89997 Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations Jennifer Parsons and Jennifer McCleary-Sills Gender Group, World Bank Group Synopsis Gender equality cannot be achieved until women have control over their sexual and reproductive health (SRH) decisions, including those that determine their risk for HIV and other sexually transmitted infections (STIs) and for unintended, inadequately spaced, and early pregnancies. Drawing on lessons from seven impact evaluations (IEs) supported by the World Bank Group (WBG), this brief highlights lessons learned about strengthening SRH interventions to support women’s achievement of their own reproductive intentions. These IEs indicate that implementing comprehensive, interactive interventions in schools and community settings, increasing access to education for girls, and promoting girls’ empowerment and agency offer promise in improving SRH outcomes, particularly among adolescents. However, more rigorous and long-term impact evaluations are needed to better understand how to effectuate sustainable impacts on these outcomes, especially adolescent childbearing. Background Control over sexual and reproductive health is a fundamental element of gender equality and is critical to achieving an array of other important development outcomes (World Bank Group, 2014). Agency over SRH denotes the ability to choose whether, when, and with whom to have sex, to ask a partner to use a condom, and to make decisions about childbearing and one’s own health (WBG, 2014). Yet around the globe, millions of women and girls are unable to exercise agency over these decisions, as evidenced by 80 million unintended pregnancies each year and high One in five girls in developing countries levels of unmet need for family planning in becomes pregnant before the age of 18. The developing countries (Singh and Darroch, lifetime opportunity costs of teen pregnancy 2012). have been estimated to range from 1 percent of annual gross domestic product in China to Early sexual initiation puts girls at greater risk as much as 30 percent in Uganda, measured for HIV and other STIs, early pregnancy, and solely by lost income (Chaaban and early childbearing (Hindin and Fatusi, 2009). Cunningham, 2011). In developing countries, Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 1 pregnancy-related causes are the largest outcomes is clear, especially now that young contributor to the mortality of girls ages 15 to people 10 to 24 comprise nearly a third of the 19—nearly 70,000 deaths annually (WBG, global population. But we are just beginning 2014; UNFPA, 2013). In Sub-Saharan Africa, to understand what works to expand the voice HIV prevalence among young people and agency of women and girls in this domain, decreased by almost half between 2001 and particularly in developing countries. 2012, but prevalence among young women remains more than double that among young Evidence from the World Bank Group men (UNAIDS, 2013). In recent years, the World Bank Group has increased investments in gender-informed There is growing evidence that improving programs, research, and impact evaluation, these SRH outcomes is directly linked to the including through numerous programs to economic status of women and their families. improve SRH outcomes and IEs to assess Women’s increased use of contraception has the success of these interventions. This brief resulted in lower fertility rates in many synthesizes learning from IEs on SRH countries. Having fewer children can improve outcomes to identify effective programming the overall well-being of women and girls, components that improve these outcomes. resulting in improved health of mothers and increased female participation in the workforce Evaluated programs target specific populations (ICRW, 2013). Lower fertility increases life and can measure a variety of outcomes. expectancy and allows women to spend less Among WBG gender IEs on sexual and time on childbearing and childrearing and to reproductive health, the outcomes most instead pursue economic opportunities outside commonly evaluated are: teen pregnancy and the home. childbearing, sexual risk behaviors including condom use, and knowledge about safe sex practices. Drawing from the 161 IE papers within the enGENDER IMPACT (eGI) database, we identified those that measured programmatic impact on these outcomes. Thirteen IEs included SRH as a component of their interventions. Of those programs, seven measured the outcomes of interest and are discussed in further detail below. (See Table 1 for a summary of included IEs). In addition, two of these interventions measured rates of early marriage among program participants, which is an important outcome given