Quasi-experimental evaluation using confirmatory procedures: A case study of an economic and social empowerment intervention to reduce girls' vulnerability to HIV in rural Mozambique.

PURPOSE: Our goal was to determine whether an economic and social empowerment intervention implemented in Zambézia Province, Mozambique reduced girls' vulnerability to HIV. We use this experience to discuss challenges of evaluating real-world interventions. METHODS: Two rounds of data were collected from 885 girls, 13-19 years, for this clustered, non-equivalent (two-stage) cohort trial. We used multi-level exact matching and difference-in-differences estimation to estimate intervention effects on two outcomes: girls' knowledge of gender-based violence and school attendance. RESULTS: Estimates of two outcomes analysed indicated no statistically significant intervention effects. Preliminary analysis of data from the intervention group revealed this study was unable to obtain accurate measures for five outcomes related to HIV vulnerability. CONCLUSIONS: Although our study did not find evidence of impact on the a priori selected outcomes, we report on our experience implementing this robust methodologic design and describe how the challenges encountered in this program setting affected our ability to attain results. We recommend prospective evaluation designs with random allocation be accommodated early during planning. When not possible, quasi-experimental studies should collect data from large samples. To reduce measurement bias, biological endpoints such sexually transmitted infections should serve as primary outcomes for programs intending to reduce sexual behaviors.

A longitudinal qualitative evaluation of an economic and social empowerment intervention to reduce girls' vulnerability to HIV in rural Mozambique.

PURPOSE: An intervention including business training and health education was implemented in Mozambique, where girls are at elevated risk for acquiring HIV. As part of a mixed-methods evaluation, we describe perceived effects of the intervention on girls' sexual behavior and school attendance. METHODS: We conducted 49 in-depth interviews (IDIs) with girl intervention participants (ages 13-19), 24 IDIs with heads of girls' households, 36 IDIs with influential males identified by girls, and 12 focus group discussions with community members after the intervention ended and one year later. RESULTS: Informants said the primary intervention benefit was realized when girls had money to stay in or return to school and/or to buy necessities for themselves and their households-reducing their need for transactional or intergenerational sex. However, some girls did not make a profit and some businesses were not sustainable. Sometimes the intervention appeared to be implemented in a way to reinforce inequitable gender norms resulting in some girls feeling shame when they reengaged in risky sex after their businesses failed. CONCLUSIONS: Earning money enabled girls to potentially reduce their vulnerability to HIV. We offer recommendations for future multi-sector interventions, including the need to address potential harms in programs serving vulnerable girls.

Correlates of Rapid Repeat Pregnancy Among Adolescents and Young Women in Uganda.

CONTEXT: Short pregnancy intervals can contribute to maternal and child morbidity and mortality. No previous research has explored factors associated with short pregnancy intervals among young women in Uganda, where adolescent pregnancy and short birth intervals are common. METHODS: Data on 626 married or cohabiting women aged 15-22 with one or two previous pregnancies were drawn from the 2011 Uganda Demographic and Health Survey. Bivariate and multivariable logistic regression analyses were used to examine characteristics associated with rapid repeat pregnancy, defined in two ways: a pregnancy occurring within 24 months or 12 months of a prior pregnancy outcome. RESULTS: Among women, 74% and 37% had experienced a rapid repeat pregnancy within 24 months and 12 months, respectively. Rural women were more likely than urban women to have had a rapid repeat pregnancy within 24 months (odds ratio, 2.4). Women aged 15-17 and those 18 or older at first union were more likely than women younger than 15 to have had a rapid repeat pregnancy within 24 months (3.8 and 3.4); those whose partner had at least a secondary education had lower odds than others of the outcome (0.6). The odds of rapid repeat pregnancy increased with the number of months between marriage and first birth (1.05). Variables associated with rapid repeat pregnancy within 12 months included urban-rural residence, region, age at first union and marriage-to-birth interval. CONCLUSIONS: Efforts to reduce rapid repeat pregnancy among young women in Uganda should focus on rural areas. Strategies to reach women during antenatal care and the postpartum period after their first birth should be prioritized.

Assessment of the psychometric properties of HIV knowledge items across five countries.

HIV knowledge is commonly measured in HIV prevention research and program evaluations, but rigorous measurement standards are not always applied. Using item response theory methods, we examined the psychometric functioning of five commonly used HIV knowledge questions in five countries with varying HIV prevalence. We evaluated the internal consistency and measurement invariance of the items. The items performed poorly in all samples and the scale as a whole did not perform equally across samples. We conclude that current ways of measuring HIV knowledge are not adequate and recommend new items be developed, tested, and validated using psychometric methods.

Qualitative study of reasons for discontinuation of injectable contraceptives among users and salient reference groups in Kenya.

Discontinuation of contraception is a major problem in Kenya. Even though they want to space or limit their births, over a quarter of contraceptive injectable users discontinue use of the method within 12 months of beginning use. Fourteen focus group discussions were conducted in Nyando District, Kenya among current contraceptive injectable users and their salient reference groups (e.g. husbands, mothers-in-law, community leaders, service providers) to understand why women discontinue using contraceptives (with a focus on injectables). Thematic analysis was performed using NVivo 8 software. Discontinuation of contraceptives in Nyando District occurs for logistical, social and medical reasons. Common reasons for discontinuation include side effects, husbands' opposition, provider and/or clinic restrictions, misconceptions about injectables, stock outs, and lack of cash to pay for family planning services. This research expands the literature by examining social influences on discontinuation by incorporating the perspectives of salient reference groups. The results suggest points of intervention for increasing continuation in this community and similar resource-poor settings.