Building the assets to thrive: Addressing the HIV-related vulnerabilities of adolescent girls in Ethiopia

In the last decade, the shape of the HIV epidemic in Ethiopia has changed dramatically. New HIV infections among adults have declined by a remarkable 90 percent (UNAIDS 2012) and new infections among children by 50 percent (UNAIDS 2013). These declines demonstrate the success of national and international commitments to combat HIV and of the initiatives to increase access to HIV prevention and treatment services born of this commitment.

As of 2014, the adult HIV prevalence rate in Ethiopia was estimated at 1.2 percent. In 2012, there were some 41,000 deaths from AIDS and 759,000 people living with HIV (NARC 2012). In the early 2000s, however, when the Population Council launched the research that evolved into the project discussed in this report, the circumstances of the epidemic were very different. The first cases of HIV infection in Ethiopia were documented in the mid-1980s, and prevalence rates were estimated to be 6.2 percent of the adult population in 1993 and 7.3 percent in 2000 (MOH 2000). By 2007 over 975,000 people were living with HIV (MOH 2007).

The Council’s experiences in Ethiopia show that when HIV prevention programs are shaped by evidence and designed for replication and scale-up, they can reach large numbers of the people at greatest risk and increase their ability to avoid infection

It was in this context that the Population Council set out to identify Ethiopia’s most vulnerable populations of young people. We found that adolescent girls—particularly married girls and those living in urban slums—were among the most vulnerable and isolated populations and faced a heightened risk of HIV infection. In response, the Council and the Ethiopian government launched “HIV Prevention for Vulnerable Adolescent Girls in Ethiopia,” a project that builds girls’ social, health, and economic assets so that they can avoid HIV, sexual and gender-based violence, and other risks.

The three programs that make up this project seek to reduce Ethiopian girls’ HIV risk by using similar methods to engage girls—and, in the case of one program, the males who play a role in their health and well-being. Biruh Tesfa mobilizes out-of-school adolescent girls in urban slums, providing them with adult female mentors and offering education on HIV and AIDS and related issues, as well as non-formal education and links to health services. Meseret Hiwott uses a similar model to reach married adolescent girls in rural areas. Addis Birhan , an offshoot of Meseret Hiwott, gathers husbands of adolescent girls into discussion groups that focus on promoting care-giving to wives and children and addressing extramarital partnerships, alcohol abuse, and violence.

Reaching vulnerable adolescent girls with information and connecting them to services are not straightforward tasks. Poor girls in Ethiopia have few opportunities to access public institutions such as youth or community centers, health services, financial institutions, and schools. They may not know that they have a right to these services, and service delivery staff might not welcome them. But the Council’s experiences in Ethiopia show that when HIV prevention programs are shaped by evidence and designed for replication and scale-up, they can reach large numbers of the people at greatest risk and increase their ability to avoid infection.

We are proud of the Council’s contribution to the remarkable decline of HIV in Ethiopia. But there is clearly more to do to achieve an AIDS-free future. We hope that other organizations and agencies will be able to use the information in this report to launch and expand successful asset-building programs to help adolescent girls achieve a healthy and productive future.

Ann K. Blanc, Vice President and Director, Poverty, Gender, and Youth Program, Population Council
Naomi Rutenberg, Vice President and Director, HIV and AIDS Program, Population Council

Three Programs: Design and Goals

In December 2007, the Population Council was awarded funding from the US President’s Emergency Plan for AIDS Relief (PEPFAR) for a project titled “HIV Prevention for Vulnerable Adolescent Girls in Ethiopia.” The goal was to create, implement, and evaluate interventions to build adolescent girls’ social, health, and economic assets so that they could avoid HIV infection, gender-based violence, and other risks. This project was one of the first large-scale, multisectoral efforts to address HIV and other risks faced by adolescent girls in the developing world.

Our work in Ethiopia had actually begun years earlier, with extensive formative research. In the early 2000s, Population Council researchers began conducting quantitative, population-based surveys among adolescent girls and boys in the slum area of Merkato in Addis Ababa and poor rural areas of Amhara Region, the second largest region in Ethiopia. We wanted to identify the populations of young people in Ethiopia who were most marginalized, understand more about their circumstances, and use data to suggest ways to reach and support them.

The Population Council and the Ethiopian Ministry of Youth and Sports (now the Ethiopia Ministry of Women, Children, and Youth Affairs) developed a simple marker for vulnerability: being between ages 10 and 14 and out of school. In both rural and urban areas at that time, girls in this age group were more likely than boys to be out of school. When determining which girls were the most likely to be out of school, we identified two groups: married adolescents in rural areas and migrants who moved from rural to urban areas, many of whom were doing domestic work. Further analysis revealed that both of these groups of girls were socially isolated, with a large proportion reporting that they had no friends and did not live with either parent. As a result, the Council and the Ministry of Youth and Sports started to develop programs for married adolescent girls and rural-to-urban migrants/child domestic workers.

