You are on page 1of 12

Hewett et al.

BMC Public Health (2017) 17:386
DOI 10.1186/s12889-017-4280-1


Cluster randomized evaluation of
Adolescent Girls Empowerment Programme
(AGEP): study protocol
Paul C. Hewett1*, Karen Austrian2, Erica Soler-Hampejsek3,4, Jere R. Behrman5, Fiammetta Bozzani6
and Natalie A. Jackson-Hachonda7

Background: Adolescents in less developed countries such as Zambia often face multi-faceted challenges for
achieving successful transitions through adolescence to early adulthood. The literature has noted the need to
introduce interventions during this period, particularly for adolescent girls, with the perspective that such
investments have significant economic, social and health returns to society. The Adolescent Girls Empowerment
Programme (AGEP) was an intervention designed as a catalyst for change for adolescent girls through themselves,
to their family and community.
Methods/design: AGEP was a multi-sectoral intervention targeting over 10,000 vulnerable adolescent girls ages
10–19 in rural and urban areas, in four of the ten provinces of Zambia. At the core of AGEP were mentor-led,
weekly girls’ group meetings of 20 to 30 adolescent girls participating over two years. Three curricula ― sexual and
reproductive health and lifeskills, financial literacy, and nutrition ― guided the meetings. An engaging and
participatory pedagogical approach was used. Two additional program components, a health voucher and a bank
account, were offered to some girls to provide direct mechanisms to improve access to health and financial services.
Embedded within AGEP was a rigorous multi-arm randomised cluster trial with randomization to different
combinations of programme arms. The study was powered to assess the impact across a set of key longer-term
outcomes, including early marriage and first birth, contraceptive use, educational attainment and acquisition of HIV and
HSV-2. Baseline behavioural surveys and biological specimen collection were initiated in 2013. Impact was evaluated
immediately after the program ended in 2015 and will be evaluated again after two additional years of follow-up in
2017. The primary analysis is intent-to-treat. Qualitative data are being collected in 2013, 2015 and 2017 to inform the
programme implementation and the quantitative findings. An economic evaluation will evaluate the incremental cost-
effectiveness of each component of the intervention.
Discussion: The AGEP program and embedded evaluation will provide detailed information regarding interventions
for adolescent girls in developing country settings. It will provide a rich information and data source on adolescent
girls and its related findings will inform policy-makers, health professionals, donors and other stakeholders.
Trial registration: ISRCTN29322231. March 04 2016; retrospectively registered.
Keywords: Adolescent girls, Randomized trial, Multi-sectoral, Zambia, Empowerment, Savings, Financial education,
Nutrition education, Sexual and reproductive health, HIV, HSV-2

* Correspondence:
Population Council, 4301 Connecticut Avenue, Washington, D.C. 2008, USA
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (, which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
( applies to the data made available in this article, unless otherwise stated.

In Zambia. 8]. the Population Coun- annually through 2017. and weak institutions and laws [4. Not surprisingly. acquiring sexually transmitted infections (STIs). markets are primary drivers for the persistent disadvan- vention targeted to vulnerable. For in- arm. in- school leaving. working in collaboration with the YWCA. have low rates of condom use would percolate up to changes in the contextual envir- and have risk behaviours that increase their chances of onment within the family and community. The theory of change underlying variety of risks and challenges in achieving positive and AGEP (Fig. The Zambia DHS also indicates that girls in Zambia it was theorized that such changes at the individual-level begin sexual activity early. of school leaving for girls during the school age years is building confidence and efficacy. health and social nificant proportion of girls enter marriage and/or begin assets would facilitate positive change across a broad set childbearing early. Women and girls who have recruited a cohort of eligible adolescent girls invited to limited social capital and are more isolated engaged in participate in the programme and girls residing in con. and providing access to secondary age girls not currently attending school [1].7 years. both health and financial services and resources. and 15–19 reporting having an STI or symptoms of a the objective of representing vulnerable adolescent girls. the National Savings and Credit and endline results in 2017. Further. Duflo argues that inequality. are more likely exposed to sex- trol areas.Hewett et al. girls in many less developed settings are lacking the needed assets and capacities required to make more Girls group meetings positive and healthier transitions that would facilitate The AGEP intervention operated for a period of two their breaking out of persistent poverty and closing the years in each site from late 2013 and early 2014 through economic and livelihoods gap between men and women. lescent girls to build their economic. nearly one in ment. and three girls aged 20–24 had married by age 18. 5]. the median age of sexual debut is Intervention design 17. with mid-term results collected cil. [9–12]. resources for the programme and evaluation. tage of women relative to men [5] Other studies have cent girls ages 10–19. half of the sites were urban in premarital sexual activity. while a more general The literature indicates that women and adolescent sample of rural residential communities were targeted. a sig. As indicated by the 2013–2014 Zambia cluding early marriage and first birth. women and girls’ opportunities are programme was initiated sequentially by site.000 the prevalence of sexual initiation among 15–19 year adolescent girls ages 10–19 across ten sites in four of olds is 27%. including similar percentage having begun childbearing [1]. approximately 1 year prior to marriage. even prior to the expected age of of critical adolescent girls’ experiences and outcomes. Making in 2015 immediately after the end of the intervention Cents International. and The goal of AGEP was to reach a minimum of 10. Provinces were purposefully condom at last sex. sexually transmitted disease and 5% having acquired As the programme was specifically designed to reach the HIV. age of sexual debut. high-density urban time girls reach the ages of 20–24 years [1]. changing aspirations. weighing feasi- use implies exposure to STIs. with a transmission of sexually transmitted infections. rural and urban adoles. strengthening social high. of access to economic assets. Health designed AGEP. only 36% reported using a areas and half rural areas. Further. a multi-sectoral and girl- Adolescent girls in less developed countries face a centered intervention. 1) proposed that working directly with ado- successful transitions to adulthood [2]. housing compounds were targeted. with 4% of adolescents bility. adverse gender norms This paper reviews the design and methods of a multi. Embedded in the design of the directly linked gender power inequality to HIV risk be- programme was a rigorous. randomized cluster evaluation of the Adolescent stance. It was posited that an asset-building framework given rates of early marriage and childbearing. after the . opportunities and labour The programme was designed as a multi-sectoral inter. the risk operates by enhancing girls’ knowledge and capacities. Implementation of the In such settings. and roles. The HIV. with 18% of primary school age girls and 59% of networks. sexual risk behaviour. For instance. of those who are currently engaging the ten provinces of Zambia. baseline in late 2013 and early 2014 and is being tracked To address these disadvantages. riskier sexual behaviours. includ- ing HIV. poverty and the lack Girls Empowerment Programme (AGEP) in Zambia. The evaluation sample was interviewed at ual violence and less likely to be tested for HIV [7. BMC Public Health (2017) 17:386 Page 2 of 12 Background inhibited by traditional practices. schooling attain- Demographic and Health Survey (DHS). randomized evaluation that haviours and exposures [6]. late 2015 and early 2016. The low rate of consistent condom selected in collaboration with the donor. The prevalence of HIV more than doubles by the most-vulnerable adolescent girls. The theory underlying AGEP defines assets as a prevalence in rural areas is nearly double that of urban store of value that can be drawn upon by adolescent areas for both early marriage (39% versus 22%) and early girls to address challenges and overcome vulnerabilities childbearing (42% versus 21%) [3]. two years post programme Bank of Zambia (Natsave) and the Zambian Ministry of termination [1].