the Smaller family size also improves children’s connection between early marriage, girls’ health and leads to higher educational sexual initiation, and teen pregnancy and attainment (WBG, 2014; UNFPA, 2012). In childbearing. However, child marriage is not addition, recent research suggests that when discussed in detail here, as it is the subject of family sizes are smaller, the economic value of another brief in this series, Preventing Child human capital increases and household Marriage: Lessons from World Bank Group decision making power among men and Gender Impact Evaluations. women becomes more equitable (WBG, 2014; Doepke, 2009). As a development objective, the need to improve sexual and reproductive health Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 2 Summary of SRH Outcomes Measured beyond a woman’s control (Jewkes and by WBG Gender IEs Morrell, 2010). Impact evaluations of SRH programs have Teen Pregnancy and Childbearing many ways of assessing and measuring sexual Teen pregnancy and childbearing are risk behaviors. These can vary a great deal important risk factors for maternal mortality, depending on the population targeted by the increased school dropout rates, and negative intervention, the desired behavior change, and long-term economic earning potential. the actual components of the intervention Lowering rates of fertility and delaying carried out in practice. Four of the seven pregnancy can increase a woman’s life included IEs measured the number of lifetime expectancy and enable her to pursue partners (Baird et al., 2010; Duflo et al., 2006; economic opportunities that improve her health Dupas, 2011; Packel et al., 2012), three and well-being, as well as that of her children measured onset of sexual activity (Baird et al., and family (WBG, 2014). 2010; Duflo et al., 2006; Dupas, 2011), and one IE measured abstaining from sexual Five out of the seven IEs assessed teen activity as well as choosing “less risky” sexual childbearing, measured as live births by girls partners (Packel et al., 2012) as outcomes before the age of 18 (Baez et al., 2011; Dupas, reflecting sexual risk behaviors. Five measured 2011; Bandiera et al., 2012; Duflo et al., 2006; condom use as an outcome of the program. Heath and Mobarak, 2012). This measure does not necessarily capture all pregnancies One intervention included in the seven IEs in among respondents. A sixth IE measured teen this analysis measured the impact of a pregnancy, based on self-reported pregnancy conditional cash transfer (CCT) behavior before age 18, without regard for the outcome change program in Tanzania aimed at of the pregnancy (Baird et al., 2010). For the reducing sexual risk behaviors and decreasing purpose of this brief, we include them as one the incidence of HIV and other STIs (Packel et category “teen pregnancy and childbearing”, al., 2012). In addition to measuring the number which was captured by six out of the seven of partners, risk behavior was also assessed IEs. Measures included questions asking about through reports of abstinence and engaging in pregnancy incidence, and whether or not girls sexual activity with “risky” partners, defined as in the interventions had “begun childbearing” having sex without a condom with a non- since the onset of the program. marital partner in the last four months (Packel et al., 2012). One final risk behavior of interest Sexual Risk Behaviors in these IEs was forced sex. Given the A number of sexual behaviors can lead to increased chance of acquiring HIV or another unwanted pregnancy and increase the risk of STI through forced sex, this outcome illustrates acquiring HIV and other STIs. These include a critical risk to women’s and girls’ agency over early onset of sexual activity, nonuse or their bodies. misuse of contraceptives, having multiple sexual partners, and drug and alcohol use Condom Use (Guttmacher, 2003). Decreasing risk behaviors Condom use is the most widely measured risk can lead to fewer unintended pregnancies, behavior among these IEs because of its link decreased incidence of HIV and other STIs, to unintended pregnancy and HIV risk, as well and contribute to greater economic opportunity as to safe-sex negotiation and contraceptive and workforce participation, particularly for decision-making power. We therefore consider women and girls (ICRW, 2013). It is important this outcome separately from other sexual risk to note that where traditional gender norms behaviors in the discussion that follows. prevail and power within sexual relationships is held by men, reducing these risks is often Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 3 Across the five interventions, condom use was spaces, skills training, and