The Population Council and its Ethiopian partners devised two programs—BIRUH TESFA (Bright Future) for out-of-school urban girls, and MESERET HIWOTT (Base of Life) for married rural girls—that responded directly to the vulnerabilities identified through this research, including heightened risk of HIV and sexual and gender-based violence. The programs were designed to give girls the assets they need to avoid these circumstances and reach their full potential. Later, in response to community demand, a third program—ADDIS BIRHAN (New Light)—was added to involve husbands and seek to improve their attitudes toward gender roles.

Biruh Tesfa and Meseret Hiwott take place in safe spaces, locations where girls are respected and where they can develop life and livelihood skills, friendships, and peer support networks and can receive support and education from trusted female mentors (Brady, Saloucou, and Chong 2007). These safe spaces also offer value to the communities in which they are established because they serve as a platform for connecting girls to services and providing them with additional resources such as health services, support for the disabled, and support for victims of violence.

Research suggests that providing girls with programs through safe spaces may have far-reaching effects. For example, in Uganda the Empowerment and Livelihoods for Adolescents program provided girls aged 14–20 with life skills, financial literacy, and vocational training through girls’ groups led by female mentors. The program was evaluated with a randomized controlled trial and was found to increase employment and condom use and to reduce fertility and reports of nonconsensual sex (Bandiera et al. 2012).

A study in South Africa found that girls who belonged to sports clubs were less likely to test positive for HIV than other girls and more likely to negotiate condom use with their partners (Campbell, Williams, and Gilgen 2002).

In rural Ethiopia, the TESFA program (Towards Economic and Sexual Reproductive Health Outcomes for Adolescent Girls) found that when married girls were provided with economic empowerment and sexual and reproductive health information, their savings and use of family planning increased significantly (Edmeades, Hayes, and Gaynair 2014).

A study in South Africa showed that girls participating in a community group were significantly less likely to be sexually experienced and more likely to have used a condom at last sex (Hallman 2011).

From the outset of program design, Biruh Tesfa, Meseret Hiwott, and Addis Birhan were intended to be replicable. The costs were considered at the design stage, and calculations of cost per beneficiary were estimated. One strategy to increase the likelihood of sustaining a program includes using local pay scales, such as government pay scales. Because government pay scales reflect the economic circumstances of the country, using them maximizes the chances for sustaining a program.

Additionally, we increased sustainability by using existing infrastructure and resources, including human resources (e.g., locally recruited women leaders). By working with government ministries and local government administrations, or kebeles, from the earliest stages, we encouraged local ownership and involvement (“buy-in”) and maximized the chances for application of the lessons learned, especially in the public sector, which has the broadest reach.

The remainder of this report describes the environment of the HIV epidemic in Ethiopia, explains the rationale for providing adolescent girls with support and services that build protective assets, and outlines the Population Council’s distinct approach of using safe spaces as a platform for reaching girls and providing them with essential protective assets. The report details the achievements of each of the three programs and summarizes key lessons learned that may be useful in launching and expanding similar programs elsewhere.

Country Context and Program Achievements

This section describes the environment in which the project for vulnerable adolescent girls in Ethiopia takes place. It also details the project's three programs: Biruh Tesfa, Meseret Hiwott, and Addis Birhan.

The female face of HIV in Ethiopia

In Ethiopia and elsewhere in sub-Saharan Africa, the HIV epidemic disproportionately affects females; the female-to-male ratio of HIV infection in Ethiopia is 3 to 2 (Central Statistical Agency [Ethiopia] and ICF International 2012; Weiss and Rao Gupta 1998; de Bruyn 1992; Erulkar, Gebru, and Mekonen 2011). Biologically, male-to-female HIV transmission is more common than female-to-male transmission. But this does not fully explain the uneven infection rate. Girls and women also face a web of inter-related socio-cultural factors—particularly economic insecurity and social isolation—that raise their HIV risk. The experience of sexual and gender-based violence is also correlated with a woman’s risk of HIV infection (Li et al. 2014).

A Population Council study in low-income areas of Addis Ababa found that 77 percent of working girls were domestic servants, work that is characterized by poor conditions, low pay, and extreme social isolation—all factors that can contribute to HIV risk (Erulkar, Mekbib, Simie, and Gulema 2006a; Erulkar and Mekbib 2007). Domestic workers participating in the research in Addis Ababa reported an average monthly salary of about US$6 (Erulkar and Mekbib 2007). Girls and women who are economically dependent on men may find it harder to negotiate for condom use within regular partnerships, even if they suspect their partner is HIV infected. Adolescent girls and young women who are struggling to make ends meet may also turn to transactional sex, that is, accepting cash or gifts in exchange for sex. In such circumstances, the power imbalance between males and females may make it more difficult for females to negotiate safe sex. The Population Council found that 44 percent of commercial sex workers in Addis Ababa were formerly domestic workers, and that 39 percent of commercial sex workers initiated sex work after escaping other abusive work, such as domestic work (Girma and Erulkar 2009).

