ADOLESCENT PREGNANCY IN THE DOMINICAN REPUBLIC World Bank Policy Note September, 2020 International Development Association Macroeconomics, Trade and Investment Global Practice Latin America and the Caribbean Region © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, 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, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2625; e-mail: pubrights@worldbank.org. ii DOMINICAN REPUBLIC GOVERNMENT FISCAL YEAR January, 1 – December, 31 CURRENCY EQUIVALENTS Currency Unit: Dominican Peso (RD$) (Exchange Rate Effective as of 5/20/2020) US$1.00 = RD$55.45 Abbreviations and Acronyms ADESS Administrator of Social Subsidies (Administradora de Subsidios Sociales) AFP Pension Fund Administrator (Administradora de Fondos de Pensiones) ARS Health Risk Administrators (Administradoras de Riesgos de Salud) ASPIRE Atlas of Social Protection Indicators of Resilience and Equity BEEP Student Progress Bonus (Bono Escolar Estudiando Progreso) BIA Benefit-Incidence Analysis CAFI Attention Center for Family and Children (Centro de Atención a la Infancia y la Familia) CAID Attention Center for Disability (Centro de Atención Integral para la Discapacidad) Center for the Integral Attention of Early Childhood ( Centro de Atención Integral a la CAIPI Primera Infancia) BCRD Central Bank of the Dominican Republic (Banco Central de la República Dominicana) CND Drugs National Council CNPS Social Protection National Council CNSS Social Security National Council COE Emergency Operations Center CONADIS Disability National Council CONANI Children and Youth National Council CONAPE Elderly National Council CORAA Aqueduct and Sewer Corporation Covid-19 Coronavirus disease SARS-CoV-2 CTC Community Technology Centers DIGEPEP General Directorate of Special Programs of the Presidency DIGEPRES General Directorate of Budgeting DR Dominican Republic ENCFT Continuous National Labor Survey ECD Infants Centers ECLAC Economic Commission for Latin America (CEPAL in Spanish) END National Development Strategy FONPER Equity Fund for Reformed Institutions GDP Gross Domestic Product GCPS Social Policy Coordination Cabinet (Gabinete de Coordinación de Políticas Sociales) GLP Liquified Gas Petroleum GNI Gross National Income IAD Agrarian Dominican Institute ICV Quality of Life Index iii IDECOOP Institute for Cooperative Development and Credit ILAE School Attendance Incentive INABIE National Institute for Student Welfare INAIPI National Institute for the Integral Attention for Childhood Development INAVI Aid and Housing Institute INDOTEL Dominican Telecommunication Institute INDRHI Institute of Hydraulic Resources INESPRE Institute for the Stabilization of Prices INFOTEP Professional Technical Training Institute INVI Institute of Housing ISSFFAA Institute of Social Security of Armed Forces JCE Central Electoral Council MAP Administrative Ministry of the Presidency MEPYD Ministry of Economy, Planning, and Development MH Ministry of Finance MICM Ministry of Industry and Commerce and Small and Medium Enterprises MINERD Ministry of Education MIREX Ministry of Foreign Affairs MISPAS Ministry of Health and Social Assistance MM Ministry of Women MT Ministry of Labor NNA Children and Adolescents ODS Sustainable Development Goals OECD Organisation for Economic Co-operation and Development ONE National Office of Statistics OPRET Office for the Reorganization of Land Transportation PAE School Feeding Program PEISE Strategic Sectorial Plan for the Economic and Social Inclusion PIB Gross Domestic Product PNPSP Pluriannual Plan of the Public Sector PNUD United Nation Development Program POA Operational Annual Plan PPA Power Purchase Agreement PPP Purchasing Power Parity PRISS Social Security Tax Recovery and Information System PROCOMUNIDAD Promotion Fund for Community Initiatives PROMESECAL Essential Medicines Program / Logistics Support Center PROSOLI Progressing with Solidarity PROVEE Program for the Protection of the Elderly in Extreme Poverty PV Elderly Protection Program PSA Adaptative Social Protection PTMC Conditional Cash Transfer Program RD Dominican Republic SDSS Social Security Dominican System SENASA National Health Insurance SFS Family Health Insurance SIGEF Integrated Financial Management System SIPEN Superintendence of Pensions iv SISALRIL Superintendence for Health and Labor Risk SISDOM Dominican Social Indicator System SIUBEN Single System of Beneficiaries SNPMRD National System for Disaster Prevention, Mitigation, and Response SNSS Health National Service System SP Social Protection SVDS Life, Disability and Survival Insurance Vice President: Carlos Felipe Jaramillo Country Director: Michel Kerf Global Director: Marcello Estevão Sector Manager: Jorge Thompson Araujo Task Team Leader: Johannes Herderschee and Miriam Montenegro v Acknowledgements This Policy Note is part of an ongoing dialogue between World Bank staff and the Dominican Republic’s Ministry of Finance and Ministry of Planning and Economic Development. The Policy Note supports the analysis presented in the Public Expenditure Review (PER) series that was developed to support the dialogues. At the Ministry of Finance, the task was managed by Mr. José Luis Actis, Advisor to the Minister, and supported by Paola Maria Vargas Antigua, Coordinator, and Camila Hernandez Villaman, Economist. At the Ministry of Planning and Economic Development, the task was managed by Martin Franco, Public Finance Unit Coordinator, UAAES, and Paula Salvador, World Bank Group Project Coordinator. Experts from sector ministries were invited to collaborate with the World Bank team on the preparation of each chapter. At the request of the Government of the Dominican Republic, the French Development Agency (Agence Française de Développement, AFD) provided financial support for this project. AFD Director Sandra Kassab was instrumental in the preparation of the analysis, and valuable contributions were made by Eleonore Pocry, Country Officer in charge of DR and Martinique and Gwenola Pellen, Procurement Officer, in Paris and Lucas Gimenez, Project Officer, in Santo Domingo. This PER series was prepared by an interdisciplinary team led by World Bank staff. Johannes (Han) Herderschee (Senior Country Economist) coordinated the team in close collaboration with Craig Kullmann (Senior Water Supply and Sanitation Specialist, SLCWA), Miriam Montenegro (Senior Social Protection Specialist, HLCSP), James Sampi Bravo (Economist, ELCMU), Anjali Shahani (Operations Officer, LCC3C), and Carmen Amaro (Operations Officer, LCCDO). The Social Protection Team was led by Miriam Montenegro (Senior Social Protection Specialist, HLCSP) and consisted of Maria Concepcion Steta (Senior Social Protection Specialist, HLCSP), Magdalena Lizardo (Consultant, ELCMU), Julian Valdés (Consultant, ELCMU), José Pellerano (Consultant, ELCMU), Daniel de la Rosa (Consultant, ELCMU), and Lisselotte Gálvez (Consultant, ELCMU). The team benefited from guidance and advice provided by Jorge Araujo (Manager, MTI), Rita Cestti (Practice Manager, SLCWA), and Pablo Gottret (Practice Manager, HLCSP). The team would also like to acknowledge the valuable overall guidance and comments provided by Tahseen Sayed Kahn (Country Director for the Caribbean), Michel Kerf (Country Director for Central America and the Dominican Republic), Alessandro Legrottaglie (Country Manager until June 30, 2020), Alexandria Valerio (Country Manager since July 1, 2020), Abha Prasad (Program Leader), Timothy Johnston (Program Leader), and Vickram Cuttaree (Program Leader). Ricardo Habalian (Operations Officer, LCC3C) prepared the first draft of the Executive Summary with contributions from team members. Juan Carlos Parra Osorio (Senior Economist, ELCPV until June 30, 2020) and Alejandro De la Fuente (Senior Economist, ELCPV since July 1, 2020) advised on poverty issues and the expenditure- impact analysis. Sincere thanks are also due to the peer reviewers who contributed at various stages of the analysis, including: Aline Coudouel (Lead Economist HSASP), Mark Charles Dorfman (Senior Economist, HEASP), Henri Fortin (Lead Financial Management Specialist, EA2G2), Chris Heymans (Senior Water and Sanitation Specialist, SAFW3), Phil Keefer (Principal Economic Advisor, IDB), Professor Thierry Madies (University of Fribourg), and Guillermo Javier Vuletin (Senior Economist, LCRCE). The team would also like to thank Oscar Parlback (Consultant) and Sean Lothrop (Consultant) for much- valued assistance in revising drafts and editing the PER. Sylvaine Cussac (Consultant) formatted this version. Giselle Velasquez (Program Assistant, ELCMU) greatly supported this work by tirelessly facilitating administrative processes. Maria Hermann (Executive Assistant in the World Bank’s Santo Domingo office) and Samahara Hernandez Angulo (Team Assistant in the Santo Domingo office) supported the numerous consultations of the visiting missions. vi The Policy Note on Teenage pregnancy was prepared by a team led by Miriam Montenegro (Senior Social Protection Specialist, HLCSP) and consisting of Maria Concepcion Steta Gandara (Senior Social Protection Specialist, HLCSP), Magdalena Lizardo (Consultant, ELCMU), Marygracia Aquino (Consultant, ELCMU), and Daniel de la Rosa (Consultant, ELCMU). The team benefited from support provided by national officials, Rosa Suarez (Social Cabinet Director), Hector Medina (Director of PROSOLI), Matilde Chavez (Director of SIUBEN), Julissa Holguin (Head of the Policy Unit at the Social Cabinet). Representatives of these different institutions implementing social protection programs also provided information during the preparation of this report. The team is grateful for peer review comments from Miriam Muller (Social Scientist, ELCPV), Timothy Johnston (Program Leader, HLCDR), and Juan Baron (Senior Economist, HSAED). vii Table of contents Abbreviations and Acronyms .................................................................................................................... iii Acknowledgements .................................................................................................................................. vi Executive Summary................................................................................................................................... xi I. Introduction ........................................................................................................................................1 II. Trends in Adolescent Pregnancy ........................................................................................................2 II.1 Pregnancy in Early and Late Adolescence ............................................................................2 II.2 Adolescent Pregnancy Rates in International Perspective ....................................................7 II.3 Factors Associated with Adolescent Pregnancy ..................................................................11 II.4 Effects of Adolescent Pregnancy .........................................................................................25 III. Policies, Programs, and Public Spending on Adolescent Pregnancy in the DR ...............................29 III.1 Policies and Plans ...............................................................................................................29 III.2 Adolescent Pregnancy Prevention Programs and Support for Adolescent Mothers ..........31 III.3 Public Spending on Adolescent Pregnancy Prevention Programs and Support for Adolescent Mothers ......................................................................................................................34 IV. The Effectiveness of Adolescent Pregnancy Prevention and Protection Programs ........................38 IV.1 Program Coverage among Late Adolescents .....................................................................38 IV.2 Participation in Social Programs and Other Interventions and the Incidence of Adolescent Pregnancy ......................................................................................................................................41 V. Conclusions and Recommendations ................................................................................................45 Bibliography .............................................................................................................................................50 Appendices ..............................................................................................................................................54 List of figures Figure 1. Share of Women Ages 20-24 Who Report Having Had Their First Child before the Age of 20, 1991-2013................................................................................................................................................... 2 Figure 2. Number of Births by Maternal Age Group, 2001-2019 ................................................................ 3 Figure 3. Female Population by Age Group, 2000-2019.............................................................................. 4 Figure 4. Share of the Population without a National Birth Certificate by Age Group and Wealth Quintile, 2002-2018................................................................................................................................................... 