Monthly mentor [15]. school-based programs whether or not the girl group was located in a cluster exclude out-of-school girls. BMC Public Health (2017) 17:386 Page 3 of 12 Fig. A were invited to participate in AGEP girls’ groups. Eligibil- are adversely affected by an already low quality of ity for these add-on components was determined by schooling [13. receive a health voucher targeted to improve access to young woman from the community who had been ten general health and sexual and reproductive health trained to initiate and guide educational sessions and services at partner public and private health providers exercises on a variety of subjects. The girls’ groups were strati. most often on weekends. The meetings were led by an older. Two community-based model was selected as school-based additional programme components were randomly sexual reproductive health programs have been known assigned to some of the girls groups. 14]. 2) an offer to open a bank savings account. that was randomized to receive the additional compo- cluded in the intervention. The design and implementation of the AGEP trained in groups at central locations at the beginning of health voucher was done in consultation with the Minis- the programme. with one refresher training approxi. A published review of the literature found that meetings also took place with site coordinators to health vouchers had been successful at increasing service . In addition. which were meant to be in. Health vouchers lustrative vignettes. 1 AGEP Theory of Change baseline evaluation survey was completed in each site. try of Health at the national. which 20 to 30 girls met at a local community space for All the girls who were assigned to the intervention one to two hours per week. address mentor questions. The assignment of girls to the fied by age (10–14 and 15–19) and by marital/fertility programme component is discussed further in the evalu- status. role play and participatory methods Girls who attended AGEP in randomly-selected areas to to maximize impact. The additional to have limited impact because teachers find it hard to components were: 1) the provision of a health voucher break out of their typical teaching styles and programs and. experiences and provide short The core component of AGEP was group meetings in reviews of curriculum topics. nents of the intervention. aged or otherwise changed status. The mentors were (Table 2). provincial and district level mately one year into the program. The group sessions were based on three core curricula (Table 1) and used il. girls stayed with their original groups as they ation design section below.Hewett et al.

maternal mortality Life skills Healthy relationships. resolving conflicts over money. choose a savings goal. passive. (11 sessions) assertive and aggressive behaviour. child feeding and growth monitoring (3 sessions)† Note: Two additional sessions developed for the end of the programme that helped girls to plan for “life after AGEP”. strategies for delaying sex. planning income and expenses. myths and facts. own versus others money. weakness. avoiding and reporting sexual violence. HIV and (3 sessions) AIDS and human rights. goal setting. abortion and stigma. anaemia–causes and symptoms Nutrition in pregnancy. dealing with setbacks in saving. Stigma and discrimination Gender and gender-based violence Sexual exploitation. managing emotions. infancy & early childhood Nutrition needs during pregnancy.Hewett et al. noted that the breadth and scientific quality of evidence was limited Table 2 Services provided with health voucher and more rigorous assessments of effectiveness were 1 General physical exam and other non-sexual reproductive needed [16]. infant feeding. gender roles. opportunity and threats in achieving dreams (1 session) Saving and earning money Why save. communication. The review. BMC Public Health (2017) 17:386 Page 4 of 12 Table 1 AGEP core curricula implemented through weekly girls’ groups meetings Health and life skills Session topics Introductory sessions What to expect (×2 sessions). goal (9 sessions) identification. risky (6 sessions) behaviour. risky ways to earn money. sexual desire. banks and bank accounts. avoiding unintended pregnancy. personal relationships Reproductive health Life cycle. however. role of food in the body (dietary (3 sessions) diversity). saving regularly Managing money Safe places to save. controlling spending. options (10 sessions) for earning money. teamwork. decision-making. pregnancy. do’s and don’ts when talking about money. Financial education Dreams Strengths. unsafe abortion. in- 4 STI consultation and treatment for girl and her partner cluding improving provider knowledge around adolescent 5 HIV counselling. reproductive (9 sessions) myths. community service and action (2 sessions) Human rights Defining human and children’s rights. nutritional 10 Gender-based violence care and referral and general health knowledge building during the Note: Availability of services varied somewhat as not all facilities were able to weekly girls’ groups. testing and referral health issues. peer-pressure. AIDs and STIs Information on transmission. make a savings plan. There was also an additional session for girls in Arm 3 providing details about the savings account † denotes sessions only for girls who were aged 15 and older at baseline. relationship of STIs and HIV. HIV testing and counselling. self-esteem. These studies suggested the potential for improving SRH 6 Antenatal care and laboratory testing care among adolescents in Zambia and the value of imple- 7 Comprehensive abortion care menting a voucher programme embedded within a rigor- 8 Consultation for other sexual reproductive health issues ous evaluation design. body changes. 9 Cervical cancer screening In addition to the sexual and reproductive. resolving conflicts role play Good money management Journey to good money management (1 session) Nutrition Building blocks of nutrition Nutritional needs of adolescent girls (types of foods). rape and gender violence. sexual and reproductive health rights. utilization and improving the quality of health services provision across a variety of health care and developing country settings. two additional programme elements provide the full list of services supported the AGEP health voucher component: 1) the . attitudes and communication practices [17]. A study of a programme that provided health services health vouchers specifically to adolescent girls in 2 Family planning consultation and method Nicaragua found that it directly reduced many of the 3 Pregnancy testing and consultation barriers to adolescent sexual reproductive health care. (4 sessions) preventing unwanted advances Leadership Defining and identifying the qualities of leadership. drugs and alcohol. conflict resolution HIV. difference between needs and wants. communications. talking (7 sessions) about money. reasons for delaying sex.