In addition, domestic workers participating in the research reported working an average of 64 hours in the week prior to the survey (Erulkar and Mekbib 2007). Domestic workers’ long working hours prevent many from accessing mainstream HIV prevention, treatment, care, and support programs for young people (Erulkar, Mekbib, Simie, and Gulema 2006b; Mekbib, Erulkar, and Belete 2005). Most domestic workers in the urban slums of Addis Ababa report having little education and scant knowledge of HIV, which also limits their ability to access HIV information and services (Erulkar and Mekbib 2007).

Further, domestic workers’ employers provide their home, their source of food, and all their security, a circumstance ripe for abuse. Council research among adolescent girls in urban slums of Ethiopia and research by others have found that domestic workers are more likely than other girls to be victims of sexual abuse, including nonconsensual sex (Erulkar and Ferede 2009; HRW 2006). Council research in low-income areas of three Ethiopian cities found that socially isolated girls were nearly twice as likely as girls with friends, social networks, or group affiliations to experience nonconsensual first sex and very early sexual initiation (Erulkar and Ferede 2009). 

Reducing the HIV vulnerabilities of adolescent girls

Few programs in Africa have sought to take a “whole girl” approach to addressing the multiple vulnerabilities to HIV infection—social isolation, economic insecurity, lack of access to services, and sexual and gender-based violence—experienced by the most marginalized adolescent girls in the poorest communities (Erulkar, Ferede, Girma, and Ambelu 2013). Further, reproductive health programs for adolescents in the developing world often ignore vulnerable married girls by focusing largely on reaching girls who are unmarried and sexually active. 


Biruh Tesfa was established in 2006 to serve out-of-school girls between ages 7 and 24 in urban slums, many of whom are migrants living apart from their families (Erulkar, Mekbib, Simie, and Gulema 2006a). The program seeks to improve girls’ ability to protect themselves by reducing their social isolation and providing them with social safety nets through mentors, peer groups, civic engagement, health information, and services to reduce sexual exploitation and abuse and to provide functional literacy skills. 

Biruh Tesfa draws on and seeks to maximize local human and community resources. Mentors are recruited from host communities. Regional bureaus and local kebele (lowest administrative level in Ethiopia) officials oversee implementation of the program, thereby heightening the visibility of these vulnerable groups in the eyes of local government and communities. 

Design and Implementation

Biruh Tesfa was originally conceived as a two-year pilot program designed for scale-up. Female mentors provide education and encouragement to girls and social support to deal with situations where sexual coercion, unfair wages, and other discriminatory or exploitative circumstances often prevail. 

Girls gather in groups according to age and educational level: those who have never been to school and those with some reading and writing ability. The group setting enables girls to develop friendship networks and cultivate relationships with caring and trustworthy adult mentors. 

Meet the Girls of Biruh Tesfa

Mentors identify out-of-school girls who are eligible for the program. Using maps from the Central Statistical Agency, a program area is divided into smaller geographical units, and each mentor is assigned a base of operation—preferably an area in which she lives. Mentors go house-to-house, making up to three visits to each house, and conduct an informal listing of household members, including their sex, age, and schooling status. The house-to-house visits allow mentors to contact girls who may otherwise be missed, such as child domestic workers[1] who are largely confined to the home (Erulkar, Mekbib, Simie, and Gulema 2006a; Erulkar, Semunegus, and Mekonnen 2011; Erulkar, Ferede, Girma, and Ambelu 2013), girls with disabilities, and child sex workers. 

[1] Households in many poor countries rely on girls and women for labor-intensive domestic tasks, such as preparing food and washing clothes. Even the poorest households can have a domestic worker who is paid very little or not at all. At times, young girls who are distant relatives may be brought from rural areas for this purpose. 


When mentors identify eligible girls in the household, they request their participation in the program. In addition, mentors discuss participation of interested girls with their parents or guardians, or—in the case of domestic workers—with their employers, who often do not initially support their participation. Mentors are responsible for following up with the girls they enroll in Biruh Tesfa. If a girl is absent for three or more group meetings, the mentor visits the girl’s home to determine why she failed to attend. In some cases, the girl’s employer has forbidden her to participate, in which case the mentor attempts to renegotiate on behalf of the girl.

Once enrolled in mentor-led groups, girls meet as often as five times a week. Group members receive notebooks and pens, identity cards, and locally made, reusable menstrual pads for older girls. The identity cards provide vulnerable girls who are far from home with an ID that gives them some degree of protection in the dangerous urban environments where they live, as well as some level of identity and personal affiliation. For many Biruh Tesfa participants, the ID is the only photograph they have of themselves. 

