Knowledge for Development Policy Brief: Malawi May 2016 Education Global Practice Beginning a Family and Adopting a Health, Nutrition, and Population Global Practice Healthy Lifestyle: Situation Analysis for Malawi Rifat Hasan, Corrina Moucheraud, and Anne Bakilana This policy brief examines two primary transi- marriage has declined somewhat since 1992, tions facing adolescents—beginning a family when it was 35 percent. The median age at and adopting a healthy lifestyle—and their first marriage was 17 years (MICS 2013–14), long-term effect on adolescents and their which represents very little change since 1992 communities as well as on Malawi’s potential when it was 17.8 years (DHS). Early marriage is to harness a demographic dividend. more common in rural than in urban areas: the percentage of currently-married adolescent Decisions Made in Adolescence Affect girls is 25 percent in rural areas compared to Health, Human Development and 19 percent in urban areas (MICS 2013–14; Economic Development p-value on t-test < 0.01) (figure 1). Median age Decisions made during youth have long-term at first sexual intercourse in 2010 (DHS) was impacts on human development, which is key to 17.2 years; this age has changed little since poverty alleviation and economic development. 2000 (when it was 16.8 years). During adolescence, two of the primary transi- tions are beginning a family and adopting a Figure 1. Share of Adolescent Girls Who Are healthy lifestyle. Youth face many choices and Married in Malawi, by Rural-Urban Location, 1996–2013/14 challenges around these key decisions, such as when to initiate sex, when to marry, when to 40 have children, whether to engage in risky behav- 36 35 iors, and what foods to consume—all of which 32 32 % women aged 15–19 who are married affect their future health and future opportuni- 30 ties. Adolescent girls also face risks related to 26 26 25 gender-based violence (GBV). Consequences of 25 21 these early decisions can have long-lasting 21 19 20 ­ effects on adolescents and their communities, potentially increasing public health costs and ­ 15 13 slowing the accumulation of human capital. 10 Trends in Marriage, Fertility, Nutrition, 5 and Health Early marriage is common in Malawi: 28 0 percent of surveyed adolescent girls be- 1992 DHS 2000 DHS 2004 DHS 2010 DHS 2013–14 MICS tween the ages of 15 and 19 reported Urban Rural being currently or previously married (MICS 2013–14). Overall, the prevalence of adolescent Source: Demographic and Health Surveys and MICS. 1 Figure 2. Fertility Rate of Adolescents (15–19 Years of Age) in Eastern and Southern Africa, Various Years, 2008–2013/14 180 160 Age-specific fertility rate (annual births 140 per 1,000 women), age 15–19 120 100 80 60 40 20 0 4 2 09 09 0 0 02 0 09 2 1 20 1 op 07 1 11 7 –1 01 01 01 –1 01 01 01 01 –0 20 20 20 8– – 6– 13 ,2 10 ,2 ,2 am i, 2 ,2 i, 2 ,2 08 06 0 e, 00 20 ia Za nda os o, , da ia ea aw nd 20 20 qu 20 th an or ,2 an a, itr e, ru al a bi so hi a, r, m nz bi , Ug bw ia Rw nd M Er Bu ca Et ny m Le Co ib Ta ila ba as Ke m oz az ag m Na M Sw Zi ad M Source: Most recent Demographic and Health Survey. 30% Fertility has declined only slowly, and ado- lescent fertility remains one of the highest in preferences of adolescents declined from an ideal of 4.3 children in 1992 to 3.4 children in The percent of adolescent Eastern and Southern Africa (figure 2). 2000, there has hardly been any change since girls in Malawi are pregnant or Fertility has declined overall (albeit slowly): the then (3.2 in 2004 and 2010). Given projected fer- already a mother. average number of children a woman would have tility and age patterns, approximately 18% of all over her lifetime if she had children at the cur- births in Malawi (between the years 2015 and rent rates of fertility is now 5 children (2013–14 2050) will be to adolescent mothers. MICS) compared to 6.7 children in 1992 (DHS). Use of modern contraception remains However, adolescent childbearing remains an lower among married adolescent girls than issue in Malawi. The median age when a woman among older women (figure 3). Between 1992 26% gives birth in Malawi is now 19 years (DHS 2010, MICS 2013–14 MICS), representing no change and 2010, modern contraceptive prevalence rate (mCPR) increased from 8 to 44 percent The percent of married since 1992. Similarly, the adolescent fertility rate among married women above age 20, com- adolescent girls in Malawi use ­ has barely changed over the 1992 to 2010 DHS pared with 2 and 26 percent, respectively, of modern contraceptives. reporting period, in both urban and rural areas. married adolescent girls (DHS). Among unmar- The fertility rate for girls 15–19 years of age who ried sexually active adolescents, the mCPR is live in rural areas was 162 in 2010 (versus 165 in 50 percent. Even when controlling for key de- 1992) and in urban areas it was 125 in 2010 (ver- mographic and socioeconomic factors, adoles- sus 