sexual education for assessed by asking a variety of questions to girls with peers and mentors. participants, including whether they were sexually active, whether they used a condom So What Works? Implications for Future at last sex, and how frequently they used Programming condoms (ever, always, often, sometimes, or never). Multiple approaches and strategies were used in each of the seven programs evaluated within Knowledge the scope of this analysis. The following In many cultural contexts, gendered social highlights programmatic components of norms that deem it inappropriate for girls to interventions that impacted the SRH outcomes learn about sexuality limit their ability to learn of interest. about SRH in school as well as from family and community members (WBG, 2014). In addition, girls and women can face stigma around being sexually active—a stigma not often experienced by boys and men (Fairhurst et al., 2004; WBG, 2014). These norms prevent girls from learning and making informed SRH decisions and can give boys and men the more prominent role in decision making around sex and contraceptive use (WBG, 2014; McCleary- Sills et al., 2012). For these reasons, increasing access to information about SRH is a key factor in allowing all individuals to make better decisions around sexual practices and avoid risky behavior. Knowledge is power, and Teen Pregnancy and Childbearing the power to make better decisions will lead to Across all six interventions, there was a improved sexual and reproductive health significant effect on teen pregnancy and outcomes. childbearing when comparing the intervention and control groups, though with some In the seven IEs discussed here, two differences in effect size and significance for specifically measured knowledge around HIV subgroups. For example, the Zomba Cash and pregnancy (Duflo et al., 2006; Bandiera et Transfer Program in Malawi showed that girls al., 2012). In one school-based intervention in who were previously not enrolled in school Kenya, knowledge was measured as were less likely to experience early pregnancy identifying different ways to protect oneself after participating in the program compared to from HIV, and participants were asked their controls. However, no change in pregnancy opinions about the efficacy of condom use to outcomes was seen among girls who were prevent pregnancy and HIV transmission already enrolled in school when the (Bandiera et al., 2012). intervention began (Baird et al., 2010). Programs in Kenya and Uganda, also targeting In a second intervention measuring knowledge adolescent girls, found statistically significant outcomes (Duflo et al., 2006), HIV knowledge decreases in the incidence of childbearing in was measured using a six-question index groups exposed to the programs compared asking about HIV transmission and testing, with the control group (Duflo et al., 2006; while pregnancy knowledge was measured by Dupas, 2011; Bandiera et al., 2012). asking one true/false question of female participants: “A women cannot become Effective programs for decreasing pregnancy pregnant at first intercourse or with occasional and childbearing included the following sexual relations.” This program provided safe components: provision of conditional cash Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 4 transfers for schooling, reduction in the cost of at least one behavior as a result of the education, provision of vocational training, and intervention. These changes included having provision of information on health and risky abstained from sex, having had fewer sexual behaviors, including the relative risk of HIV partners, and having had less risky partners. infection by partner’s age. In addition, one IE Interestingly, significantly more men than measuring the effects of garment factory job women reported any change in behavior— opportunities on education and childbearing notably abstaining from sex and having less outcomes for girls (Heath and Mobarak, 2012) risky partners. found declines in births before age 16 and 18 among girls whose towns had a factory. Only one of these IEs (Bandiera et al., 2012) Programs promoting abstinence-only measured the outcome of forced sex, but the education for teens did not show any change in results were staggering. For girls who pregnancy or childbearing outcomes (Dupas, participated in this intervention in Uganda, the 2011). incidence of forced sex decreased by half. More detailed discussion of this analysis of In regard to onset of sexual activity, three IEs changes in sexual coercion can be found in saw statistically significant decreases in the another brief in this series: Gender-Based proportion of participants who had ever had