Because many participants (particularly those from rural areas) lack exposure to basic health services, members receive wellness checkups and are provided with vouchers for basic medical services, including screening and treatment for HIV and other sexually transmitted infections, through a network of public and private providers. Participants who need further diagnostic and curative services are referred to participating providers, using the voucher to subsidize the cost of services.  When mentors provide vouchers, they also offer to accompany girls and young women, particularly those who may not feel comfortable  going to a clinic for the first time.

Program services such as these ensure that girls have access to health care, that rape victims can obtain support services, and that a shelter is available for victims of sexual abuse.

Well-designed initiatives based on formative research can reach and support the most disadvantaged slum-dwelling girls.

Program Reach

Benefiting from local ownership in Biruh Tesfa, the program has grown significantly, reaching more than 63,000 girls in 18 cities by 2013. The profile of participants reveals their extreme vulnerability. Nearly half (44 percent) have lost one or both parents, over half (52 percent) have no education, and 82 percent have fewer than five years of schooling. One-third are domestic workers, one-quarter are daily laborers, and two-thirds are migrants.

Program Outcomes

According to the final evaluation using weighted logistic regressions, girls in the program sites were more than twice as likely to report having social support as girls in the control site. They were also twice as likely to score highly on HIV knowledge questions, to know where to obtain voluntary HIV counseling and testing, and to want to be tested compared to girls in the control site (Erulkar, Ferede, Girma, and Ambelu 2013). Because only a minority of girls (31 percent) were sexually experienced, the evaluation did not assess changes in sexual behavior.

An evaluation of Biruh Tesfa between September 2013 and February 2014 examined whether the program could improve girls’ literacy and numeracy. Among the girls in the experimental area who had never been to school, the mean score on the reading test increased significantly, from 0.46 to 0.66 between baseline and endline (p<0.05). The mean score on the numeracy test also increased significantly, from 2.6 to 3.5 between baseline and endline (p<0.001). Similar improvements in test scores were not seen among girls in the control area who had never been to school (Erulkar and Medhin 2014). 

Biruh Tesfa successfully adapted the rigorous house-to-house recruitment method tradition­ally used by community-based distribution agents or health extension workers. This resulted in one-third of eligible girls—normally a population that is very hard to reach—attending Biruh Tesfa. Without this effort, the program would not have achieved the coverage and scale that it did, given that the target group was generally housebound and carrying a significant household burden. 

Program Modification and Expansion

Biruh Tesfa was initially pilot-tested in the low-income Merkato area of Addis Ababa, the location of the main bus terminal and the arrival point for many migrants. The program was ultimately expanded to urban areas of the Amhara and Tigray regions, as well as other low-income areas of Addis Ababa, especially towns along the major transport routes that were initial destinations for migrant girls. Biruh Tesfa costs about US$52 per beneficiary per year to implement (Sewall-Menon et al. 2012).[2]

[2] This figure reflects costs to implement at the field level, including setting up staff with one field coordinator and payment for mentors. Additional costs include subsidies for health services, learning supplies (notebooks, pens, and pencils), t-shirts, and feminine hygiene products. The figure does not include evaluation costs (Sewall-Menon et al. 2012).

Qualitative information gathered during program monitoring helped to shape the services offered by Biruh Tesfa. Building on the original mentorship program, additional components such as the link with local health clinics were added as the needs of participants were identified. For example, one participant confided to her mentor that she had been raped. The mentor asked whether she had gone to the clinic. However, as a young girl from a rural area she had no experience of clinics, and told her mentor that she did not know how to go about it. 

Biruh Tesfa initiated onsite wellness checkups with local nurses to give girls the experience of having a physical exam. Through partnerships with local clinics, girls gained access to free services. Girls can request the accompaniment of a mentor if they are concerned about going alone for the service. Service statistics from the voucher program showed that 89 percent of girls requested accompaniment. The voucher system and mentor accompaniment were instrumental in introducing girls to the formal health system. For 70 percent of voucher users, it was their first time attending a clinic (Erulkar and Medhin 2014). All participating service providers were trained using the physicians’ “National Guideline for Survivors of Sexual Assault,” ensuring that rape victims received adequate care.

After learning through program monitoring that girls lack access to feminine hygiene products, the Council collaborated with a local female entrepreneur who makes washable, reusable menstrual pads. Biruh Tesfa now includes discussions on menstruation in the group meetings, a subject with which many participants were unfamiliar. 

Biruh Tesfa has received acclaim at the international, national, and local levels. It has been designated as a best practice on integrating gender and HIV by the US President’s Emergency Plan for AIDS Relief (PEPFAR) and the USAID-funded AIDS Support and Technical Assistance Resources (AIDSTAR-One) (USAID 2009). 