Hewett et al. was hypothesized to reinforce girls’ money health and banking facilities. At and the ability to deposit or withdraw funds with no public facilities. the fees. The girls who were eligible for the savings adolescents. the health As noted previously.5 (US $0. AGEP reimbursements were low minimum opening balance of KW 2. facilitate economic independ. promote economic asset building. The sampling frames of management skills. and to the pilot period suggested that adolescent girls had a variety District Health Office for overall management of the of sources of income that may be used to build their sav- programme (5%). budgeting and savings. participating health facility providers and staff were had been shown in other contexts to improve financial given a two-day training on adolescent girl-friendly literacy. to the health facilities for supplies or loans were provided to AGEP girls. The bank account. AGEP was designed to serve a voucher started after one year of the programme and minimum of 10. HIV and STIs.000 vulnerable adolescent girls in continued one year after the AGEP girls groups ended. anchored to proximally-located cial services. Monthly monitoring and quality assurance visits crease savings and improve positive opinions about HIV to health facilities were also conducted [19]. Zambia aged 10 to 19 in ten sites located within four of the ten provinces of Zambia. Even in resource poor environments among vulner- designed for AGEP and supported by the Ministry of able populations. as health services are generally free. Savings accounts. the representation of the target population and consider- provision of a bank account was designed to provide a ation of the feasibility of operating AGEP while also mechanism for knowledge and skills to be operational. whose rates their own. while in rural areas sites were credit.50) based on negotiated fees for each service provided. It should be noted that unlike other programme elements. The study provinces and Savings accounts the number of sites per province were selected pur- While the girls groups provided capacity building in posefully through discussions with the donor regarding money management. All accounts. although for girls under age 18 years the ac- were agreed on with MoH and were uniform for all counts required their co-signatories to withdraw funds. sessions provided an overview of AGEP and the mecha. for per-diems to the district health staff to assist based on focus group discussions conducted during the in monitoring service provision (20%). conditions: that they were located within 15 km from a . The AGEP “Girls Dream” savings adolescents. separately. identifying and addressing barriers to ado. public facilities. and 2) a results-based finan. but also covered topics partnership with the National Savings and Credit Bank geared to improve the quality of service delivery for of Zambia (NatSave). conducting a randomized evaluation. For urban areas. were used as studies had shown in other settings randomly selected among areas that met the following limited results for such programs [20] and that micro. a list of high-density resi- ence and provide assets in cases of emergencies or other dential compounds was created. NatSave branch to provide additional information and to The results-based financing component of the health open accounts. including cash from parents. The training prevention methods [23]. The bank account component was implemented in nisms of the health voucher. Reimbursements were a function of ings. potential sites for urban and rural areas were developed grow a culture of savings. and sites selected ran- basic needs. Savings cing of service provision at health facilities [18]. the total number of girls with vouchers obtaining agricultural production. account were provided an orientation in the girls’ group lescent health care and the sexual and reproductive sessions and a field trip was organized to the nearest health rights of adolescents. Intervention population due to delays in finalizing the arrangements. savings accounts were found to in- Health. communicating and counselling money. At private and non-governmental tailored to the financial needs of the girls. Girls who were under the age of 18 years voucher also provided a basis from which to reimburse were required to have co-signatories present at the the health facility and provider for service use among account opening. during the period that the vouchers were usable. domly from the list. as opposed to micro. The bank account was specifically adolescent girls. when appropriately designed for adolescents. The girls who opened accounts could deposit on payments to the facilities for each service. BMC Public Health (2017) 17:386 Page 5 of 12 training of health facility staff in providing adolescent credit was not an appropriate first exposure to formal fi- girl-friendly health services. paid piecemeal work. Study sites within ized in practice. nancial services for vulnerable adolescents [21]. were distributed by fixed percentages: to While no direct financial resources in the form of grants the providers (50%). with a very organization facilities. including values clarification about family account provided adolescent girls a formal place to store planning. increase the self-efficacy and savings behaviours health services that followed a set curriculum specifically [22]. provided to girls provinces were selected randomly from a sampling who would not otherwise typically have access to finan. market research (25%). frame of potential sites. selling goods made and transac- services and the number of services that they accessed tional sex [24].

which also permits characterization of the were assigned the health voucher in addition to the vulnerability of girls younger than those who already weekly girls’ group sessions. CSAs contained approximately 750 in the programme with the most vulnerable girls se. the measure captures adolescents who are ing the bank account. Study outcomes The overall objective of AGEP was to make a meaningful Randomised cluster design change in the traditionally observed trajectories of vulner- The AGEP evaluation was designed as a multi-arm ran. Clusters assigned to the first from their participation in AGEP. It was also expected that the impact of the of AGEP. although the average number of households varies considerably as population densities vary geographically. BMC Public Health (2017) 17:386 Page 6 of 12 health facility and within 65 km of a NatSave bank Table 3 Randomisation arms of AGEP branch. intervention arm were assigned all components. Local political and community leaders were programme on these longer-term outcomes would be invited to participate in the lottery. per- The objective of the AGEP evaluation was to implement haps a few hundred metres long and wide.Hewett et al. inde- pendently for rural and urban areas. The as- bute the changes in girls’ lives to their participation in signment of CSAs to the experimental and control arms AGEP. The lottery was con- the girls acquire through the programme. Given that the ducted via a two-step selection process in which a CSA AGEP evaluation has multiple components that may was first randomly selected for participation among all lead to improved outcomes. The process of change was (arms) of AGEP. an indicator of vulnerability was created intervention arm were assigned only the weekly girls’ using whether girls were behind in school for their age group sessions. many children fall behind in school due itional component of the intervention. Instead. which was subsequently used as a sam. sessions + health voucher tics of their households. these indicators were not directly used to identify vulnerabil- ity. It was expected that the programme would have was conducted through a random selection process at a a positive impact on the longer-term demographic. with the more immediate the study are displayed in Table 3. vided for a rigorous assessment of the marginal benefit. while clusters in the final have been married. pregnant or dropped out of school. given the primacy of educa. public lottery. Evaluation methods and design CSAs in urban areas were spatially relatively small. as well as the incremental cost-effectiveness of each add- going process. changes in the girls’ assets and empowerment deriving ciated components. along with their asso. cluster. while CSAs a rigorous randomized design to more confidently attri. All girls selected to late entry. the study was designed and CSAs in the site and then an AGEP arm assigned statistically powered to evaluate the impact of each of through a second random selection process. areas. neither interventions nor placebo exposures. The experimental and control arms of posited to advance in phases. translating into positive . these components on adolescent girl outcomes. includ- Conceptually. becoming pregnant. the Zambia Central Statistical Office (CSO) for the 2010 pling frame to determine eligible girls for participation national census. and temporary with. Clusters in the second intervention arm as a proxy. Early in the school. Control No components of the AGEP intervention riage. repetition of grades. households in urban areas and 300 households in rural lected from the list to receive invitations to participate. all of these events are a reflection of to the intervention that was randomly selected for that some degree of personal and household vulnerability. girls in clusters for the control arm received The information used to develop the measure of vulner. in rural areas encompassed multiple kilometres. Study arm AGEP components AGEP was also designed to draw participants who Intervention arm 1 Weekly girls group mentor-led sessions expressed multiple levels of vulnerability through their Intervention arm 2 Weekly girls group mentor-led residential location and the socio-economic characteris. The public lottery was conducted to reproductive and health outcomes of participating maximize the transparency and community acceptance adolescents. for participation in AGEP within a cluster were assigned drawal from school. conducted at a mediated through a set of individual assets or skills that centrally-located public facility. tion in a child’s life at these ages. able adolescent girls in Zambia as they pertained to early domised cluster trial where clusters within the ten study family formation. low schooling attainment and poor sex- sites were randomised to receive different combinations ual and reproductive health. dropping out of school). ability was collected along with other household socio. A cluster in the AGEP evaluation was delineated by economic characteristics via a household survey in all Census Supervisory Areas (CSAs) that were compiled by study clusters. To avoid over-representing girls Intervention arm 3 Weekly girls group mentor-led sessions + health in AGEP who have already experienced the outcomes voucher + bank account that were to be prevented by the programme (early mar. The multi-arm evaluation pro- at risk of adverse outcomes.