4 Figure 5. Adolescent Pregnancy Rates by Maternal Age Group and Rural/Urban Location, 1991-2013 ..... 5 Figure 6. Ratios between the Incidence of Adolescent Pregnancy in Rural and Urban Areas by Maternal Age Group, 1991-2013 ................................................................................................................................ 5 Figure 7. Adolescent Pregnancy Rates by Maternal Age Group and Household Wealth Quintile, 1996 and 2013 ............................................................................................................................................................ 6 Figure 8. Ratios between Adolescent Pregnancy Rates in the Poorest Income Quintile and All Other Quintiles by Maternal Age Group, 2007 and 2013 ...................................................................................... 7 Figure 9. Late Adolescent Fertility Rates by Region and Income Group, 2017 ............................................ 7 Figure 10. Late Adolescent Pregnancy Rates in the Dominican Republic versus Global and Regional Averages, 1960-2017 .................................................................................................................................. 8 Figure 11. Human Capital Index Scores and Late Adolescent Pregnancy Rates in High-Income and Upper- Middle-Income Countries, 2017 ................................................................................................................. 9 Figure 12. Maternal Mortality Rates and Late Adolescent Pregnancy Rates in High-Income and Upper- Middle-Income Countries, 2017 ............................................................................................................... 10 viii Figure 13. Neonatal Mortality Rates and Late Adolescent Pregnancy Rates in High-Income and Upper- Middle-Income Countries, 2017 ............................................................................................................... 10 Figure 14. Infant Mortality Rates and Late Adolescent Pregnancy Rate in High-Income and Upper- Middle-Income Countries, 2017 ............................................................................................................... 11 Figure 15. The Late Adolescent Pregnancy Rate and Real GDP per Capita in the Dominican Republic, 1960-2017................................................................................................................................................. 11 Figure 16. Share of Adolescents Ages 15-19 Who Report Having Been Pregnant by Region, 2018 .......... 14 Figure 17. Share of Late Adolescents Who Report Having Been Pregnant by Current Educational and Employment Status, 2018 ......................................................................................................................... 16 Figure 18. Reasons for Leaving School or University among Late Adolescents Who Left during Their First Pregnancy, 2018 ....................................................................................................................................... 16 Figure 19. Reasons for Leaving School or University among Late Adolescents Who Left after the Birth of Their First Child by Household Socioeconomic Status, 2018..................................................................... 17 Figure 20. Share of Late Adolescents with Fewer Years of Schooling than Would Be Expected for Their Age by Pregnancy History and Socioeconomic Status, 2018 ..................................................................... 17 Figure 21. Age of First Sexual Intercourse among Late Adolescents by Socioeconomic Status and Pregnancy History, 2018 ........................................................................................................................... 19 Figure 22. Reasons for Postponing Sexual Intercourse among Late Adolescents Who Have Never Had Sexual Intercourse, 2018 .......................................................................................................................... 19 Figure 23. Negative Consequences of Early Marriage or Sexual Union Cited by Late Adolescents, 2018 . 20 Figure 24. Positive Aspects of Early Marriage or Sexual Union Cited by Late Adolescents, 2018 ............. 20 Figure 25. The Annual Budget for Adolescent Pregnancy Prevention under PROSOLI by Program, 2015-18 .................................................................................................................................................................. 36 Figure 26. Share of Late Adolescents Who Have Had Sexual Intercourse Who Report Using a Modern Contraceptive Method by Socioeconomic Characteristics and Pregnancy History, 2018 ......................... 39 Figure 27. Sources of Contraception among Late Adolescents Who Have Had Sexual Intercourse by Socioeconomic Characteristics and Pregnancy History, 2018 ................................................................... 39 Figure 28. Preferred Source of Assistance with Reproductive-Health-Related Concerns by Socioeconomic Characteristics and Pregnancy History, 2018 (%) ...................................................................................... 41 List of tables Table 1. Late Adolescent Pregnancy Rates by Socioeconomic Characteristics and Location, 2018 .......... 13 Table 2. Incidence of Lost or Aborted Pregnancies among Late Adolescents by Rural/Urban Area and Socioeconomic Status, 2018 (%) ............................................................................................................... 15 Table 3. Educational Attainment among Late Adolescents by Pregnancy History and Education Enrollment or Employment Status, 2018 .................................................................................................. 18 Table 4. Probability of Late Adolescent Pregnancy ................................................................................... 22 Table 5. Effect early and late adolescent pregnancy on labor-market participation, educational attainment, and reliance on social programs among female heads of household ages 20-30 ................. 27 Table 6. Expected Decrease in the Overall Adolescent Pregnancy Rate under the PREA, 2019-2023....... 30 Table 7. Government Programs Designed to Prevent Adolescent Pregnancy and Support Adolescent Mothers, 2012-2018 ................................................................................................................................. 32 Table 8. Programs Designed to Prevent Adolescent Pregnancy, Support Adolescent Mothers, or Provide Sexual and Reproductive Education by Number of Beneficiaries, 2015-2019 .......................................... 