134 in 1992). Approximately one-third—30 cents were approximately 50 percent less likely 50% percent—of adolescent girls are pregnant or than women aged 20-49 to use modern con- ­ already a mother (MICS 2013–14). Moreover, the traception.1 Unmet need—the proportion of The percent of unmarried proportion of adolescents who have given birth currently married women who want to stop or s ­ exually active adolescent or are pregnant is higher in rural than in urban space childbearing but are not using contra- areas, at 32 and 24 percent, respectively, and ception (or who are pregnant with a mistimed girls in Malawi use modern this difference is statistically significantly or unwanted pregnancy)—has declined among contraceptives. (p-value < 0.01) (2013–14 MICS). Additionally, (as explored in Bakilana, Moucheraud, McConnell, 1   Results of a multivariate regression model using data and Hasan 2016), early childbearing puts chil- from the 2013–14 MICS, on whether the current use of dren born to adolescent mothers at elevated risk modern methods of family planning is associated with for death, illness, and poor nutrition. Variation adolescence, current marital status, religion, region of residence, urban or rural residence, and household also exists by region: 26 percent in Northern and wealth quintile; the model was restricted to women who Central compared to 35 percent in Southern. were not pregnant but who were ever sexually active. Furthermore, while fertility Full findings are available from the authors upon request. 2 Figure 3. Use of Modern Contraceptive Methods Figure 4. Estimated Maternal Mortality Ratio in by Married Women in Malawi, by Age, 1992–2010 Malawi, 1990–2015 DHS, 2013–14 MICS 1,200 70 Maternal deaths per 100,000 live births (estimate) 65 % of currently married women, ages 15–19, 20–49, using 1,000 60 52 a modern method of contraception 800 50 634 44 600 40 424 29 400 30 28 25 20 17 200 13 10 8 0 2 1990 1995 2000 2005 2010 2015 0 Eastern Africa region Malawi 1992 DHS 2000 DHS 2004 DHS 2010 DHS 2013–14 MICS Source: Trends in maternal mortality: 1990 to 2015: Ages 15–19 Ages 20–49 estimates by WHO, UNICEF, UNFPA, World Bank Group and the United Nations Population Division. Geneva: Source: Demographic and Health Surveys and MICS. World Health Organization; 2015. married women in Malawi, although bigger skilled attendant at recent birth.3 However, gains were seen among women over age 20 there is evidence of ­ inequity—higher proportions than among adolescents (DHS), indicating that of women in urban areas (93 percent) use skilled there is room for more progress in reaching ad- delivery than women in rural areas (85 percent) olescent girls. according to the 2013–14 MICS, and this is a Maternal mortality declined in Malawi statistically significant difference (p-value < 0.01). but has more or less plateaued since 2005 There is a similar difference between urban and (figure 4). The maternal mortality ratio (MMR) rural adolescents (p-value 0.07). The rates also was estimated to be 634 maternal deaths per differ by regions (86 percent in Central and 100,000 live births in 2015, which represents Southern, and 92 percent in Northern). a 34 percent decline since 1990 (when the In contrast to skilled attendance at birth, ratio was 957) (WHO et al. 2015). While sub- use of the recommended antenatal care (ANC), stantial, the decline was insufficient to meet however, has been declining, with only 44 per- the fifth Millennium Development Goal target. cent of women receiving four or more ANC vis- Progress in Malawi also more or less stagnated its prior to delivery (2013–14 MICS), compared after 2005, versus the Eastern Africa regional 62 percent in 1992 (DHS). This decline in ANC average MMR which continued its decline. visits suggests that interventions delivered The percentage of recent births that were at- during ANC visits such as tetanus toxoid vacci- tended by a skilled provider (doctor, nurse, or nation, screening for and treatment of infec- midwife) increased for all women, from 53 per- tions, identification of warning signs during cent in 1992 (DHS) to 86 percent in 2013–14 pregnancy, and nutrition counseling are not (MICS)2; for adolescents, it is now 90 percent reaching those who need them—all of these and this is significantly higher than for women are important for not only maternal health and over age 20 (p− value < 0.01), among whom it is nutrition but also early childhood development. 