Cluster mean difference-in-difference results of the impact of women's group intervention on women's health indicators
| Control | Intervention | DID* estimate (95% CI) | p Value | |||
|---|---|---|---|---|---|---|
| Pre | Post | Pre | Post | |||
| Sexual health | ||||||
| Unmet need for spacing (%) | 5.5 | 4.8 | 5.2 | 3.8 | −0.7 (−2.5 to 1.0) | 0.391 |
| Unmet need for limiting (%) | 21.6 | 20.6 | 22.5 | 17.6 | −3.9 (−7.1 to −0.7) | 0.018 |
| Total unmet need (%) | 27.1 | 25.3 | 27.7 | 21.4 | −4.6 (−8.0 to −1.2) | 0.009 |
| Knowledge of ≥3 methods of modern contraception (%) | 86.5 | 87.7 | 92.9 | 98.3 | 4.2 (2.0 to 6.3) | <0.001† |
| Women accessing modern contraception (%) | 51.0 | 52.1 | 54.1 | 57.5 | 2.4 (−1.3 to 6.1) | 0.209 |
| Awareness of ≥3 STIs (%) | 0.8 | 0.7 | <0.0 | 15.8 | 16.0 (14.7 to 17.2) | <0.001† |
| Awareness of ≥1 way to prevent STIs (%) | 36.2 | 43.5 | 18.4 | 96.8 | 71.0 (68.0 to 74.1) | <0.001† |
| Awareness of ≥1 way to treat STIs (%) | 34.3 | 51.8 | 24.5 | 97.4 | 55.4 (52.2 to 58.5) | <0.001† |
| Reported ideal age for marriage (years) | 18.5 | 18.6 | 18.4 | 18.3 | −0.1 (−0.3 to <−0.1) | 0.014 |
| Reported ideal age for first pregnancy (years) | 20.9 | 21.0 | 20.7 | 20.7 | −0.1 (−0.3 to <0.1) | 0.069 |
| Mean ideal spacing between pregnancies (years) | 4.6 | 4.5 | 4.7 | 4.0 | −0.6 (−0.7 to −0.5) | <0.001† |
| Nutritional health | ||||||
| Women's dietary diversity score (mean) | 3.9 | 4.1 | 4.0 | 4.3 | 0.2 (0.1 to 0.3) | 0.002† |
| Night blindness (%) | 12.3 | 12.5 | 6.3 | 9.6 | 3.0 (0.8 to 5.2) | 0.008 |
| Knowledge of ≥3 ways to maintain good nutrition (%) | 27.0 | 29.3 | 45.6 | 94.5 | 46.6 (43.6 to 49.6) | <0.001† |
| Knowledge of ≥3 ways to prevent anaemia (%) | 1.8 | 2.2 | 3.2 | 66.2 | 62.8 (60.9 to 64.6) | <0.001† |
| Morbidity | ||||||
| Any illness or injury over the previous 3 months (%) | 21.3 | 10.0 | 19.4 | 11.0 | 3.0 (0.2 to 5.7) | 0.033 |
| Sought care for mild or severe self-reported illness (%) | 92.1 | 93.2 | 92.0 | 93.4 | 0.2 (−5.2 to 5.6) | 0.938 |
| Women's participation in healthcare decision-making (%) | 44.7 | 33.5 | 55.6 | 58.3 | 14.0 (10.6 to 17.4) | <0.001† |
Response rates for all outcomes were 89% or higher, apart from care seeking, which was 20.3% and 10.4% for pre-intervention and post-intervention surveys, respectively, because it was only reported if the respondent had been ill in the previous 3 months. Also, dietary diversity score response rates were 93.3% and 87.4% because scores were not recorded if the respondent had fasted or feasted in the 24 h before the survey.
*Difference-in-difference derived from a random effects model interaction term between pre/postsurvey and control/intervention arm and accounting for the stratified, clustered study design.
†Significant at the 0.05 level using the Bonferroni correction for multiple testing (ie, p value of 0.003).
STIs, sexually transmitted infections.