34 Table 9. Public Spending on Adolescent Pregnancy Prevention Programs, Support for Adolescent Mothers, and Sexual and Reproductive Health Education, 2018-19......................................................... 35 ix Table 10. Share of Late Adolescents Who Have Visited a Health Center in the Last 12 Months by Type of Health Service Requested, Type of Facility, and Pregnancy History, 2018 ................................................ 37 Table 11. Share of Late Adolescents Who Report Having Attended a School or University Class in Which Human Anatomy and Reproductive Physiology were Taught by Socioeconomic Characteristics and Pregnancy History, 2018 ........................................................................................................................... 38 Table 12. Sources of Sexual and Reproductive Health Information among Late Adolescents by Pregnancy History, 2018 (%) ...................................................................................................................................... 40 Table 13. Statistical Correlations between Participation in Social Programs and Other Interventions and the Probability of Late Adolescent Pregnancy .......................................................................................... 43 Table A. 14. Legal Framework Supporting Adolescent Pregnancy Prevention and Protection Programs . 54 Table A. 15. Mean age at first sex among adolescents aged 15-19, by instruction in SRHE, area of residence, family socioeconomic group and years of school completed, 2018 ........................................ 57 x Executive Summary In the Dominican Republic (DR), decades of robust economic growth have yielded sustainable, broad- based improvements in living standards and poverty reduction, yet persistently high rates of adolescent pregnancy continue to adversely affect the lives of women and girls across the country. The DR’s pregnancy rate among girls aged 15-19 is almost 50 percent higher than the regional average for Latin America and the Caribbean and more than three times the average for upper-middle-income countries worldwide. If current trends continue, the DR’s adolescent pregnancy rate will soon exc eed the Sub- Saharan Africa average for the first time since the 1960s. In the DR, as in countries around the world, high rates of adolescent pregnancy are associated with increased infant and maternal mortality, diminished educational attainment among women and girls, greater socioeconomic inequality, more persistent intergenerational poverty, and a wide range of negative social outcomes, including domestic violence and substance abuse. To achieve its economic and social development objectives and retain its position as a regional leader in poverty reduction and shared prosperity, the DR must effectively address both the causes and consequences of adolescent pregnancy. To date, the government’s efforts to reduce the incidence of adolescent pregnancy have b een limited, fragmentary, and sporadic. Spending on policies and programs that directly or indirectly focus on adolescent pregnancy is very modest, and a pervasive lack of monitoring and evaluation precludes a rigorous empirical assessment of the impact of the current approach. However, statistical analysis sheds light on the individual-, household-, and community-level factors that are associated with a higher risk of adolescent pregnancy. After controlling for cross-correlations, the probability of becoming pregnant during the late adolescence is positively associated with being involved in an early marriage or sexual union, having a lower level of educational attainment, and being neither engaged in the labor market nor enrolled in school. While in absolute terms adolescent pregnancy rates are highest in rural areas and border regions and among lower-income households and immigrant women who were born in Haiti, these relationships vanish when circumstantial factors have been accounted for. After controlling for cross-correlations, the highest adolescent pregnancy rates are found in urban areas and in the regions of Higuamo and Valdesia, and among non-poor households and women born in the DR. Having been pregnant in adolescence is associated with lower levels of educational attainment and increased reliance on social programs. Female heads of household in their 20s and 30s who were adolescent mothers are less likely than their peers to have completed primary, secondary, or tertiary education and are more likely to depend on public social programs such as Comer es Primero and Seguro Familiar de Salud del Regimen Subsidiado. Moreover, 31.1 percent of adolescents who report having been pregnant did not attend school or college during their first pregnancy, and 48.7 percent dropped out after the birth of their first child. Almost all negative outcomes were more common among female heads of household who had become pregnant before the age of 15. Data limitations notwithstanding, there is evidence that individual-, household-, or community-level participation in certain programs and initiatives is associated with lower rates of adolescent pregnancy. Having received sexual and reproductive health education at school or college is significantly correlated with lower pregnancy rates, and the gap between girls who have and have not had access to reproductive health education widens among the poorest households. Adolescents at all income levels are more likely to seek sexual-health-related information and support from a health center than from any other source, xi and access to sexual health services at a health center is especially crucial for adolescents from moderately and extremely poor households. Municipal-level participation in the Jornada Escolar Extendida program and household-level participation in Comer es Primero and Escuelas de Familia are associated with a lower probability of late adolescent pregnancy, while public outreach campaigns have been limited, and no information is available on their effectiveness. The analysis presented in this policy note yields six key recommendations. To enhance the effectiveness of its efforts to reduce adolescent pregnancy, the government should: (i) implement a comprehensive strategy to bolster sexual and reproductive health education, particularly at early adolescence; enhance access to family planning services, and strengthen socio-emotional skills, life skills, and the development of self-esteem and personal agency among adolescent girls; (ii) issue a Presidential Decree defining the role of each public agency involved in implementing the PREA, establish monitoring and evaluation mechanisms, and allocate adequate budgets across agencies; (iii) expand social programs that are associated with