86 percent. Even when controlling for key demo- Undernutrition among adolescents is a graphic and socioeconomic factors, adolescents concern, as a sizable proportion of adoles- were approximately 40 percent more likely than cents are classified as thin or anemic. women over age 20 to have had a According to the 2010 DHS, 81 percent 3   Results of a multivariate regression model using data   Skilled birth attendance is calculated for births in the 2 from the 2013–14 MICS, on whether a recent birth 3 years prior to the survey for DHS and for births in the (during the preceding 2 years) was attended by a doctor, 2 years prior to the survey for MICS. A more recent nurse or midwife, is associated with adolescence, timeframe generally results in higher proportions for current marital status, religion, region of residence, skilled attendance given the secular increase overall urban or rural residence, and household wealth quintile. during the time period. Full findings are available from the authors upon request. 3 of adolescent girls are classified as having a than age 20 years (29 percent versus 12 per- normal body mass index (BMI), 13 percent cent, respectively; p < 0.01) (MICS 2013–14). have a below-normal BMI, while 6 percent are However, adolescents were less likely to report classified as overweight or obese. In addition, being the final decision maker (solely or jointly) 28 percent of adolescent girls were classified about their own health care than older women: as anemic, and among adolescent girls who 47 percent versus 56 percent (p < 0.01) (2010 were pregnant, the figure is substantially DHS). There were statistically significant dif- higher at 41 percent, compared to 38 per- ferences in reporting decision-making power cent of pregnant women 20 years of age or for health care by women’s educational status: older (DHS 2010) (p-value on difference 0.1) only 48 percent of women with no education (figure 5). Low pre-pregnancy BMI, short reported being the final decision maker, versus ­ stature, and anemia can cause physical and 54 percent of women with primary level edu- emotional health problems, and maternal cation and 69 percent of women with second- underweight is a leading risk factor for pre- ­ ary education and beyond (2010 DHS). ventable deaths and disease. For example, pregnant adolescent girls with a low BMI may Policy Framework in Malawi be at increased risk for pregnancy and ob- Malawi has policies that aim to address the stetric complications and poor birth outcomes, health of women and adolescent girls, both 4.2% including obstructed labor, prematurity and low around fertility and healthy behaviors. The birthweight. Anemia during pregnancy is asso- multisectoral National Population Policy in- The percent of adolescent girls ciated with low birth weight of the child. cludes among its aims the enhanced scale-up Adolescent girls are exposed to risky of health and social services that help address in Malawi are HIV positive. sexual behaviors, with 4.2 percent of girls population challenges, particularly family plan- age 15–19 being HIV positive—and yet they ning, education, and empowerment of youth have less autonomy over their health care and women. The National Health Policy, the decisions. HIV prevalence is almost four times Health Sector Strategic Plan, and the Sexual as high among young women (ages 15–24 Reproductive Health and Rights (SRHR) years) in urban areas compared to those in Strategy aim to address sexual and reproduc- rural areas (11.2 percent and 3.7 percent, re- tive health and rights (including family planning) spectively, in 2010). Adolescents were more of women, men, and young people. Women’s likely to have received an HIV test during the empowerment, voice and agency in decision- preceding 12 months than other age groups making, and GBV are intricately linked with ad- (among ever sexually active women, 74 per- olescent pregnancies and use of contraception cent of adolescents and 51 percent of women and should be taken into consideration in de- older than age 20 years had received an HIV signing and implementing programmatic re- test over the preceding year, p value on differ- sponses. The National Youth Policy also ence < 0.01) (MICS 2013–14). Adolescents were specifically addresses youth-relevant topics in- also more likely to have used a condom during cluding comprehensive SRH education, and the their last sexual encounter than women older minimization of early marriage and of GBV. Health is one of the priority areas of the Gender Figure 5. Prevalence of anemia among pregnant Policy and its Implementation and M&E Plan. and non-pregnant women, by age group, 2010 While the policies in place attempt to com- prehensively address the health and nutrition 45 40.6 needs of adolescent girls, effective translation % of women with mild, medium, or 40 37.6 of policies and subsequent implementation 35 remains a challenge. A policy space analysis 27.9 28.4 indicates that there has been progress in im- severe anemia 30 25 plementation, including high-level support 20 from the government on SRH issues, creation 15 of technical working groups and other coordi- 10 nation bodies for various partners working on SRH, capacity-building for adolescent SRH, 5 creation of youth-friendly health service cen- 0 ters, clearly articulate program goals, a well- Anemia prevalence Anemia prevalence among among non- developed M&E