The hypothesis tests ment and reducing the acquisition of HIV and the Herpes reflect the causal pathways as illustrated in the theory of Simplex Virus Type – 2 (HSV-2). Weekly girls groups meetings specific mechanisms. vices. friends who can be counted on Social Percentage of girls who hold positive gender normative believes Social Average score on financial literacy scale Economic Percentage of girls who have opened a formal bank account Economic Average score on knowledge of sexual reproductive health Health Average number of modern contraceptives known Health Percentage having comprehensive knowledge of HIV Health Percentage of girls accessing sexual and reproductive health services Health Mediating outcomes indicators Percentage of girls working for cash or in-kind in the past year Economic Percentage of girls with a modest amount of savings Economic Percentage of girls who have used a condom during last sex with a non-marital/non-cohabiting partner Sexual and reproductive health Percentage of girls who are using modern contraception Sexual and reproductive health Percentage of girls who have engaged in transactional sex Sexual and reproductive health Longer-term outcomes indicators Outcome domain Percentage of girls ever married Socio-demographic Percentage of girls ever pregnant and given birth Socio-demographic Percentage of girls experiencing unwanted pregnancy Socio-demographic Percentage of girls completing grade 7 and grade 9 Educational HIV prevalence among girls Sexual and reproductive health Herpes Simplex Virus Type-2 (HSV-2) prevalence among girls Sexual and reproductive health . 2) increasing self-confidence and weekly meetings in which girls met with a mentor and Table 4 Primary and secondary study outcome indicators and measures Immediate programme outcomes Asset domain Percentage of girls with high self-efficacy Social Percentage of girls with strong role models and social support Social Average number of friends. all three intervention arms included and building skills. The first set of hypotheses reflect the expectation As the concept of empowerment. by reducing early marriage. a array of assets that girls can build upon and utilize to set of representative indicators was used to epitomise their affect change. BMC Public Health (2017) 17:386 Page 7 of 12 behavioural change and. improved attendance in school and sexual and reproductive health knowledge. tions and. 3) increasing adolescent aspirations and motiva- over the course of four years in which empowered adoles. 4) improving access to key resources and ser- cents would realize more positive transitional outcomes. programme outcomes indicators in Table 4. reflected through HIV and pregnancy risk der normative beliefs. economic and health and included intervention on what are conceptualized here as mediat- measures of self-efficacy. social connections. change (Fig. The study will test several hypothesis to assess the im. are presented as mediating outcomes indicators in tional and health. that AGEP is having a strong and positive effect on the term outcomes are potentially broad and multifaceted. The indicators used to reflect these hypotheses ranged across three domains. The longer-term indicators of programme impact tunities. to longer-term change efficacy. A second The indicators of empowerment ranged across the broad set of hypotheses is made regarding the impact of the domains of social. Table 4. These are measured as immediate operationalization in the evaluation as noted in Table 4. access to savings reduction behaviours. Finally. 1) by improving factual knowledge As noted above. mediating and longer. ultimately. 1). educa. reflected in specific ways. The impact of the intervention is posited to have operated through four general and domain.Hewett et al. increasing educational attain- pact of AGEP and its components. socio-demographic. positive gen. the overall objective and ultimate goal. of the programme is to improve the wellbeing of adolescent Hypotheses girls. ing factors. early pregnancy and births. financial literacy. among and possessing greater economic resources and oppor- others.