As part of the Population Council’s broader work with vulnerable urban and rural girls in Ethiopia, Biruh Tesfa has received numerous commendations from the Ethiopian government, including the Federal Ministry of Youth and Sports (currently Women, Children, and Youth Affairs) and the Amhara Regional Youth and Sport Bureau (currently, Women, Children, and Youth Affairs).


Meseret Hiwott was established in 2008 to support married girls aged 10–24 in rural Amhara, the second largest region in Ethiopia and the region with the highest rate of early marriage. Girls married early are at a distinct disadvantage (Erulkar 2013a; Erulkar, Mekbib, Amdemikael, and Conille 2009). Early marriage in Ethiopia is centered in the poorest rural communities, which are dominated by conservative gender norms (Erulkar 2013b). In a recent study in Ethiopia, among girls married before age 15, 79 percent had never been to school, 89 percent experienced arranged marriage, and 71 percent first met their husbands on their wedding day (Erulkar 2013b). 

Meseret Hiwott provides girls with wider social networks, as well as the assets to avoid HIV infection, improve their reproductive health, and better negotiate the power dynamics that often relegate them to inferior status within their household. The program also increases access to income and productive resources by offering training in financial literacy.

In late 2008, men in Meseret Hiwott project communities expressed interest in a male-only program of a similar nature, and Addis Birhan was born. Addis Birhan is a male-led HIV prevention program focusing on reducing behaviors that raise the risk of marital transmission of HIV and on promoting husbands’ healthy support of the family. Unlike Meseret Hiwott, husbands of any age are eligible for Addis Birhan. Participating husbands do not have to be married to girls of adolescent age or to girls who participate in Meseret Hiwott. The program was designed to contribute to achieving the objectives of Meseret Hiwott: increasing gender equity, encouraging supportive and understanding husbands, and promoting positive reproductive health behavior. In Ethiopia-based research, men with gender-equitable attitudes were more likely to report healthy behaviors with intimate partners, such as discussing and using condoms and other contraceptives (Pulerwitz et al. 2010).

The two programs were implemented in 20 districts in Amhara.  By mid-2013, over 225,000 married girls aged 10–24 had taken part in Meseret Hiwott, and over 130,000 married boys and men aged 10–85 had participated in Addis Birhan (Erulkar and Tamrat 2014).

Meet the Men of Addis Birhan

Design and Implementation 

In Meseret Hiwott, trained adult female mentors go house-to-house to identify and recruit married girls. In Addis Birhan, trained adult male mentors go house-to-house to identify and recruit married men. Married girls and married men are allowed to enter the programs regardless of whether their spouse participates. Participants meet in sex-segregated groups in locally available spaces, such as community halls, under trees, or in participants’ houses. 

Girls in Meseret Hiwott are offered a 32-hour curriculum that covers topics such as self-esteem; sexually transmitted infections, including HIV; voluntary HIV counseling and testing (VCT); anti-retroviral therapy (ART); reproductive health; menstruation management; family planning; safe motherhood; gender and power dynamics; and financial literacy. Additional topics of specific interest to married girls include spousal communication, gender roles that affect distribution of labor in the household and childrearing, and the risks of early marriage. Participants are referred to government health extension workers for medical services, such as HIV-related services and family planning.

Men in Addis Birhan are offered a 30-hour curriculum that was adapted—with significant changes to ensure that it was locally appropriate—from an existing curriculum developed by the Population Council, Promundo, and EngenderHealth. It includes  topics to help men become more supportive of their wives and family. The program promotes non-judgmental discussion, self-exploration, and self-expression. Interactive group sessions include role-playing and storytelling to spark discussion on topics such as assistance with domestic duties and childcare; couples attending health clinics together; and prevention of domestic violence. Because a large proportion of participants have never been to school and cannot read, pictures and other simple graphics are used as teaching aids. 


The two programs were evaluated together using data gathered from married adolescent girls to examine changes associated with programs for married girls and their husbands. The evaluation focused on outcomes related to husbands’ support and assistance with domestic duties, positive health behavior including use of family planning and VCT, and domestic and sexual violence. Researchers used a post-test cross-sectional research design, which entailed a population-based survey of 1,010 married girls aged 12–24 in communities where the pilot programs were implemented (Erulkar and Tamrat 2014). 

Girls who participated in Meseret Hiwott were more likely than other girls to use family planning, obtain voluntary HIV counseling and testing, and negotiate spousal accompaniment to clinic visits. For example, after controlling for age, schooling, and age at marriage, participating girls were nearly 8 times more likely to report having received VCT than nonparticipants, and if their husbands also participated they were over 18 times more likely to report having undergone VCT (Erulkar and Tamrat 2014). 

