χ2 p<0.001)), indicating positive intervention effects.

Gender attitudes and norms: The proportion of adolescents with gender-equitable attitudes was evaluated based on the Gender Equitable Attitude scale, a composite of six items assessing the degree of acceptance surrounding gender equity in domains such as household chores, education, and males having the final say in household decision-making. At endline, the average score of all respondents on the Gender Equitable Attitude scale was 3.5 out of 5 (the average of responses across all items on 5-point Likert scales). At endline, ATWA adolescents had higher average scores than adolescents in the control group (3.9 versus 3.1 among controls (p<0.001). This intervention effect was particularly strong for boys who had lower agreement at baseline, but saw greater increases in scores on gender equity by endline. This finding indicates that the ATWA intervention was particularly useful in impacting boys' attitudes towards gender-equitable behaviours and decisions.

Attitudes on GBV: At endline, the vast majority of all respondents agreed that no one has the right to force another person to engage in sexual activity (93.4%). More adolescents in the ATWA intervention agreed with this statement relative to adolescents in the control group (95.9% versus 91.2% in the control group (χ2 p=0.043)). Interestingly, a smaller proportion of adolescents strongly agreed (65.4% in the overall sample) that it is unacceptable to use pressure or force to persuade a girlfriend to have sex if she didn't want to, with ATWA having a positive intervention effect on levels of agreement that it is unacceptable to pressure or force a girlfriend to have sex (68.7% versus 62.4% (p<0.001)).

Views on FGC: The research found that a majority (92.8%) of respondents stated a girl should not be circumcised. When stratified by subgroup, a lower proportion of adolescents in the control group (89.2%) compared to the intervention group (96.7%) agreed that girls should not be circumcised, and this difference reached statistical significance (χ2 p<0.001).

Self-efficacy: A number of survey items were included to assess adolescent self-efficacy with respect to general self-efficacy, self-efficacy surrounding puberty/change, self-efficacy surrounding unwanted sexual attention, self-efficacy in practicing safe sex, and ability to resist peer pressure. The study found positive intervention effects in adolescent self-efficacy on all 11 of the self-efficacy measures. For example, 95.6% of the ATWA adolescents strongly agreed that trying hard can improve their situation in life, compared to 86.9% of the adolescents in the control group (χ2 p<0.001).

In conclusion, the report states that overall, "ATWA's Year 2 outcome evaluation findings show many significant improvements in SRHR-related attitudes and knowledge among adolescents of all age groups. These changes may be harbingers of future shifts in ATWA long-term outcome measures as adolescents progress in their pubertal development and grow into young adults. Additional rounds of data collection could examine this longitudinal connection between attitudes and knowledge in adolescence on behavior and SRHR outcomes into young adulthood. Further rounds of data collection could also provide insights into whether ATWA's immediate intervention effects on knowledge and behavior are sustained over time."

Click here in order to download an 8-page summary brochure sharing the results of this evaluation.

Click here in order to download the French version of this 108-page report in PDF format.