Hewett et al. Baseline . reproductive health. sites. including given the budget available for fieldwork [25]. therefore. with the threshold determined by the total number of girls needed for recruitment into the Health voucher programme at each site. In the immediate randomized trial to determine the most efficient sample term. ples were stratified by the younger and older cohorts ception is hypothesized to reduce unwanted pregnancy (10–14 and 15–19) and urban and rural. that the provision of a bank ac- week. girls will groups in the weekly girls groups and certain aspects of have more positive assessment of their overall health. performance and. Table 4 served as the focal indicators for the sample size crease efficacy and decision-making control over the calculations. generally discuss their experiences and to socialize count will have positive impact on informal and formal with each other. The economic resources is hypothesized to reduce the need weekly girls’ group meeting is hypothesized. these changes are hypothesized to improve never married at baseline. The effect of sample size determination based on known effect sizes..0 for a multi-site through the weekly girls groups alone. budgeting and managing savings and expendi. 20% for the and treatment for STIs. attain. for comparisons between the intervention arms them- tures. Bank accounts The minimally detectable effect sizes were estimated As noted in Table 1 above. the bank account compo. As limited evidence was available for a utilization of health facilities and resources. money management plies and/or pay school fees. Participa. also hypothesized to depress the need for adolescent geted the girls’ social assets. It is per cluster was the most efficient combination. and improve mediating behaviours. The sam- acquisition in the longer-term. In general. Given the randomized design. while access to contra. In the domain of schooling at. these hypothesizes will result in followed a parallel process for identifying and selecting strong and positive relationships between the indicator vulnerable girls. by obtaining counselling. hypothesized to reduce the prevalence of sexual exchange tion in the groups is also hypothesized to reduce social and. is hypothesized to reduce HIV younger cohort and 35% for the older cohort. the curricula were adjusted for age-appropriateness by cohort. Sample the use of modern contraception. group. financial and nutritional knowledge. savings behaviour. Increased access and control over sible to assess the impact of the girls groups alone. ultimately. The group meetings also tar. including the number of girls to rely on older and potentially riskier partners and friends reported. for adolescent girls to rely on household resources from to have positive effects on the girls’ sexual and repro. which they have less control to obtain personal items ductive health. relationships and opportunities are. e. specifically a ranking of girls by their of randomization to the girls group only study arm and vulnerability score and randomly selecting girls below a the indicators noted along this continuum of outcomes. Access to bank accounts is and self-advocacy skills. the health voucher is hypothesized to increase the size (number of clusters and respondents per cluster) use of sexual and reproductive health services. selected among all invited AGEP participants who were tainment. For comparability. in particular. estimates of effect sizes were not obtained goals. set threshold. as well as to in. minimally detectable effects were estimated using Opti- guishable empirically over and above what is realized mal Design Plus Software Version 3. the weekly girls’ group for comparison of each programme component against curriculum includes 19 sessions on setting financial the control. the heath voucher component is expected to be distin. the longer-term outcomes in access to health facilities and resources. schooling sup- well as build their communications. reduce girls’ acquisition of sexually trans- isolation and provide access to community spaces. Research population and sample sizes move aspirations. The age stratifi- and birth outcomes. ultimately. it is hypothesized that as cation was necessary as there was no age mixing of specific health issues are prevented or resolved. If confirmed. Given a total of ten study entrepreneurial aspirations. the health funder and the programme was to improve the well- voucher intervention is hypothesized to directly increase being of adolescent girls. As the ultimate objective of the As with the bank account component. it was determined that a total of 40 clusters per nent of AGEP was designed to provide a direct link and arm (four clusters per arm per site) and 20 participants access to financial services to catalyse change. including HIV and HSV-2. stratified by site and age school participation. selection of the control girls ment. it is pos. Participants for the research study were randomly such as attending school. in-turn. the quality of those relationships and exchange-based sexual relationships. These changes are whether they report a role model in their lives. While the lessons themselves are hypothesized to selves as little prior information was available to guide directly increase knowledge and improve financial and such an estimation approach. Improved sexual and sizes were estimated for the evaluation after four years. care with conservative estimates of attrition. as such as clothing and hygiene products. These mitted infections.g. hypothe- sized to weaken regressive gender norms and roles. BMC Public Health (2017) 17:386 Page 8 of 12 each other to engage with a curriculum topic for the hypothesized.

With ACASI the respondent listened Research instruments with headphones to pre-recorded questions and response A household survey instrument was implemented prior to categories. ± 10 pp 18%. study. Never married girls were targeted dress potential spillover impacts. ± 14 pp 15%. ± 14 pp 89%. residential status were obtained from the 2007 Zambia recent household shocks.00. while. ± 6 pp 70%.80 to detect programme impact . may occur over time regarding: educational attainment The sample size calculations indicated that 3200 girls and schooling transitions. while facilities. financial literacy designed to serve as external controls given the potential and knowledge. power of 0. ± 10 pp 11%. ± 6 pp 6%. ± 14 pp Ever use of modern contraception 15%. study impact indicators: 800 girls (within 40 study clus. ± 11 pp 26%. ± 13 pp 13%. ± 8 pp 48%. high-density components. ± 9 pp Completed grade 9 28%. The screen. would have already experienced the key outcomes that Survey instruments were implemented by electronic were meant to be impacted by the programme. relationship sta- would be needed at the end of the study to assess all key tus and sexual partners. Statistical power distances to key resources. household asset ownership. All questionnaires were at baseline as a significant proportion of older girls translated into Nyanja. living arrangements designated to be sampled for urban areas in nearby and household resources. the sample size was inflated for non-eligibility and CAPI is a process of data capture in which the interviewer non-response at baseline. The would control for site fixed effects [25]. ± 10 pp 38%. HIV risk perception. n/a c 19%. ± 12 pp Notes: Younger cohort members are expected to be ages 14–18 at endline. ± 8 pp 21%. Surveys conducted after the baseline also for contamination between AGEP and control girls include questions of exposure to the intervention to ad- within the same site. ACASI maximizes confi- members with basic socio-demographic information on dentiality and privacy of response. including sexual behaviour. and 10 subjects per cluster b For each age cohort estimates are based on: 5 sites. To data capture.80. ± 13 pp 45%. ± 10 pp HSV-2 prevalence 21%. ± 12 pp 21%. Thus. Audio Computer-Assisted Self-Interviewing (ACASI) was used. gender ters) in each of the three AGEP arms and in the control attitudes. behav- imal detectable effect sizes. ± 12 pp 55%. housing quality. ± 10 pp 82%. etc. ± 8 pp 55%. BMC Public Health (2017) 17:386 Page 9 of 12 (control) estimates of study indicators by age group and each member. ± 10 pp 83%. by outcome and study iour. ± 8 pp 26%. ± 10 pp 60%. 1600 in internal control sexual violence. ± 14 pp Ever given birth 16%. ± 10 pp 19%. ± 7 pp 15%. and abortion. Computer-Assisted Personal-Interviewing achieve the target sample of 3600 girls at the end of the (CAPI) was used for questions that are non-sensitive. For sensitive questions. Table 5 provides baseline survey is included as an Additional file 1. ± 8 pp 54%. ± 14 pp Completed grade 7 61%.Hewett et al. All estimates assume alpha of 0. unwanted pregnancy clusters and 800 in external control areas. attrition from the cohort and reads the question from a computer screen and enters the refusals for biological specimen collection. ± 12 pp 55%. access to financial resources and Demographic and Health Survey [26]. labour force arm. ± 15 pp HIV prevalence 6%. as no one could hear or Table 5 Estimates of prevalence in the control and minimally detectable effect sizes in percentage points (pp) of programme impact by age-cohort and by rural and urban stratifications Totala Ruralb Urbanb Younger Older Younger Older Younger Older Ever had sex 40%.05 were set. the participant’s response directly into a handheld or tablet final estimated sample to be collected at baseline was device. and 10 subjects per cluster c Not powered at 0. ± 11 pp 26%. ± 11 pp 34%.efficacy and locus of control.80 and an alpha coefficient of 0. The estimates of prevalence in the control sample and min. ± 11 pp 76%. ± 14 pp 41%. ± 15 pp 79%.05. literacy. surveys instruments measure changes in attitudes. banks. n/a c 7%. simultaneously reading the question on the device for selecting cases in designated control clusters. transition status. 4 clusters per arm per site (40 clusters per arm). ± 9 pp 47%. ± 11 pp 17%. if desired and if the participant was liter- baseline to determine eligibility for the intervention and ate. These girls and clusters were numeracy and cognitive skills. and site fixed effects a For each age cohort estimates are based on: 10 sites. ± 9 pp 43%. sexual and physical coercion and violence. 7200 or 4800 in AGEP clusters. self. and ability. ± 6 pp 63%. Comprehensive annual surveys were the effect size variability was fixed at 0. n/a c 6%. The participant entered a response by touching a household instrument included a roster of all household designated number or option. An additional 400 girls (within 20 clusters) were participation and savings behaviour. older cohort members are expected to be ages 19–23 at endline. including schools. ± 13 pp 75%. marriage and marital dissolution. mobility and migration. Bemba and Kaonde. ± 11 pp 85%. health of 0. 4 clusters per arm per site (20 clusters per arm). sexual activity. ± 15 pp 64%. ± 4 pp 12%. roads. ± 13 pp Ever married 15%. as analyses conducted at baseline and for four subsequent years. ± 13 pp 10%. social assets and cognitive skills that stratifications.