Participants in the married girls’ groups were over 2.5 times more likely to report receiving assistance with domestic work from their husbands than those who did not participate. Wives from couples in which both the husband and wife participated in the groups were over 8 times more likely to report that their husbands provided them with domestic support than wives in couples in which the husband did not participate (Erulkar and Tamrat 2014). 

While both programs showed promising results, there are limitations in the study design. The post-test, cross-sectional evaluation was based solely on data collected following initiation of the programs. Because no baseline data were collected from the target group of married adolescent girls, it was difficult to identify changes associated with the programs. In addition, the design makes it difficult to determine the role of selectivity in the results. In other words, it is unclear to what extent the characteristics of people who join programs such as these may bias the results. For example, people who join community programs may be more highly motivated or more likely to believe in gender equity and may be more likely to use family planning or VCT, regardless of the nature of the program. The evaluation indicates that younger married girls were less likely to join the program, suggesting potential biases in enrollment. In addition, this analysis is limited to reports by the girls themselves, and not those of their husbands.

Nevertheless, the consistency of the results across various outcomes suggests that positive norms and behaviors are associated with program participation for rural, married adolescent girls and husbands.

Lessons Learned and Recommendations for Replicating the Programs

This section discusses lessons learned and best practices from the Biruh Tesfa, Meseret Hiwott, and Addis Birhan programs in Ethiopia. It also makes recommendations that will be useful in efforts to replicate the programs elsewhere.

Lessons from the “HIV Prevention for Vulnerable Adolescent Girls” project and recommendations

These three programs in Ethiopia have provided many key lessons about the best ways to reach and help girls, as well as ways to use evidence to improve programs. 

Biruh Tesfa, Meseret Hiwott, and Addis Birhan reach different participants in different geographical areas, but they share many similarities that can be adapted to the design, implementation, and evaluation of programs to benefit adolescent girls elsewhere.

Population Council evaluations indicate that these programs reduce social isolation and increase health knowledge, including knowledge about avoiding HIV infection. Preliminary evidence suggests that girls are building their social capital and protective assets in ways that reduce their HIV risk and increase their prospects for a safe and more productive life. Our results also demonstrate that when evidence-based insights about girls’ circumstances and needs and about obstacles to their health and wellbeing are used to inform program design, participants, gatekeepers, and the community respond positively and the programs can be scaled up successfully.

When developing programs to reduce gender inequalities and improve health, the Population Council’s experience in Ethiopia and elsewhere suggests the following six strategies:


Formative research, particularly population-based surveys, can identify the most vulnerable populations. In this case we identified rural married girls and urban out-of-school girls as being especially vulnerable to HIV infection. 

Data from this research were also used to characterize the circumstances and needs of these girls (i.e., housebound, low levels of education, few or no friends) as a guide in shaping program design. By reflecting the priorities and needs of target populations, researchers can increase the likelihood of program participation and effectiveness.


The mentorship model employed in these programs builds trust and inspiration among participants. The girls we sought to enroll often lacked caring adults in their lives. Mentors filled this role and also served as higher-status advocates for girls who otherwise lacked protective adults.


Programs can avoid unintentionally serving those who are relatively more elite or advantaged in the geographic setting by using strategies that have been proven to reach even the hardest-to-reach girls. Because girls are often confined to the home and tend to have far more limited mobility than their male counterparts (Hallman and Roca 2007), program mentors in Biruh Tesfa and Meseret Hiwott reach and enroll them by going house to house—a method traditionally used by community-based distribution agents or health extension workers. This strategy ensures that the hardest-to-reach girls are identified and invited to join the program. It also allows mentors to negotiate with gatekeepers, such as employers, husbands, and in-laws, to permit the girls’ participation. This method of recruitment not only results in reaching a higher proportion of a population in need of program services, but also allows mentors to observe and understand the home circumstances of the girls in their groups. This equips the mentors to better deal with future challenges that girls may face. 


Measuring important components of girls’ empowerment will help to build the evidence base—and the case—for investing in girls.

While it is essential to document program implementation and changes in attitudes, the most salient results deal with changes in the lives of the girls themselves. A clear indication that a program is girl-centered is that it has girl-level outcome measures. Research has identified several types of protective social capital (friendship networks, membership in groups that meet regularly) and safety nets (someone to turn to in an emergency or a shelter in which to spend the night)  (Erulkar, Mekbib, Simie, and Gulema 2004). These can be measured and documented.

Programs can measure how well prepared girls are to deal with the challenges they face by assessing their protective assets, such as having someone to turn to in a crisis and having specific safety plans. Some protective assets can be attained within relatively short periods of time. For example, girls can acquire personal documentation and a more explicit sense of risk in their environment, and they can formulate specific plans to avoid such risk (Bruce 2011).

