Table 3.
Estimated causal effect of DREAMS on social support and self-efficacy in 2018 and in 2019
| Outcome and year | Setting | Age group | % with outcome in total study population | Estimated % with outcome if none benefit from DREAMS | Estimated % with outcome if all benefit from DREAMS | Difference in estimated % |
| % (95% CI) | % (95% CI) | % (95% CI) | ||||
| Social support, 2018 | Nairobi | Overall | 50.7 | 40.2 (33.1 to 47.6) | 52.7 (48.5 to 56.7) | 12.5 (4.2 to 20.9) |
| 15–17 | 51.5 | 34.4 (24.8 to 44.3) | 55.5 (50.4 to 60.8) | 21.1 (9.6 to 31.8) | ||
| 18–22 | 49.6 | 47.6 (38.4 to 56.6) | 49.2 (42.8 to 55.6) | 1.6 (-10.1 to 13.3) | ||
| Gem | Overall | 36.2 | 27.9 (23.9 to 32.9) | 43.3 (39.6 to 46.8) | 15.4 (10.2 to 19.8) | |
| 13–17 | 33.8 | 26.9 (22.0 to 35.2) | 39.7 (35.4 to 43.6) | 12.8 (4.0 to 19.8) | ||
| 18–22 | 39.6 | 28.8 (22.5 to 35.5) | 47.8 (42.9 to 53.4) | 19.0 (10.6 to 27.1) | ||
| uMkhanyakude | Overall | 45.7 | 41.7 (38.4 to 45.4) | 49.4 (46.1 to 52.8) | 7.7 (2.5 to 12.0) | |
| 13–17 | 47.1 | 42.8 (38.2 to 47.9) | 49.6 (45.6 to 53.3) | 6.8 (-0.3 to 12.9) | ||
| 18–22 | 44.0 | 40.4 (35.9 to 45.1) | 49.2 (43.2 to 55.0) | 8.8 (1.6 to 15.8) | ||
| Social support, 2019 | Nairobi | Overall | 56.3 | 49.4 (42.6 to 56.2) | 58.2 (54.1 to 62.3) | 8.8 (1.2 to 16.7) |
| 15–17 | 57.3 | 46.1 (36.2 to 55.6) | 60.0 (54.6 to 65.0) | 14.0 (3.0 to 25.0) | ||
| 18–22 | 55.2 | 53.3 (43.9 to 61.3) | 56.0 (49.5 to 62.5) | 2.6 (-7.3 to 14.1) | ||
| Gem | Overall | 40.4 | 35.5 (31.1 to 39.6) | 43.3 (39.3 to 47.3) | 7.8 (2.1 to 14.0) | |
| 13–17 | 37.0 | 31.2 (26.0 to 37.3) | 40.7 (35.7 to 46.3) | 9.5 (1.9 to 17.3) | ||
| 18–22 | 44.9 | 43.5 (35.7 to 51.1) | 48.2 (41.6 to 55.0) | 4.7 (-4.5 to 14.1) | ||
| uMkhanyakude | Overall | 45.4 | 44.2 (40.5 to 47.7) | 46.0 (42.5 to 49.5) | 1.8 (-3.2 to 6.6) | |
| 13–17 | 45.4 | 43.2 (38.0 to 48.1) | 46.7 (42.7 to 50.7) | 3.6 (-2.6 to 10.2) | ||
| 18–22 | 45.5 | 45.6 (41.2 to 50.4) | 44.9 (39.1 to 50.4) | −0.6 (-7.5 to 6.6) | ||
| Self-efficacy, 2018 | Nairobi | Overall | 53.7 | 54.7 (46.4 to 60.7) | 54.9 (51.2 to 59.2) | 1.2 (-6.9 to 10.0) |
| 15–17 | 52.1 | 53.7 (43.3 to 64.5) | 52.1 (47.4 to 57.3) | −1.6 (-13.7 to 9.9) | ||
| 18–22 | 55.7 | 53.6 (43.8 to 63.2) | 58.4 (52.8 to 64.5) | 4.8 (-6.4 to 16.6) | ||
| Gem | Overall | 37.2 | 36.1 (30.9 to 40.6) | 38.5 (35.3 to 41.8) | 2.4 (-4.1 to 8.5) | |