a lower probability of adolescent pregnancy, including Jornada Escolar Extendida, Comer es Primero, and Escuelas de Familia; (iv) develop new programs to expand access to modern contraceptive methods and ensure that students receive reproductive health education in public school; and (v) strengthen the provision of comprehensive reproductive health services for adolescents in public hospitals and clinics, including care for adolescent mothers in order to avoid repeated adolescent pregnancy. xii I. Introduction 1. Over the past several decades, robust economic growth and sustained poverty reduction have made the Dominican Republic (DR) a regional success story, yet the country’s adolescent pregnancy rate remains far above the average for Latin America and the Caribbean (LAC) and is closer to the levels observed in Sub-Saharan Africa (SSA). Since the late 1990s, the DR’s adolescent fertility rate has declined at a slower pace than those of peer countries in both LAC and SSA. For example, for the period of 2008- 2017, the average growth rate of adolescent fertility rate for DR reached (-0.99%) compared to (-1.5%) in SSA. Adolescent fertility rate in 2017 in DR reached 94.3 (births per 1000 women aged 15-19) compared to 102.8 in SSA. If current trends continue, the DR’s adolescent fertility rate will exceed the SSA average in 2033 for the first time since the late 1960s. 2. Adolescent pregnancy poses severe risks to girls and young women, including elevated rates of maternal1 and child mortality, and is strongly correlated with sexual abuse2 and domestic violence. Adolescent pregnancy also sharply constricts opportunities for educational attainment, economic productivity, and autonomy over reproductive decisions and the overall trajectory of their lives.3 The consequences of adolescent pregnancy are especially severe if they occur in the early adolescence (defined as ages10-14) rather than late adolescence (ages 15-19),4 but even late adolescent pregnancy is associated with deeply negative consequences for young women and their families. 3. This report analyzes the DR’s persistently high rate of adolescent pregnancy, identifies its main determinants and consequences, and evaluates the effectiveness of policies and programs designed to help reduce adolescent pregnancy and mitigate its adverse personal and social impacts. This report presents a set of stylized facts on adolescent pregnancy in the DR and uses an econometric model to estimate the statistical significance of risk factors for adolescent pregnancy and its negative effects at the individual and household levels. Finally, an assessment of early adolescent versus late adolescent pregnancy reveals important differences in their causes and implications. The information used in this study is derived from household surveys and national administrative records. 5 1 Chen et al. (2006), based on information from the United States, show that adolescent pregnancy increases birth risks regardless of socioeconomic status and quality of prenatal care. Count-Agudelo et al. (2005) present similar conclusions using information from Latin America. 2 Baumgartner, J. (2009). 3 Azevedo, J. et al (2012). 4 Conde-Agudelo et al. (2005) report that the risk of dying from pregnancy-related causes is four times higher in 15-year-olds or less than in women ages 20-24, while for ages 16-19 the risk is almost similar to that of women aged 20-24. Similarly, early neonatal death is higher in children of adolescents 15 years of age or younger compared to other women ages 16-24. 5 Orc Macro/Measure Dhs+/CESDEM (1991, 1996, 2002, 2007, 2013): Demographic and Health Surveys (ENDESA); ONE-UNICEF (2016): Multi-Indicator Cluster Survey (MICS 2014); ONE (2018): National Multipurpose Household Survey (ENHOGAR 2018) and Central Bank of the Dominican Republic (2018): Ongoing National Labor Force Survey (Encuesta Nacional Continua de Fuerza de Trabajo, ENCFT). 1 II. Trends in Adolescent Pregnancy II.1 Pregnancy in Early and Late Adolescence 4. In the DR, an unusually large share of women report having had their first child before the age of 20, and a much smaller but still substantial share report having had a child before the age of 15. In 2013, over 40 percent of women ages 20 to 24 reported having had their first child before the age of 20: 17.1 percent had their first child between the ages of 18 and 19, 20.7 percent between the ages of 15 and 17, and 3.1 percent between the ages of 10 and 14 (Graph 1). In other words, almost one-quarter of women ages 20 to 24 reported having had their first child at an age when they should have been completing either primary (10-14 years) or secondary education (15-17 years). MacQuarrie et al. (2017) find that the DR’s adolescent fertility curve begins rising at age 12, below the minimum age in countries such as Colombia, Guatemala, Haiti, Honduras, and Peru. 6 In 2013, the share of girls between the ages of 10 and 14 that were either married or cohabitating with an intimate partner was 11 percent, the highest rate in LAC. 7 Rates of child marriage and early sexual unions are strongly correlated with early adolescent pregnancy.8 Figure 1. Share of Women Ages 20-24 Who Report Having Had Their First Child before the Age of 20, 1991-2013 47.5 45.0 42.5 40.0 37.5 % Women age 20-24 35.0 32.5 30.0 27.5 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 10-14 15-17 18-19 15-19 10-19 Age when having first child Estimate Lower bound Upper bound Source: Authors’ elaboration based on data from ENDESA 1991, 1996, 2002, 2007, 2013. Note: 95% confidence intervals. 5. The DR’s adolescent pregnancy rate began rising in 1991, peaked in 2002, and remained broadly constant until 2013, the most recent year for which comparable data are available. In 1991, 33.1 percent of women ages 20 to 24 reported having had their first child before the age of 20. By 2002, this share had reached 43.7 percent, and in 2013 it was statistically unchanged at 40.1 percent. The overall increase in 6 MacGuarrie et al. (2017). Fig A6.6. 7 MacGuarrie et al. (2017). Pag. 15. 8 Bank World-UNICEF (2017). Page. 4. 2 adolescent pregnancy was driven by a rising share of women who reported having had their first child during late adolescence. The share of women who reported having had their first child between the ages of 15 and 17 rose from 14.8 percent in 1991 to 20.7 percent in 2013, while the share who reported having had their first child between 18 and 19 ticked up from 15.7 percent in 1991 to 17.1 in 2013, and the share who reported having had their first child between 10 and 14 remained statistically unchanged. 