It is also important to include a range of outcomes, rather than limit evaluations to a narrow range of measures. This is because, as girls (and the community as a whole) become sensitized, many circumstances that were previously stigmatized and considered taboo topics for conversation or reporting—such as gender-based violence—may become destigmatized and redefined as  reportable events. In girls’ programs that include discussion of gender-based violence, evaluations may actually detect increased reporting of such violence because of the increased acceptability of reporting.  


Before, during, and after program implementation, use quantitative and qualitative research to generate the evidence required to identify best practices, refine critical program elements, and eliminate ineffective approaches. 

Qualitative monitoring after programs are underway can provide important information for improving programs. Every six months, for example, Biruh Tesfa program managers interview between 10 and 20 girls, as well as 5 to 10 mentors. Managers use these interviews to explore new areas for expansion or program modification, and they modify the questions at each round to elicit different types of information.


Programs cannot succeed if they achieve their short-term goals but fail to turn to local agencies to implement them in the long term. By working with governments and local NGOs from the earliest stages, researchers can encourage local ownership and involvement (“buy-in”) and continuing resources to support initial efforts and their scale-up once proven successful.

Program sustainability is fostered by close collaboration with governmental and local nongovernmental partners in program planning, implementation, and evaluation. The sharing of knowledge and ownership is critical, since local partners will ultimately be responsible for conducting programs. In the instances of Biruh Tesfa, Meseret Hiwott, and Addis Birhan, each program collaborated with numerous governmental and nongovernmental partners. Community ownership was further enhanced by recruiting male and female leaders from within the community to serve as mentors for program participants, and by encouraging the participation of kebele administrations. 

Engaging large numbers of local residents creates lasting, normative changes within a community. As indicated by findings from the evaluation of Meseret Hiwott and Addis Birhan, this approach has the potential to increase men’s support for girls’ and women’s assets. 

Making programs attractive to intended recipients, a key factor in sustainability, can be ensured when programs are based on formative research. Such research can provide information on what categories of people are not being reached by other initiatives.

Collaborating organizations and local administrations that provide both monetary and “in kind” support for programs help to make them more cost-effective. For instance, local kebeles  played a critical role in contributing the safe spaces used for program activities.


Bandiera, O., N. Buehren, R. Burgess, M. Goldstein, S. Gulesci, I. Rasul, and M. Sualiman. 2012. Empowering Adolescent Girls: Evidence from a Randomized Control Trial in Uganda, enGender Impact: The World Bank’s Gender Impact Evaluation Database. Washington, DC: World Bank.

Brady, Martha, Lydia Saloucou, and Erica Chong. 2007. “Girls’ adolescence in Burkina Faso: A pivot point for social change.” Ouagadougou: Population Council.

Bruce, Judith. 2011. “Violence against Adolescent Girls: A Fundamental Challenge to Meaningful Equality,” GIRLS FIRST! Perspectives on Girl-Centered Programming. New York: Population Council.

Campbell, Cathy, Brian Williams, and Denise Gilgen. 2002. “Is social capital a useful conceptual tool for exploring community level influences on HIV infection? An exploratory case study from South Africa,” AIDS Care 14(1): 41–54.

Central Statistical Agency [Ethiopia] (CSA) and ICF International. 2012. Ethiopia Demographic and Health Survey 2011. Addis Ababa, Ethiopia and Calverton, Maryland, USA: CSA and ICF International.

de Bruyn, M. 1992. “Women and AIDS in developing countries,” Social Science & Medicine 34: 249–262.

Edmeades, J., R. Hayes, and G. Gaynair. 2014. “Improving the lives of married adolescent girls in Amhara, Ethiopia.” Washington, DC: International Center for Research on Women. 

Erulkar, Annabel S. 2013a. “Adolescence lost: The realities of child marriage,” Journal of Adolescent Health 52(5): 513–514.

———. 2013b. “Early marriage, marital relations and intimate partner violence in Ethiopia, International Perspectives on Sexual and Reproductive Health 39(1): 6–13.

Erulkar, Annabel and Abebaw Ferede. 2009. “Social exclusion and early or unwanted sexual initiation among poor urban females in Ethiopia,” International Perspectives on Sexual and Reproductive Health 35(4): 186–193.

Erulkar, Annabel, Abebaw Ferede, Woldemariam Girma, and Worku Ambelu. 2013. “Evaluation of ‘Biruh Tesfa’ (Bright Future) program for vulnerable girls in Ethiopia,” Vulnerable Children and Youth Studies 8(2): 182–192.

Erulkar, Annabel, Herma Gebru, and Gebeyehu Mekonnen. 2011. “Biruh Tesfa (‘Bright Future’) program provides domestic workers, orphans and migrants in urban Ethiopia with social support, HIV education and skills.” Program brief. Addis Ababa, Ethiopia: Population Council.

Erulkar, Annabel and Girmay Medhin. 2014. “Educational and health service utilization impacts of a girls’ safe spaces program in Addis Ababa, Ethiopia.” Report. Addis Ababa, Ethiopia: Population Council.