Violence Prevention: Lessons from World sex compared with the control groups (Baird et Bank Impact Evaluations. al., 2010; Duflo et al., 2006; Packel et al., 2012). However, in one program targeting girls Condom Use who were in school as well as those who had Three out of the five IEs that measured dropped out prior to the intervention, the condom use showed significantly higher use reduction was statistically significant only among those exposed to the intervention among the girls who had dropped out prior compared with those who were not (Bandiera (Baird et al., 2010). A reduction in the number et al., 2012; Duflo et al., 2006; Packel et al., of partners reported by program participants 2012). These three programs include girls’ was seen in only one of these three IEs—the clubs in Uganda that promote vocational CCT program for school girls in Malawi (Baird training and health education simultaneously et al., 2010). However, once again reductions through mentoring, a school-based program in were statistically significant only among girls Kenya encouraging students to debate the role who had dropped out of school prior to the of condoms and to write essays on how to program. Evidence of greater programmatic protect themselves against HIV/AIDS, and an impact among school leavers suggests that the HIV and STI testing and prevention CCT CCT intervention is more protective for girls program in Tanzania for men and women ages with the highest risk profile. In one program, 18 to 30. Of the remaining IEs, the Zomba which provided risk information to teenagers in program in Malawi (Baird et al., 2010) saw no Kenya (Dupas, 2011), there was not a programmatic impact on self-reported condom decrease in the number of partners overall, but use, while the final IE measuring teenager’s there was a significant decline in the number of response to HIV risk information in Kenya cross-generational partners among teenage (Dupas, 2011) showed an increase in condom girl participants, and an increase in sexual use among girls exposed to the intervention, activity among teen boys, indicating a possible but that increase was not statistically shift from girls choosing older (i.e., riskier) significant. These results suggest that partners to choosing partners closer to their components of effective interventions aiming to own age. increase condom use incorporate interactive education and discussion sessions among Sexual Risk Behaviors peers and mentors, and provide cash incentive The evaluation by Packel et al. (2012) found to reduce sexual risk behaviors. that at the four-month follow-up, over half of study participants reported they had changed Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 5 From these five IEs, we know that key components of effective interventions for reducing sexual risk behaviors overall include those that: give girls access to free education, specifically allowing girls who were previously not in school to return to the classroom; provide safe spaces for skills training and sexual education for girls, including discussions around empowerment in relationships with men; and inform girls and boys about the increased risk of cross- generational sexual activity. For older men and women, there is some evidence that providing financial incentives to change behavior can reduce sexual risk behaviors, especially among men. Components of Effective Programming Programs that aim to improve SRH are Knowledge implemented both in school curricula and in The IEs assessing knowledge change found community-based interventions outside of the significant increases in knowledge among girls classroom. Ideally, SRH programs would be and boys in schools implementing teacher comprehensive in nature, aiming to reduce risk training with condom debates and essay behaviors, increase condom use and SRH writing on the topic. Increased knowledge knowledge, with the long-term outcome of about condom use as effective protection empowered decision making as evidenced by, against HIV and about correct condom use for example, delayed pregnancy, decreased preventing pregnancy was documented unintended pregnancy, and reduced among the intervention group. Girls in this adolescent childbearing. group also increased their knowledge of correct condom use preventing HIV Reducing the cost of school attendance for transmission. Among those exposed to just girls and ensuring access to education is a key the teacher training, the only significant component to increasing knowledge, reducing outcome around knowledge was among risk behaviors, and decreasing teen pregnancy boys—evidenced by an increase in the number and childbearing. Incorporating interactive who mentioned abstinence as a