6. Despite the persistently high adolescent pregnancy rate, national statistics indicate that the number of births to adolescent girls has declined over the past decade. The number of births among girls ages 10 to 14 has dropped substantially since 2001, and this trend accelerated in 2011, while births among girls ages 15 to 19 have also fallen steadily since 2011 (Figure 2). Increased social awareness of the risks and challenges associated with adolescent pregnancy increased in the late 2000s, when the government launched several programs designed to reduce the adolescent birth rate. As the share of the female population between the ages of 10 and 14 remained unchanged from 2005 to 2016 and declined thereafter, demographic change did not offset the decline in the adolescent birth rate (Figure 3). However, national statistics should be interpreted with caution due to the large share of Dominicans who lack a birth certificate. In 2002, 13.2 percent of the population did not have a birth certificate, including 22.3 percent of children under the age of 4 and 11.7 percent between the ages of 5 and 9. Birth registration has improved in recent years, and the share of the population lacking a birth certificate fell to 2.1 percent in 2018. However, this share rises to 7 percent among households in the lowest income quintile, which is also the quintile with the highest rate of adolescent pregnancy (Figure 4). Figure 2. Number of Births by Maternal Age Group, 2001-2019 70,000 3,500 60,000 3,000 Ages 15 and over 50,000 2,500 40,000 2,000 Ages 14 and under 30,000 1,500 20,000 1,000 10,000 500 0 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 < 15 15-19 20-24 25-29 30-34 35-39 Source: Authors’ elaboration based on ONE data (2019). 3 Figure 3. Female Population by Age Group, 2000-2019 490,000 960,000 Women aged 10-14 and 15-19 480,000 950,000 470,000 940,000 Women aged 10-19 460,000 930,000 450,000 920,000 440,000 910,000 430,000 900,000 420,000 890,000 410,000 880,000 2001 2003 2005 2007 2009 2011 2013 2015 2017 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2019 10-19 10-14 15-19 Source: Authors’ elaboration based on ONE data: https://www.one.gob.do/demograficas/proyecciones- de-poblacion Figure 4. Share of the Population without a National Birth Certificate by Age Group and Wealth Quintile, 2002-2018 25.0 22.3 20.0 % Population 15.0 13.2 13.4 11.7 10.0 8.0 7.9 6.9 6.1 5.0 4.0 5.0 3.0 3.4 2.1 1.8 1.5 0.6 0.7 0.1 - Entire 0-4 5-9 10-14 Q1 Q2 Q3 Q4 Q5 population Age Wealth quintile 2002 2007 2013 2018 Source: Authors’ elaboration based on ENDESA 2002, 2007, 2013 and ENHOGAR 2018. 7. Between 1991 and 2013, the adolescent pregnancy rate in rural areas was consistently higher than the rate in urban centers, but the urban rate increased significantly over the period. In the urban area, adolescent pregnancy rates rose substantially between the 1990s and 2000s, with the largest increase observed among 15-year-old girls. In rural areas, the adolescent pregnancy rate spiked between the mid-1990s and the late 2000s, but by 2013 it had almost returned to its 1991 level. Meanwhile, the adolescent pregnancy rate in rural areas increased substantially, and in 2013 the urban and rural rates for girls ages 15 to 17 were close to converging. 4 Figure 5. Adolescent Pregnancy Rates by Maternal Age Group and Rural/Urban Location, 1991-2013 60.0 % Women age 20-24 50.0 40.0 30.0 20.0 10.0 - 1996 2007 2013 1991 2002 2007 2013 1991 1996 2002 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 2013 1991 1996 2002 2007 Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban 10-14 15-17 18-19 15-19 Age of having first child Estimate Lower bound Upper bound Source: Authors’ elaboration based on data from ENDESA 1991, 1996, 2002, 2007, 2013. Note: 95% confidence intervals. 8. Between 1991 and 2013, the gap between the adolescent pregnancy rates in rural and urban areas narrowed for every age group. Among girls ages 10 to 14, the ratio of births in rural and urban areas was 2.1 to 1 in 1991, but by 2013 this ratio had fallen to 1.8 to 1. For most years, the rural/urban ratio is highest among girls ages 10 to 14, which suggests the factors that contribute to early adolescent pregnancy are especially ingrained in rural areas. Meanwhile, the rural/urban ratios for late adolescent pregnancy ratios have tended to converge over time, as rural rates have fallen while urban rates have risen (Figure 6). Figure 6. Ratios between the Incidence of Adolescent Pregnancy in Rural and Urban Areas by Maternal Age Group, 1991-2013 2.50 2.10 1.87 % rural / % urban 2.00 1.81 1.56 1.60 1.50 1.24 1.31 1.20 1.16 1.07 1.00 0.50 - 10-14 15-17 18-19 15-19 10-19 Age of having first child 1991 1996 2002 2007 2013 Source: Authors’ elaboration based on data from ENDESA 1991, 1996, 2002, 2007, 2013. 9. Adolescent pregnancy rates are inversely correlated with household income level, and this pattern likely contributes to intergenerational poverty and inequality. Adolescent pregnancy rates systematically decline as income levels rise, and households in the bottom two quintiles have the highest 5 rates across all maternal age groups and time periods (Figure 6). These disparities have remained large and relatively stable over time: in 1991 and in 2013 more than 60 percent of women in the poorest income quintile reported having their first child before age 20, compared to just over 20 percent of women in the wealthiest quintile. Rates of early adolescent pregnancy have fallen to near zero among women from households in the wealthiest quintile but remain significant among women from households in the poorest quintile. In 2013, for every girl in the wealthiest quintile who had her first child between the ages of 10 and 14, there were 5.2 girls in the poorest quintile and 5.8 in the second poorest. This gap was smaller but still significant among women who had their first child between the ages of 18 and 19, with ratios between the wealthiest and poorest quintiles of less than 1 to 2 in 2013. The gap was also narrower than it had been in 2007, due in part to an increase in the incidence of pregnancy among 18-19-year-olds in the wealthiest quintile (Figure 8). Figure 7. Adolescent Pregnancy Rates by Maternal Age Group and Household Wealth Quintile, 1996 and 2013 80.0 70.0 60.0 % Women aged 20-24 50.0 40.0 30.0 20.0 10.0 - Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 2007 2013 2007 2013 2007 2013 2007 2013 2007 2013 10-14 15-17 18-19 15-19 10-19 Wealth quintile, survey year and age of having first child Estimate Lower bound Upper bound Source: Authors’ elaboration based on data from DHS 1996, 2007 and 2013. 