Erulkar, Annabel and Tekle-Ab Mekbib. 2007. “Invisible and vulnerable: Adolescent domestic workers in Addis Ababa, Ethiopia,” Vulnerable Children and Youth Studies 2(3): 246–256.

Erulkar, Annabel S, Tekle-Ab Mekbib, Helen Amdemikael, and Garry Conille. 2009. “Marriage and sexual experience of adolescent girls and women in West Gojam Zone, Amhara region, Ethiopia,” Ethiopian Journal of Reproductive Health 3(3): 27–33.

Erulkar, Annabel S., Tekle-Ab Mekbib, Negussie Simie, and Tsehai Gulema. 2004. “Adolescent life in low income and slum areas of Addis Ababa, Ethiopia.” New York: Population Council.  

———. 2006a. “Migration and vulnerability among adolescents in slum areas of Addis Ababa, Ethiopia,” Journal of Youth Studies 9(3): 361–74.

———. 2006b. “Differential use of adolescent reproductive health programs in Addis Ababa, Ethiopia,” Journal of Adolescent Health 38: 253–260. 

Erulkar, Annabel, Belaynesh Semunegus, and Gebeyehu Mekonnen. 2011. “Biruh Tesfa provides domestic workers, orphans, and migrants in urban Ethiopia with social support, HIV education, and skills.” Transitions to Adulthood brief no. 21. New York: Population Council.

Erulkar, Annabel and Tigest Tamrat. 2014. “Evaluation of a reproductive health program to support married adolescent girls in rural Ethiopia,” African Journal of Reproductive Health 18(2): 68–76.

Girma, Woldemariam and Annabel Erulkar. 2009. “Commercial sex workers in five Ethiopian cities: A baseline survey for USAID targeted HIV prevention program for most-at-risk populations.” Addis Ababa: Population Council. 

Hallman, Kelly. 2011. “Adolescent social exclusion: The gendering of HIV risk in KwaZulu-Natal, South Africa.” In J. Klot and V. Nguyen (eds.), The Fourth Wave: An Assault on Women–Gender, Culture and HIV in the 21st Century, Paris: Social Science Research Council and UNESCO, pp. 53–79.

Hallman, Kelly and Eva Roca. 2007. “Reducing the social exclusion of girls,” Transitions to Adulthood brief no. 27. New York: Population Council.  

Human Rights Watch (HRW). 2006. “Swept under the rug: Abuses against domestic workers around the world.” Human Rights Watch 18.

Li, Ying, Caitlin M. Marshall, Hilary C. Rees, Annabelle Nunez, Echezona Ezeanolue, and John Ehiri. 2014. “Intimate partner violence and HIV infection among women: A systematic review and meta-analysis,” Journal of the International AIDS Society doi: 10.7448/IAS.17.1.18845.

Mekbib, Tekle-Ab, Annabel S. Erulkar, and Fekerte Belete. 2005. “Who are the targets of youth programs: Results of a capacity-building exercise in Ethiopia,” Ethiopian Journal of Health Development 19(1): 60–62.

Ministry of Health (MOH), Disease Prevention and Control Department. 2000. AIDS in Ethiopia (3rd Ed.). 

Ministry of Health (MOH) and Federal HAPCO. 2007. “Single Point HIV Prevalence Estimate,” Addis Ababa, Ethiopia: June. 

National AIDS Resource Center (NARC). 2012. “HIV/AIDS Estimates and Projections in Ethiopia, 2011-2016.” Accessed 26 August 2014. <>.

Pulerwitz, Julie, Stephanie Martin, Manisha Mehta, et al. 2010. “Promoting Gender Equity for HIV and Violence Prevention: Results from the Male Norms Initiative Evaluation in Ethiopia,” Washington, DC: PATH.

Sewall-Menon, Jessica, Judith Bruce, Karen Austrian, et al. 2012. “The cost of reaching the most disadvantaged girls: Programmatic evidence from Egypt, Ethiopia, Guatemala, Kenya, South Africa, and Uganda.” New York: Population Council.

UNAIDS. 2012. “UNAIDS: World AIDS Day Report 2012,” Geneva: UNAIDS.

———. 2013. “2013 Progress Report on the Global Plan,” Geneva: UNAIDS.

USAID. 2009. AIDSTAR-One Technical Compendium. “Biruh Tesfa (Bright Future),” Excerpt from Integrating Multiple Gender Strategies to Improve HIV and AIDS Interventions: A Compendium of Programs in Africa, Arlington, VA: AIDSTAR-One, John Snow, Inc. (May). 

Weiss, E. and G. Rao Gupta. 1998. Bridging the Gap: Addressing Gender and Sexuality in HIV Prevention. Washington, DC: International Center for Research on Women.

PDF Download