testing procedures was conducted prior to testing of As girls were invited to participate in AGEP. The relevant costs for variable. over-identification will quantitatively assess the credibil- pating in AGEP. with the relative to the girls group intervention alone. the impact of AGEP on empowerment and longer- tory setting in rural Kenya [28]. HIV status was determined via capillary The primary objective of the evaluation is to assess blood draws obtained from finger pricks. constructed to generate estimates of the incremental Appropriate tests. including a controlled labora. safe sex practices and treatment options. The second stage will then use the predicted the programme services. difference-in-differences (DID) calculation will be pants were provided information about HSV-2 detection. many specimens for the main study to assure the quality of choose not to do so.Hewett et al. Indeterminate results date for any baseline differences between AGEP and were not retested since their prevalence was so low as to control clusters that may exist in the outcomes. The respondent was term outcomes. and the Hansen statistic for progress achieved on non-health indicators from partici. intermittently missed the weekly laboratory testing protocols. as well as dom assignment. intervention will be calculated to assess whether cost- Biological markers were collected from adolescents effectiveness varies by study arm. Both tests have a very underlying processes by which change occurred in the high sensitivity (100%) and specificity (>99%) in con. analysis will be conducted to understand the Determine™ and Uni-Gold™ [27]. After specimen collection all partici. Participant-specific out-of-pocket and stage will predict programme participation. A decision analytic model will be baseline covariates to test the relevant hypotheses. assessment after Round 5 the mediating and longer- ing health centers. and Round 5 (at the end of the Analysis evaluation). Important as Health guidelines. made for each indicator. HIV tests were the impact of AGEP on adolescents’ intermediate em- conducted by trained and certified voluntary counselling powerment and longer-term outcomes using the and testing (VCT) staff. It is important to recognize the statis- to assess the incremental cost-effectiveness of the add. To accommo- compared to Western Blot [30]. using the ex- indirect costs data for participation in AGEP and uptake ogenous randomised assignment as the key instrumental of services were also obtained. The HSV-2 biological specimens were collected via The average treatment effect of the programme on the finger prick. serum was derived from the whole empirically assessed.. The mid-term assessment symptoms. sessions or left the programme before the programme was completed. the Wald F-statistic. key areas of interest. we conducted serial testing using well. for which the that are both unadjusted and adjusted for baseline co- sensitivity and specificity have been found to be high variates. including F-tests on excluded instru- costs per negative health outcome averted and positive ments. after Round 3 (at the end of AGEP). Therefore a secondary analysis will be Economic evaluation conducted that uses a measure of participation intensity The study also includes an economic evaluation to rigor. as the indicator of impact rather than the study’s ran- ously assess the cost of implementing AGEP. A two-stage instrumental vari- programme delivery costs for the Population Council ables estimation approach will be used in which the first and its partners. while in the final numbers to receive their test results at AGEP participat. Post-test counseling was guided be conducted using the original randomised assign- by National Guidelines for Testing and Counselling [29]. not including the costs related participation intensity from the first stage along with to project evaluation. In accordance with Ministry of above-noted hypothesis as test criteria. A sample of whole blood was collected and adolescents in clusters randomised to AGEP relative stored in microtainers and laboratory tested. observed baseline covariates will also be these analyses are the real resource costs of delivering included. tical estimation problems that arise from self-selection on components (health voucher and savings account) into AGEP. after Round 3 focused on the immediate and mediat- spondents were provided vouchers with identification ing indicators as primary outcomes. Direct characteristics of girls predicting both their uptake of programme costs were collected from AGEP budgets and exposure to the programme and key outcomes and and financial reports and included both start-up and behaviours of interest. and the degree of participation in it. both approaches accounting for strata and (92% and 98%) respectively in African populations when intraclass correlations within clusters. BMC Public Health (2017) 17:386 Page 10 of 12 see the question being read. aged 15 and older at baseline (Round 1). To conduct to girls in clusters randomised to the control will be the laboratory test. a make it impractical. . an intent-to-treat analysis (ITT) will provided the test results. A validation of the HSV-2 laboratory term outcomes will be the empirical focus. Incremental cost-effectiveness ratios ity of the selected instrumental variable. The analysis will estimate models blood and the Kalon ELISA assay used. ment to study arm as the primary indicator of impact. As a first approach for assessing trolled clinic evaluations. nor the response option (ICERS) comparing the add-ons to the girls group only selected. Re.

