Effectiveness of a girls’ empowerment programme on early childbearing, marriage and school dropout among adolescent girls in rural Zambia: study protocol for a cluster randomized trial

Approximately 7.3 million girls below age 18 give birth in low- and middle-income countries (LMICs) every year [1]. Maternal complications are estimated to be the fourth most common cause of death in girls aged 15–19 in LMICs [2], and the risks of prematurity and low birth weight are high in adolescent pregnancies [3‐5], with consequent higher morbidity and mortality [4, 6‐10]. The younger the mother is, the higher is the risk of complications both for her and the child [3], and childbearing before age 16 is of particular concern [4, 6‐8].

Early pregnancy is often associated with early marriage and school dropout, and poverty contributes to all three. Observational studies from low-income countries indicate that young women who quit school early are more likely to marry and become pregnant earlier than those who stay in school [11‐15]. Increased schooling has also been associated with better health of women and their children [16, 17]. In the last decades primary school enrolment has increased significantly in many poor countries. However, enrolment at secondary level is much lower than at primary level in most LMICs, particularly for girls [18]. This may be due to limited availability of and longer distances to school, higher fees, or to early marriage or pregnancy. Moreover, there may be a preference to support boys’ education rather than girls’ [19]. In many societies marrying off a girl may be regarded as better to secure her future than schooling, and the bride-price paid to the girl’s family may be an important source of income. Once a girl is married, she is expected to start childbearing. Moreover, where access to cash is severely limited, many unmarried adolescent girls engage in sexual relationships, even if not socially acceptable, to receive gifts and cash [20‐22].

We did a systematic search¹ in 16 databases² for randomized controlled trials (RCTs) in LMICs that assessed interventions targeting adolescent childbearing and marriage. Recent systematic reviews on adolescent pregnancy [23, 24] and childbearing [25] and early marriage [26, 27] were also examined for relevant original articles. We focused on trials that included girls younger than 18. Three trials examined the effects of support to girls to reduce the cost of schooling. In Kenya, girls provided with free school uniforms were less likely to drop out before completing primary school, and proxy reports by classmates indicated reduced risks of early marriage and childbearing [28]. A trial in Zimbabwe found that a programme targeting orphan girls, offering payment of school fees and free uniform, led to an 80% reduction in school dropout, and 60% reduction in marriage rates in the next 2 years [29]. A trial in Malawi found that payment of school fees combined with small cash transfers to adolescent girls and their families resulted in lower prevalence of human immunodeficiency virus (HIV) and Herpes simplex virus type 2 (HSV-2) in the next 18 months among girls enrolled in school at baseline [30]. The trial also found that unconditional cash transfers led to reduced marriage and pregnancy rates whereas no significant change was seen in these outcomes in the group provided with cash transfers conditional on school attendance [31]. Quasi-experimental evaluations of unconditional and conditional cash transfer programmes in Kenya [32], Pakistan, [33] and Mexico [34] have also found effects on adolescent childbearing and/or marriage. More recently, two RCTs on the effects of conditional cash transfer programmes on HIV have been conducted in South Africa [35]. They found no effect on HIV incidence [22]. Thus effects of cash transfers appear to be context dependent [24], and there is a need for further evaluations of such programmes in new settings, alone and in combination with other interventions [22].

Comprehensive sexual and reproductive health education (CSRHE) programmes can equip adolescents with knowledge and skills to prevent unwanted pregnancy even if they are sexually active. Reviews of evaluations of sexual and reproductive health education programmes worldwide indicate that effective curricula are intensive [36], include student active teaching [37], and incorporate discussions around gender and power dynamics [38]. Our systematic literature review of RCTs identified five cluster randomized controlled trials (CRCTs) from LMICs that studied the impact of sexual and reproductive health (SRH) education programmes on early childbearing and marriage. The findings were mixed: a CRCT in Kenya found that pupils exposed to a curriculum informing them of the higher risk of HIV infection associated with having older sexual partners, had 28% lower pregnancy rates 12 months later [39], whereas training of teachers in the national HIV/AIDS curriculum promoting sexual abstinence until marriage, indicated no effect on adolescent childbearing after 2, 3, 5 and 7 years [28, 39, 40]. This is in line with a systematic review of randomized studies assessing abstinence-only programmes in the US, finding no protective effects on self-reported sexual behaviour, sexually transmitted infection (STI) diagnosis, or pregnancies [41]. In Uganda, a CRCT assessing a combination of a life skills programme (which included comprehensive sexual and reproductive health education) and vocational skills training, found lower probability of adolescent girls in the intervention arm having children after 2 years compared to the control [42]. In South Africa, a CRCT evaluated the impact of a 50-hour interactive HIV prevention programme focusing on SRH and gender relations was evaluated. The intervention was delivered over 8 weeks to girls and boys (in- and out-of-school) aged 16–19. After 2 years the proportion of girls who were pregnant tended to be higher in the intervention arm (OR 1.45; 0.92–2.28), there was no difference in HIV incidence between the two arms, whereas the HSV-2 incidence was 33% lower in the intervention than control group [43]. Another CRCT in South Africa studied a comprehensive teenage pregnancy prevention programme with 12 weekly sessions among pupils in grade 8 and found no impact on pregnancy rates after 8 months [44]. In Tanzania, a CRCT examined the effects of an intervention with the following four components: (1) a teacher-led, SRH school programme in grades 5–7; (2) training and supervision of health workers in youth-friendly health services; (3) community-based condom promotion and distribution by adolescents; and (4) 1-week mobilization in each community and annual youth health weeks. An evaluation 3 years after the start of the intervention found effects on SRH knowledge and self-reported condom use among the youths, but no impact on the incidence of HIV, HSV-2 or pregnancy despite high coverage and high quality of the implementation [45]. Qualitative evaluation during the first years of the intervention indicated that adolescents found it difficult to use the new knowledge and skills because of attitudes and practices in the community. The authors suggest this may imply that SRH education may not be effective on its own, but needs to be combined with efforts to “address broader sexual norms” and future aspirations [46‐48].