way to protect components of SRH education curricula in oneself from HIV. The intervention in Uganda schools in addition to teacher training and found statistically significant increases in both explanation of risk reduction also decreases HIV knowledge and pregnancy knowledge risk behaviors and increases condom use. among program participants (Bandiera, 2012). Given gender norms and stigma around sexual health, adolescent girls are a particularly The results of these two IEs measuring vulnerable group with respect to SRH knowledge outcomes suggest that successful outcomes. These IEs indicate that providing interventions to increase knowledge around safe spaces where girls can learn vocational HIV, pregnancy, and SRH as a whole combine training, receive SRH education from peers traditional sexual education curricula with and mentors, and increase their empowerment interactive and didactic learning methods in a and agency in relationships leads to improved classroom-based approach. In addition, SRH knowledge, reduced risk behavior, and programs directed at teen girls exclusively lower levels of teen pregnancy and work best when the interventions are combined childbearing. Providing cash incentives can to teach vocational skills and health education, influence sexual risk behaviors such as while increasing empowerment and agency. condom use, particularly among men. The greater observed increase in reported condom Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 6 use among men compared to women likely included in this analysis as each outcome was reflects underlying differences in the power measured by only a single program. and control men have in regard to decision making around sexual behaviors, in particular Conclusion condom use. Future programming should also It is evident that a multi-pronged approach to address decision-making power around SRH education and programming is necessary contraception and sexual risk and aim to to increase knowledge around HIV and increase women’s control over these critical pregnancy, reduce teen pregnancy and early SRH decisions. childbearing, and decrease sexual risk behavior not only among adolescents but also Limitations among adults. Based on the results of these While there appear to be a number of IEs, approaches that combine SRH education consistent findings in what works to improve in classrooms with peer engagement and SRH outcomes, some limitations to this discussion seem to be the most effective analysis exist. All of the programs within our approach to improving SRH outcomes in scope are based in in Africa (Malawi, adolescents. Keeping girls in school and Tanzania, Kenya, Uganda), which limits the specifically targeting girls for interventions are extent to which these findings may be key to ensuring better SRH outcomes. More generalized to other contexts. In addition, a rigorous evaluations of what works to promote majority of our programs address SRH among girls’ and women’s agency over their sexual adolescents, and the findings may not be and reproductive decisions are needed and applicable to adults and those in more stable should occur over longer periods of time to relationships. Other IEs identified by more accurately measure long-term effects, enGENDER IMPACT included measures of particularly for childbearing and pregnancy adherence to antiretroviral therapy, counseling, outcomes. testing, and care seeking for HIV, and STI prevalence. IEs of those programs were not enGENDER IMPACT enGENDER IMPACT is an online gateway for Gender-Related Impact Evaluations. At www.worldbank.org/engenderimpact you will find profiles summarizing key information about World Bank Group funded gender-related impact evaluations. These profiles are organized around priority areas for policy action, including: reducing health disparities, shrinking education and skills gaps, increasing economic opportunities, boosting voice and agency, and addressing gender-based violence. enGENDER IMPACT aims to share knowledge from previous evaluations and encourage more and better evaluations in key gender topics. Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 7 References Baez, J. E., Alam, A., and Del Carpio, X. V. (2011). Does Cash for School Influence Young Women’s Behavior in the Longer Term? Evidence from Pakistan. Baird, S., Chirwa, E., McIntosh, C., and Özler, B. (2010). The short‐term impacts of a schooling conditional cash transfer program on the sexual behavior of young women. Health economics, 19(S1), 55-68. Bandiera, O., Buehren, N., Burgess, R., Goldstein, M., Gulesci, S., Rasul, I., and Sulaiman, M. (2012). Empowering Adolescent Girls: Evidence from a Randomized Control Trial in Uganda. Working Paper. *Results reported here are based on an updated paper published in 2014: http://www.theigc.org/sites/default/files/Bandiera%20et%20al%202014.pdf Chaaban, J., and Cunningham, W. (2011). “Measuring the Economic Gain of Investing in Girls: The Girl Effect Dividend,” Policy Research Working Paper 5753, World Bank, Washington, DC, 2011. Doepke, M., and Tertlit, M. (2009). Women’s Liberation: What’s in It for Men? Quarterly Journal of Economics 11, (92). 