6 Figure 8. Ratios between Adolescent Pregnancy Rates in the Poorest Income Quintile and All Other Quintiles by Maternal Age Group, 2007 and 2013 6.0 5.0 4.0 Ratio 3.0 2.0 1.0 - Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 Q1/Q2 Q1/Q3 Q1/Q4 Q1/Q5 2007 2013 2007 2013 2007 2013 2007 2013 2007 2013 10-14 15-17 18-19 15-19 10-19 Quintile ratios, Survey year and Age of reporting first child Source: Authors’ elaboration based on data from DHS 1996, 2007 and 2013. II.2 Adolescent Pregnancy Rates in International Perspective 10. The DR’s has the highest adolescent pregnancy rate outside of SSA. In 2017, the DR’s late adolescent fertility rate was 94.3 births per 1,000 women ages 15 to 19. This was the highest rate in LAC, the third highest rate among upper-middle-income countries, and the 26th highest rate worldwide. All 25 countries with higher rates were in SSA (Figure 9). Figure 9. Late Adolescent Fertility Rates by Region and Income Group, 2017 Source: Authors’ elaboration based on World Bank data. Note: The late adolescent fertility rate is defined as the number of births per 1,000 women ages 15 to 19. 7 11. The DR’s adolescent pregnancy rate is diverging from the regional average , and if its trajectory remains unchanged, it will again surpass the SSA average. The DR managed to reduce its adolescent fertility rate by an average of 1.9 percentage points per year between 1960 and 1982, but the rate fluctuated during the 1990s and early 2000s, then settled into a more modest decline of 1 percentage point per year between 2008 and 2017. The DR is failing to keep pace with the average annual decline in the adolescent pregnancy rate for the LAC region, and in the next few years it is projected to converge with and then exceed the average for SSA (Figure 10). Figure 10. Late Adolescent Pregnancy Rates in the Dominican Republic versus Global and Regional Averages, 1960-2017 180.0 167.1 Fertility rate women aged 15-19 160.0 140.0 154.0 120.0 112.9 102.8 100.0 106.3 108.1 80.0 94.3 86.0 60.0 63.0 40.0 42.5 20.0 - 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 World Sub-Saharian Africa Latinoamérica and Caribe Dominican Republic Source: Authors’ elaboration based on World Bank data. Note: The adolescent pregnancy rate is defined as the number of births per 1,000 women ages 15 to 19. 12. Reducing the incidence of adolescent pregnancy will be vital to improve the DR’s performance on critical social indicators, many of which are significantly lower than what the country’s income level would predict. The DR has a Human Capital Index (HCI) score of 0.49, lower than the average of upper- middle-income countries (0.57) but higher than the expected value for a country with its rate of late adolescent pregnancy. Further improvements in the country’s HCI score will almost certainly require reducing the adolescent pregnancy rate in conjunction with measures to improve the quality and coverage of education, health, and social protection. (Figure 11). 8 Figure 11. Human Capital Index Scores and Late Adolescent Pregnancy Rates in High-Income and Upper-Middle-Income Countries, 2017 1.00 0.90 PRT USA 0.80 ESP GBR NZL Human capital index TTO CHL ARG 0.70 AZE BRA SYC PRY NAM JOR JAM URY 0.60 DZA BRB GTM GUY PAN DOM, 94.3 BWA ZAF VEN MUS CRI SUR IRQ 0.50 FJI CUB MEX SGP ECU 0.40 PER COL 0.30 0.20 0.10 0.00 0.0 20.0 40.0 60.0 80.0 100.0 Fertility rate among women ages 15-19 High income Upper middle income Linear (Upper middle income) Source: Authors’ elaboration based on World Bank data. 13. Rates of adolescent pregnancy are closely correlated with maternal and infant mortality, as well as other negative health outcomes. At 95.0 deaths per thousand live births, the DR’s maternal mortality rate is far above the average for upper-middle-income countries (65.3 deaths per thousand live births). While adolescent pregnancy is only one of many factors that contribute to maternal mortality, successful efforts to reduce adolescent pregnancy will be vital to lower the DR’s maternal mortality rate (Figure 12). The DR’s neonatal mortality rate is extremely high for a country of its income level at 20.0 deaths per thousand live births, almost double the average for upper-middle-income countries (10.5 deaths per thousand live births), and the mortality rate for children under one year is 24.8 deaths per thousand live births, about 1.5 times the average of upper-middle-income countries (16.2 deaths per thousand live births) (Figure 13 and Figure 14). The DR’s neonatal and infant mortality rates are even higher than its late adolescent pregnancy rate would predict, underscoring the urgent need to address other urgent health- related challenges in that extend beyond adolescent pregnancy. 9 Figure 12. Maternal Mortality Rates and Late Adolescent Pregnancy Rates in High-Income and Upper- Middle-Income Countries, 2017 250.0 Maternal mortality rate NAM 200.0 GUY BWA 150.0 SUR PRY ZAF VEN DZA GTM PER COL IRQ 100.0 JAM MUS BRA ECU JOR TTO DOM, 94.3, 95.0 CUB FJI MEX 50.0 CRI AZE SYC PAN ARG BRB USA CHL URY PRT GBR NZL SGP ESP 0.0 0.0 20.0 40.0 60.0 80.0 100.0 Fertility rate among women ages 15-19 High income Upper middle income Linear (Upper middle income) Source: Authors’ elaboration based on World Bank data. Note: Maternal mortality rate (model estimated) per 100,000 live births. Figure 13. Neonatal Mortality Rates and Late Adolescent Pregnancy Rates in High-Income and Upper- Middle-Income Countries, 2017 30.0 25.0 BWA Neonatal mortality rate 20.0 DOM, 94.3, 20.0 GUY NAM IRQ 15.0 DZA VEN TTO GTM AZE FJI JAM ZAF PRY 10.0 JOR SUR MUS BRASYC PAN BRB PER MEX COL ECU CRI ARG 5.0 CHL URY USA NZL PRT GBR ESP CUB SGP 0.0 0.0 20.0 40.0 60.0 80.0 100.0 Fertility rate women age 15-19 High income Upper middle income Linear (Upper middle income) Linear (Upper middle income) Source: Authors’ elaboration based on World Bank data. Note: Neonatal mortality rate per 1,000 live births. 