framework, and rigorous pregnant women pregnant women monitoring efforts and mechanisms to revise Ages 15–19 Ages 20+ policies and programs based on results. In fact, these policies in Malawi specifically in- Source: Demographic and Health Survey 2010. clude several approaches recommended by 4 the global literature base, including promotion •• Malawi has one of the highest adolescent of family planning by community health fertility rates in Eastern and Southern workers (Health Surveillance Assistants), Africa with little change since 1992, and comprehensive sex education that includes 30 percent of girls are already mothers or messages about HIV prevention, training of pregnant. health workers to strengthen SRH service de- •• Married adolescent girls use modern livery, and creation of youth-friendly family contraception less than other married planning services. However, impediments to women. fully successful implementation remain. •• Undernutrition is a concern as adolescent Stakeholders reported insufficient budgetary girls are often thin or anemic. allocations for planned activities in the SRHR •• Adolescent girls engage in risky sexual Strategy with only a small percentage of bud- behaviors and have less autonomy over get earmarked, indicating inadequate govern- decision making regarding healthcare, ment financial commitment to these issues. putting them at risk for illness and This has implications for sufficient capacity to death. address adolescent SRH and further institu- While there have been some improve- tional development as well as technical ments for adolescents with regard to groups not having enough financing to have starting a family and adopting a healthy optimal impact. Furthermore, high-level lifestyle, further implementation progress government support is not matched by all ­ is necessary to help them to make better stakeholders, including opposition to adoles- decisions and prepare them to be healthy, cent SRH issues from the Ministry of productive, and contributing members of Education, religious and community leaders. society. Only by addressing sexual and repro- In addition, the multi-sectoral Population ductive health, nutrition, and risky sexual be- and Youth Policies, which are also related to haviors will the negative long-term effects of adolescent health and nutrition, have bene- decisions made during adolescence be averted fited from visible support from the President and the demographic dividend be harnessed in and feature in development plans as well as Malawi. across ministerial plans and budgets. Both policies have created new central institutions, and the NYP has created units throughout References Bakilana, Anne, Corrina Moucheraud, Christin Malawi, which have the potential to serve as McConnell, and Rifat Hasan. 20106. “Early a platform for program implementation by Childhood Development: Situation Analysis for partners within and outside government. The Malawi.” Policy Brief: Malawi, World Bank, policies are accompanied by strategic/opera- Washington, DC. National Statistical Office (NSO). 2015. “Malawi MDG tional/action plans, which include clear pro- Endline Survey 2014.” NSO, Zomba, Malawi. gram goals and M&E frameworks. However, NSO and ICF Macro. 2011. “Malawi Demographic and they are under-resourced and experience Health Survey 2010.” NSO and ICF Macro, Zomba, staffing shortages, jeopardizing effective im- Malawi, and Calverton, Maryland, USA. plementation. Furthermore, the low visibility NSO [Malawi] and Macro International. 1994. “Malawi Demographic and Health Survey 1992.” NSO and of the National Population Policy may con- Macro International, Zomba, Malawi, and Calverton, tribute to its overall challenges with staffing. Maryland, USA. NSO [Malawi], and ORC Macro. 2001. “Malawi Key Findings and Conclusions Demographic and Health Survey 2000.” Zomba, Substantial gaps remain to be addressed Malawi, and Calverton, Maryland, USA. ———. 2005. “Malawi Demographic and Health regarding key decisions made by adoles- Survey  2004.” NSO and ORC Macro, Calverton, cents about starting families and adopting Maryland. healthy lifestyles: Trends in maternal mortality: 1990 to 2015: estimates by WHO, UNICEF, UNFPA, World Bank Group and the © 2016 International Bank for Reconstruction •• Early marriage remains common with 28 United Nations Population Division. Geneva: World and Development / The World Bank. Some rights reserved. The findings, interpretations, percent of adolescent girls married. Health Organization; 2015. and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. This work is subject to a CC BY 3.0 IGO license (https://creativecommons.org/licenses/by/3.0/ igo). The World Bank does not necessarily own each component of the content. It is your responsibility to determine whether permission is needed for reuse and to obtain permission from the copyright owner. If you have questions, email pubrights@ worldbank.org. SKU K8734 5