| 13–17 | 32.8 | 32.7 (27.1 to 37.7) | 34.7 (29.8 to 38.9) | 2.0 (-4.3 to 9.6) | ||
| 18–22 | 43.5 | 40.3 (34.5 to 46.0) | 43.4 (36.7 to 48.9) | 3.2 (-3.9 to 9.7) | ||
| uMkhanyakude | Overall | 41.6 | 37.9 (34.5 to 41.4) | 46.4 (42.9 to 49.6) | 8.6 (3.4 to 13.1) | |
| 13–17 | 36.4 | 32.4 (27.6 to 37.3) | 39.4 (35.6 to 43.3) | 7.0 (0.9 to 12.9) | ||
| 18–22 | 48.3 | 44.9 (40.6 to 49.8) | 55.5 (50.0 to 61.4) | 10.6 (3.2 to 17.8) | ||
| Self-efficacy, 2019 | Nairobi | Overall | 54.6 | 50.5 (43.8 to 58.2) | 56.3 (52.2 to 60.2) | 5.7 (-2.7 to 13.9) |
| 15–17 | 53.2 | 50.6 (41.3 to 61.2) | 53.9 (49.0 to 59.0) | 3.3 (-8.7 to 14.3) | ||
| 18–22 | 56.2 | 50.5 (41.9 to 59.5) | 59.1 (53.2 to 65.5) | 8.6 (-3.0 to 19.8) | ||
| Gem | Overall | 34.5 | 31.8 (27.3 to 36.0) | 35.6 (31.1 to 39.0) | 3.8 (-4.0 to 9.1) | |
| 13–17 | 30.1 | 27 (21.6 to 31.7) | 32.8 (28.2 to 37.8) | 5.7 (-0.1 to 13.4) | ||
| 18–22 | 41.4 | 41 (33.6 to 49.8) | 41.2 (34.2 to 46.6) | 0.2 (-11.7 to 9.6) | ||
| uMkhanyakude | Overall | 48.4 | 45.5 (42.1 to 49.3) | 51.0 (47.4 to 54.4) | 5.4 (0.5 to 10.1) | |
| 13–17 | 42.6 | 38.7 (33.6 to 44.1) | 45.3 (41.4 to 49.7) | 6.7 (0.3 to 12.6) | ||
| 18–22 | 56.1 | 54.5 (49.9 to 59.4) | 58.3 (52.2 to 64.0) | 3.8 (-3.6 to 11.2) |
2018 denominator: AGYW followed up in 2018 (Overall totals: Gem 1171; Nairobi 836; uMkhanyakude 1853).
2019 denominator: AGYW followed up in 2019 (Overall totals: Gem 1018; Nairobi 852; uMkhanyakude 1712).
Method: Propensity-score regression adjustment.
Outcome definition social support: Binary outcome variable constructed where a high level of social support was defined as a ‘yes’ response to at least three out of four questions: ‘Is there a female in your community from whom you can borrow money in an emergency?’; ‘Do you have at least one trusted female friend?’; ‘Do you know a woman in your community, other than a mother or guardian, whom you could turn to if you had a serious problem?’; ‘Do you have a safe and private place to meet with girls and young women who are like you?’Outcome definition self-efficacy: Binary outcome variable constructed based on a series of ten questions comprising a general self-efficacy scale, where a cut-off value of ≥3.5 was used to define higher self-efficacy (yes).
AGYW, adolescent girls and young women; DREAMS, Determined, Resilient, Empowered, AIDS-free, Mentored and Safe.