financial literacy. contributed text and provided and feedback. a lack of Empowerment Programme. The rigorous randomized evaluation and cohort study to evaluate AGEP outlined in Additional file 1: AGEP Baseline Survey. cation and sexual and reproductive health.Hewett et al. as demo. Human Services 1991). (XLSX 2922 kb) this protocol paper was designed to address information gaps and the weaknesses recognized in the existing litera. information on vulnerable and disadvantaged populations. ESH. including sexual and http://www. The AGEP programme is unique in the amount government’s official policies. with the expectation that improved outcomes for girls would lead to life-long improvements in the well-being of Additional file women and their children. Association graphically the “cohort of young people age 10–24 is Acknowledgements the largest in history. in the areas of early marriage and childbearing. The authors also thank the UK Department for International Development (DFID) core concept of mentor-led. KA. All study investigators and study research protocol included an embedded economic staff were certified in human subject’s protection training prior to study evaluation that collected programme implementation and program. the benefits from finding out what Bank of Zambia. assent and guardian written informed consent was obtained for all participants under the age of 18. Valerie Roberts who whose primary objective was to build and empower girls has been a champion of AGEP. PcH. enhanced the potential impact external reviewer. JB. and FB designed the study with review and suggestions from UKAID-DFID staff and an independent consulting company hired as an count and health voucher. The economic evaluation will provide and older. NJ. The Population Council IRB has established procedures that adhere to the U. added in terms of impact. ESH. AGEP: Adolescent Girls ture. Federal guidelines for human subjects as set forth in Title about the investment needed to scale the programme 45. NATSAVE: National Savings and Credit the literature. KA. Part 46 of the Code of Federal Regulations (Department of Health and and the incremental costs of investments in each inter. and other stakeholders to develop policies and programs Ethics approval and consent to participate The research protocol was approved by the Population Council Institutional for adolescent girls. consortium members. edu- prove important adolescent transitions of girls in Zambia. AGEP was not. reproductive health. .” while epidemiologically devel. To achieve this additional aim. the first for the behavioural survey and the second for the an opportunity to compare AGEP with alternative biomarker data collection. the Zambia’s Research Ethics Committee (UNZA-REC – protocol 008–11-12) with federal-wide assurance number is FWA00000338. All study participants provided written consent if they were age 18 participant costs. by 1) enhancing their knowledge and skills. We thank the lescent health problems” [32]. CSA: Census Supervisory Area. KA and JB drafted the study protocol and of the group learnings by providing girls direct access to instruments. under what conditions and how for adoles. Materials related to this study can be found on the Population Council: ment across a range of domains. Abbreviations ACASI: Audio Computer-Assisted Self-Interviewing. UNZA-REC: University of works. which has a Department of Health and the investment needed for added programme com. government. weekly girls’ group meetings for investing in rigorous evaluation and. ESH. PcH drafted the manuscript. FB. BMC Public Health (2017) 17:386 Page 11 of 12 Discussion programmes that address the needs of adolescent girls AGEP was an intervention whose objective was to im. FB. a bank ac. PcH. the National Savings and Credit Bank of Zambia and the are “characterized by high levels of all types of ado. MB reviewed resources that typically did not exist for them or have the draft manuscripted. The multi-arm randomized evalu- ation was designed to assess whether each of these Competing interests programme components provided additional value The authors declare that they have no competing interests. oping countries face multi-burdens from diseases and the Young Women’s Christian Association (YWCA). life skills and nu- trition. Government of the Republic Of Zambia. HSV-2: Herpes simplex virus. high barriers to use. AGEP is being implemented by a consortium led by the Population Council. CAPI: Computer Assisted Personal Interviewing.S. in particular. Delineating costs of the programme Review Board (PCIRB – protocol #581). As noted in Development. 2) building Funding their self-confidence and efficacy. Making Cents International. DFID-Research and Evidence Division. and 4) providing access to services and re. JB. with over two years of Availability of data and materials weekly programme activities devoted to girls’ empower. Ministry of Health. limited to the weekly Author’s contributions girls groups. Consent for publication Determining the impact of AGEP is important in order Not applicable.popcouncil. and Human Services (DHHS) federal-wide assurance number of ponents can provide additional critical information FWA00000279. however. of investment in this effort. STI: Sexually transmitted infections. All authors read and approved the final manuscript. and Mott MacDonald. YWCA: Young Women’s Christian cents in a setting such as Zambia is large. The AGEP programme and evaluation was funded by UK aid from the UK rations. Research Ethics Committee. MOH: Ministry of Health. as well as members of the AGEP Evidence Scrutiny AGEP was a multi-sectoral intervention built off of the Committee. to provide needed evidence for governments. 3) changing their aspi. The protocol was also reviewed by the University of vention component. including non-randomized study designs. HIV: Human immunodeficiency virus. A two staged consent process was cost-effectiveness information for each study arm and conducted. type-2. DFID: Department for International and short-term durations of assessments [31]. Zambia. donors. as two additional components. however the views expressed do not necessarily reflect the UK sources.