379–420. Duflo, E., Dupas, P., Kremer, M., and Sinei, S. (2006). Education and HIV/AIDS prevention: evidence from a randomized evaluation in Western Kenya. World Bank Policy Research Working Paper. Dupas, P. (2011). Do Teenagers Respond to HIV Risk Information? Evidence from a Field Experiment. American Economic Journal: Applied Economics, 3(1) 1-34. Fairhurst, K., Ziebland, S., Wyke, S., Seaman, P., and Glasier, A. (2004). Emergency Contraception: Why can’t you give it away? Qualitative findings from an evaluation of advance provision of emergency contraception, Contraception 70, 25–29. Guttmacher Institute. (March/April 2003). Perspectives on Sexual and Reproductive Health. Vol 35, No 2. Heath, R., and Mobarak, A., (2012). Does Demand or Supply Constrain Investment in Education? Evidence from Garment Sector Jobs in Bangladesh. Hindin, M., and Fatusi, A. (2009). Adolescent Sexual and Reproductive Health in Developing Countries. International Perspective on Sexual and Reproductive Health, 35 (2), 58-62. Jewkes, R. and Morrell, R. (2010). Gender and sexuality: Emerging perspectives from the heterosexual epidemic in South Africa and implications for HIV risk and prevention. J Int AIDS Soc; 13(6). Joint United Nations Programme on HIV/AIDS (UNAIDS), Global Report (2013). UNAIDS Report on the Global AIDS Epidemic 2013. New York: UNAIDS. McCleary-Sills, J., McGonagle, A., and Malhotra, A. (2012). Women’s Demand for Reproductive Control: Understanding and Addressing Gender Barriers (Washington, DC: International Center for Research on Women. Packel, L., Dow, W., De Walque, D., Isdahl, Z., and Majura, A. (2012). Sexual behavior change intentions and actions in the context of a randomized trial of a conditional cash transfer for HIV prevention in Tanzania. World Bank Policy Research Working Paper, (5997). Stoebenau, K., Pande, R., and Malhotra, A. (2013). Has fertility decline contributed to improvements in women’s lives? International Center for Research on Women Fertility & Empowerment Working Paper Series. 012-2013-ICRW-FE. 1-40. Singh, S., and Darroch, J. (2012). Adding It Up: Costs and Benefits of Contraceptive Services, Estimates for 2012. New York: Guttmacher Institute and UNFPA. UNFPA, State of the World’s Population: Motherhood in Childhood: Facing the challenge of adolescent pregnancy, (New York: UNFPA, 2013). UNFPA, State of the World’s Population: By Choice, Not By Chance: Family Planning, Human Rights and Development, (New York: UNFPA, 2012). World Bank Group (2014). Voice and Agency: Empowering Women and Girls for Shared Prosperity. Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 8 Photo Credits: 1. Woman with her child, India, Photo: John Isaac 2. Statisticians entering data into the database for further processing and analysis, Turkmenistan. Photo: World Bank 3. High school students in La Ceja, Department of Antioquía, Colombia. Photo: Charlotte Kesl 4. A student explains to her class how to prevent HIV/AIDS, Cambodia. Photo: Masaru Goto Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 9 Table 1 Summary of SRH outcomes measured by WBG Gender IEs Outcomes of Interest Measured in IE Teen Sexual Condom Knowledge Pregnancy Risk Use of safe sex and Behaviors practices Childbearing Impact Evaluation Citation The Short-Term Impacts of a Schooling Conditional Cash Baird et    Transfer Program on the Sexual Behavior of Young Women in al., 2010 Malawi Education and HIV/AIDS Prevention: Evidence from a Duflo et     randomized evaluation in Western Kenya al., 2006 Empowering Adolescent Girls: Evidence from a Randomized Bandiera,    Control Trial in Uganda et al., 2012 Do Teenagers Respond to HIV Risk Information? Evidence Dupas,    from a Cluster Randomized Control Trial in Kenya 2011 Sexual Behavior Change Intentions and Actions in the Context Packel et   of a Randomized Trial of a Conditional Cash Transfer for HIV al., 2012 Prevention in Tanzania Does Demand or Supply Constrain Investments in Education?Heath and  Evidence from Garment Sector Jobs in Bangladesh Mobarak, 2012 Does Cash for School Influence Young Women's Behavior in Baez et  the Longer Term? Evidence from Pakistan al., 2011 Advancing Women’s Sexual and Reproductive Health: Lessons from World Bank Group Gender Impact Evaluations 10