The need to find ways to prevent early marriage and pregnancy is high on the political agenda in Zambia. As many as 35% of young rural 18-year-old girls have given birth, and the median age at marriage was 18 years in 2013/2014 [49]. According to the 2010 Zambia Census of Population and Housing (ZCPH), the pregnancy-related mortality ratio among girls aged 15–19 years is 80% higher than among those aged 20–24 [50]. Adolescent childbearing rates are much higher in rural than urban areas, and higher among girls who are out-of-school compared to those still attending (36% vs 5% at age 17 in rural areas) (unpublished data from 2010 ZCPH).

In preparation for this trial, we conducted formative research to explore contributors to early childbearing, marriage and school dropout. In-depth interviews (IDIs) and focus group discussions (FGDs) were conducted with girls, parents, teachers, health workers and community leaders in December 2014 and January 2015. Poverty was raised as a primary reason for early childbearing, early marriage and school dropout; girls mentioned a desire for gifts/money as an important reason for engaging in sexual relations; and parents put pressure on their daughters to get married to secure their daughters economically and to obtain the bride price. Moreover, many families could not afford uniforms or school fees at secondary school level. Other factors leading to early pregnancy were misconceptions linked to contraception, e.g. that girls could end up as infertile if they use hormonal contraception before their first pregnancy, social barriers to seeking contraception, and school dropout. Girls who have dropped out of school were regarded as ready for marriage after menarche. The SRH curriculum in Zambian schools is comprehensive on paper [51], but the interviews with teachers indicated that in practice a high proportion of them tend to focus on sexual abstinence as the only way to avoid pregnancy or sexually transmitted infections (STIs), including HIV. Hence misconceptions about modern contraceptives can prevail. Pregnancy was also described as leading to school dropout and marriage, although some girls reenter school after having given birth.

Since early pregnancy and marriage in Zambia have multiple causes, our contextual assessment implies that it is likely that a programme that targets several contributing factors will have a greater effect than a single focus intervention. This is in line with the existing literature, which has increasingly recognized that multicomponent interventions are needed to achieve a substantial impact on complex issues such as adolescent childbearing, early marriage and school dropout [52‐55]. Further, as long as a high proportion of girls never enrol in secondary school, preventing early marriage and childbearing also needs to reach out-of-school girls. Informed by our literature review, the formative research, discussions with key stakeholders such as the Ministry of General Education (MoGE), the Ministry of Community Development and Mother and Child Health (MCDMCH) and local stakeholders, we developed an intervention package that targets what we identified as the main causes of early pregnancy: (1) an economic component targeting poverty and school dropout, with the aim to increase school attendance and secondary school enrolment, and to reduce parental pressure for early marriage and girls’ dependence on having a boyfriend to receive basic goods; and (2) a community component, including a youth club, that aims to enhance SRH knowledge and skills, and perceived supportive community norms regarding pursuit of education and postponement of pregnancy and marriage. Our hypothesis is that the timing of support is important, and payment of school fees, which reduces the cost of schooling, may be decisive at stages when children approach transition points in the education system, e.g. between primary and secondary school, when costs would otherwise increase.

As far as we are aware, this will be the first cluster RCT to measure the effect of a package combining economic support and a community component to prevent adolescent childbearing in a LMIC. The intervention components have been carefully selected to ensure they will be feasible and sustainable to implement. Increased schooling among adolescent girls is likely to empower them economically [61] and cognitively, and combined with postponed childbearing this can enable them to better protect the health of their children [17] and themselves and moreover increases the probability that their future children will complete secondary school [5]. The findings from this study will be highly relevant for programmes aiming to improve adolescent reproductive health in Zambia and in similar contexts.