Performance of commercially available enzyme immunoassays for detection of antibodies against herpes simplex virus type 2 in African & HIV in the 21st Century. P. 2005. 11. Trop Med Int • We provide round the clock customer support Health. Gender. Nguyen V-K. 23. USA. A novel economic intervention to reduce HIV risks among school-going AIDS orphans in rural Uganda. Mshana G. D. United States Agency for International Development: 5 Departments of Economics and Sociology. The use of vouchers for reproductive • Our selector tool helps you to find the most relevant journal health services in developing countries: systematic review. 26.S. Chong E: Evaluation of a savings and micro-credit program for Tropical Medicine. World Bank: Washington. World Bank: A guide to competative vouchers in health. Salam RA. conventional testing Technologies for Human Immunodeficiency Virus Type 6. Culture Hayes RJ. University of Zambia. Das JK. Plot 3670. Mazige B. Ashburn K. Providing access to economic assets for girls and interventions in developing countries: making the investments count. Adolescent health interventions: 9. 2016. N. 2011. 12. 4.. 2004. 3Population 20. 2010.C. ZMIR61FL. Sateren W. Birx DL et al. 2008. Torres V: Adolescent Girls Empowerment Program 1. 32. and Testing of Children. No.: The International Bank for Reconstruction and Development/The World Bank. Padian NS.42(7):2961–5. Duflo E. Ssewamala FM. 2011. Glynn JR. 2004(Aug 2004):3850–5852. Robb ML.A. 6London School of Hygiene and 21. Gendered socioeconomic conditions and HIV risk behaviours 29. Washington D. Philadelphia. Malia JA. Todd J. Dickson K. program review. USAID. Fluvia 101. Foglia GF. In. In. 2(6):889–898. 2008. In. Girl centered program design: a toolkit to develop strengthen & expand adolescent girls programs. Bruce J. Hallman K: Social exclusion: The gendering of adolescent HIV risks in 30.C. 7. University of London. Bangpan M. Mwaleshi Road. Kenya. Edited by Klot JF. Use of rapid and 2012. 3o 4a.. Ministry of Health and ICF international 2014.S. Maryland Making Cents International. 2010.A. New York. In. Ross D. Makokha M. G. France: populations.50(4):1051–79. 2007. Weiss HA.4(1):37–50. health voucher mechanisms. Council. Can economic empowerment reduce vulnerability of girls and young women to HIV? Emerging insights. J Econ Lit. D. Institute of Education. microfinance and sustainable livelihoods approaches: a literature and 4 Independent consultant. • We accept pre-submission inquiries 16. Calverton. Laila Rahman. 2011. Microbiol. New York. Brown DW. D. P. 15-17 Tavistock Place.O. Central Statistical Office (CSO) MoHM. Spybrook J. Demographic and Health Survey 2007. Olympia Park. In. UNESCO. Population Council: Curriculum on adolescent-friendly health services and 1 Population Council. 2011. 2012. Locust Walk. One Dag Hammarskjold Plaza. Pettifor AE. 19104. Hill C. Alicea S. 3718 Washington D. Akter R: Enhancing adolescent financial capabilities through financial education in Bangladesh. 25. Rob U.59(3):S1–3..S. evidence gaps. Population Council: Received: 11 February 2017 Accepted: 21 April 2017 Nairobi.attitudes and practices related to sexual and • Thorough peer review reproductive health care for adolescents? A case study from Latin America.42(1):102–4. London WC1H 9SH. South Africa. In: The Fourth Wave: Violence. a workshop on adolescent girls' livelihoods. New York.10(11):1996–2004. Sexual power and HIV risk.59(S4):S88–92. 17. KwaZulu-Natal.S. Congdon R. Calero EK. Paris. Plummer MJ. Martinez A.A: Population Council. 2008. 22. NY 10017. Rees HV. J Clin South Africa Emerg Infect Dis. African Journal of AIDS Research. A systematic J Adolesc Health. National HIV/AIDS/STI/TB Council: Zambia HIV Rapid Test Algorithm. Zambia: Ministry of Health. National Research Council and Institutes of Medicine: Growing up global: Design Plus Empirical Evidence: Documentation for the “Optimal Design” The changing transitions to adulthood in developing countries. Erulkar A. Bannon WM Jr. Adolescent girls livelihoods 2016. Lassi ZS. Warner A. 2012. meeting:. Zambia.Hewett et al.C. World Bank. Maryland USA: Central Statistical Office. U.A. 7Population vulnerable young women in Nairobi. Austrian K.S.A. [Zambia]. Muthengi E. Nairobi. 2015. Donald Royster IV. and we will help you at every step: 15. Women empowerment and ecnomic development. J Adolesc Health. Sebstad J: Building assets for safe and productive lives: A report on . 28. Mani R. • Inclusion in PubMed and all major indexing services Trop Med Int Health 2006. Warren CE. Gorter AC: Health Vouchers for Adolescent Girls Empowerment Program. Unpublished report: Independent consultant. U. De Deken B. NY U. In. review of the evidence. Submit your next manuscript to BioMed Central In. 4. Ministry of Health Macro International Inc. University of Pennsylvania. and ICF International: (TDRC). 2005. Measham DM. Spain. • Maximum visibility for your research C. and ICF international. In. Central Statistical Office (CSO) MoHM. Tropical Diseases Research Centre 3. James-Wilson D.biomedcentral. Population Council: New York. Rockville. Lusaka. 27. Hayes R.16(1):84–96. Zaman B. Amin S. Health Educ Res. International Center for Research on Women: Washington D. and Macro International Inc. 13. 2012. Zambai: National HIV/AIDS/STI/TB Council. Kenya. Rockville.: National Academies Press. [Zambia]. USA. Austrian K. van Dyck E. 10101 Lusaka.A. and ICF International: Youth-Inclusive Market Research Report-Round One In. 14. Reducing adolescent Girls' vulnerability to HIV infection: examining Council. Maryland: CSO and DTA. UK. 2005. BMC Public Health (2017) 17:386 Page 12 of 12 Author details 19. 8.: Zambia Zambia Demographic and Health Survey 2013–2014 [Dataset]. In. Meuwissen LE: Can a comprehensive voucher programme prompt changes • Convenient online submission in doctors’ knowledge. Wight D. 4301 Connecticut Avenue. References 24. Raudenbush S: Optimal 2. USA: Central Statistical Office. Lusaka. Bhutta ZA. Bloom H. In. U. Buve A. 2009. J Clin Microbiol. Government 2005. 2013. 2008. World development report: gender equality and development. Ministry of Health GotRoZ: Zambia National Guidelines for HIV Counseling among young people in South Africa. of the Republic of Zambia. Box 17643-00500. 2014. Washington. New York: Population Council. Parry J. Ismayilova L.C U. Hallman K. Fatusi A. Wamoyi J. 1 serologic screening in a rural Kenyan reference laboratory. NY: Population Council.C. young women in low-and-lower middleincome countries. Bellows NM. Unpublished report: Zambia Demographic and Health Survey 2013–2014. U. 5. Safe and smart savings products for vulnerable adolescent girls in kenya and uganda: evaluation report. Editorial: young People’s sexual and reproductive health 10. Ghati D. 18.22(4):500–12. 08019 Barcelona. 2012. Washington Software In. 2004. Obasi A. J Adolesc Health. Renzullo PO. A process evaluation of a school-based adolescent sexual health intervention in rural Tanzania: the MEMA kwa Vijana programme. USA. 2004. Bellows B.Y. Central Statistical Office (CSO) MoHM. In. and research priorities.A. Ainul S.: Population Council. London: EPPI-Centre. New York. 2010. Social Science Research Unit.S. Submit your manuscript at www. Kibaya R. NY. 31. 2 Population Council. Wasunna KM. edn.