10 Figure 14. Infant Mortality Rates and Late Adolescent Pregnancy Rate in High-Income and Upper- Middle-Income Countries, 2017 40.0 Infant mortality rate (less 30.0 BWA NAM ZAF than one year) GUY IRQ GTM DOM, 94.3, 24.8 FJI VEN 20.0 DZA AZE TTO SUR PRY JOR BRA PAN MUS JAM SYC COL ECU BRB PER MEX 10.0 ARG CRI URY USA CHL NZL CUB PRT GBR SGP ESP 0.0 0.0 20.0 40.0 60.0 80.0 100.0 Fertility rate women age 15-19 High income Upper middle income Source: Authors’ elaboration based on World Bank data. Note: The infant mortality rate is the number of deaths of children under one year of age per 1,000 live births. II.3 Factors Associated with Adolescent Pregnancy 14. In the DR, the link between the growth rate of income per capita and the adolescent pregnancy rate weakened in the early 1980s.9 Various structural, economic, and cultural factors have slowed the country’s progress in reducing adolescent pregnancy in recent decades. This section presents stylized facts for the key socioeconomic and behavioral variables associated with adolescent pregnancy, followed by a series of estimations for econometric models designed to determine the statistical significance of risk factors associated with adolescent pregnancy. Due to data limitation, this analysis is performed solely for late adolescent pregnancy using the National Multi-Purpose Household Survey (Encuesta Nacional de Hogares de Propósitos Múltiples, ENHOGAR) for 2018. Figure 15. The Late Adolescent Pregnancy Rate and Real GDP per Capita in the Dominican Republic, 1960-2017 200 $250,000 Births per 1,000 women aged $200,000 DOP (2010=100) 150 $150,000 100 15-19 $100,000 50 $50,000 0 $- 1990 1995 1960 1965 1970 1975 1980 1985 2000 2005 2010 2015 Adolescent fertility rate GNI per capita Source: Authors’ elaboration based on World Bank data. Note: The late adolescent pregnancy rate is defined as the number of births per 1,000 women ages 15 to 19. 9 https://www.bancomundial.org/es/country/dominicanrepublic/overview 11 15. In 2018, 19.1 percent of women between the ages of 15 and 19 reported having been pregnant, and 4.3 percent of teenage girls had more than one pregnancy before the age of 20. Among women ages 15 to 19, 13.8 percent were already mothers, 3.4 percent were pregnant at the time of the survey, and 1.9 percent had lost or aborted a pregnancy or their baby was stillborn. Research shows that women who become pregnant at an early age are more likely to have pregnancies very close to each other, and are likely to suffer severe socioeconomic and emotional consequences, including lost educational opportunities, lower lifetime earnings, and higher rates of drug use.10 16. Late adolescent pregnancy rates are highest in rural areas and regions along the Haitian border. The rural adolescent pregnancy rate is 23.7 percent, well above the urban rate of 18.1 percent. The highest adolescent pregnancy rates in the DR are registered in the border regions of Enriquillo (27.7 percent), Cibao Noroeste (25.6 percent) and El Valle (23.1 percent). In all regions except Cibao Noroeste and Cibao Norte, late adolescent pregnancy rates are higher in rural areas than in urban centers. The rural/urban disparity in adolescent pregnancy rates is widest in Higuamo, at 30.8 percent versus 17.6 percent (Figure 14 and Table 1). 10 Freites et al. (2018); and Charles et al. (2016) 12 Table 1. Late Adolescent Pregnancy Rates by Socioeconomic Characteristics and Location, 2018 Total Rural Urban Once pregnant 19.1 23.7 18.1 Mother 13.8 18.4 12.8 Currently pregnant, non-mother 3.4 3.6 3.3 Pregnancy loss, stillborn baby, currently not pregnant 1.9 1.6 2.0 or mother Number of pregnancies 1 pregnancy 14.5 16.5 14.1 2 pregnancies 3.3 5.1 2.9 >2 pregnancies 1.1 1.6 0.9 No answer 0.2 0.5 0.2 Development region Central Cibao 18.1 22.7 16.0 Northwest Cibao 25.6 23.8 26.2 Northern Cibao 19.8 19.5 19.9 Cibao Sur 20.2 24.5 17.7 Del Yuma 20.3 28.1 19.3 The Valley 23.1 26.4 21.6 Enriquillo 27.7 33.2 26.7 Higuamo 20.4 30.8 17.6 Metropolitan 16.1 21.0 15.7 Valdesia 20.4 23.3 19.2 Quality of Life Index (ICV) ICV1 41.4 47.0 38.5 ICV2 31.3 28.5 32.2 ICV3 16.5 18.7 16.1 ICV4 7.1 11.4 6.7 Place of birth Dominican Republic 18.7 23.1 17.7 Haiti 39.8 47.9 37.7 Other 8.1 7.5 8.2 Marriage Union Once married or united 71.3 70 71.7 Never married or united 3.2 3.2 3.2 Desire to get pregnant I wanted to get pregnant 36.1 30.9 37.5 You'd want to get pregnant later 62.0 65.7 61.0 I didn't want to have any more children. 1.9 3.4 1.5 Source: Authors’ elaboration based on ENHOGAR-2018 data. 13 Figure 16. Share of Adolescents Ages 15-19 Who Report Having Been Pregnant by Region, 2018 Source: Authors’ elaboration based on ENHOGAR-2018 data. 17. Adolescent pregnancy rates are closely correlated with multidimensional poverty indicators. The share of women ages 15 to 19 who reported having been pregnant was 41.4 percent for households living in extreme poverty, more than five times higher than the share for non-poor households and more than two times higher than the share for vulnerable households.11 While Haitian-born adolescents are 2- 5 times more likely to become pregnant than adolescents born in the DR or in other countries, this correlation is driven by differences in poverty status: after controlling for household-level poverty, the pregnancy rate among Haitian-born adolescents is lower than it is for their non-Haitian-born peers (Table 1 and Box 1). 18. Child marriage and early sexual unions greatly increase the likelihood of adolescent pregnancy, and such relationships are most common among adolescents from poor households. While 71.3 percent of adolescents who reported having been married or in a sexual union also reported having become pregnant at least once, this rate falls to 3.2 percent among those who reported never having been in such a relationship (Table 1). Overall, 23.8 percent of girls ages 15-19 reported being, or having been, married or in a sexual union. Of these, 23.5 percent reported that the relationship began when they were younger than 15. The early adolescent pregnancy rate is highest among the poorest households, while the late adolescent pregnancy rate is highest among households classified as moderately poor or vulnerable (Figure 17). A joint 2017 study by the World Bank and UNICEF found that, in the DR, early sexual unions are associated not only with higher rates of adolescent pregnancy, but also lower completion rates for secondary education, higher rates of exposure to physical, sexual, and verbal violence, and lower rates of birth registration for children. 19. About two-thirds of late adolescents who became pregnant did not want to become pregnant at the time. Only 36 percent of those who experienced adolescent pregnancies reported wanting to become pregnant, while 62 percent wanted to postpone pregnancy, and 1.9 percent did not want to have 11 The DR’s Unique Beneficiary System (Sistema Único de Beneficiarios, SIUBEN) assigns households to one of four socioeconomic groups based on a multidimensional Quality of Life Index (Indice de Calidad de Vida, ICV). Households in extreme poverty are classified as ICV-1; moderately poor households are classified as ICV-2; non-poor but vulnerable households are classified as ICV-3; and economically secure households are classified as ICV-4. 14 any more children (Table 1). The share of unwanted pregnancies among late adolescents was higher in rural areas (69.3 percent) than in urban centers (62.5 percent). The very high rate of unplanned or unwanted pregnancies among late adolescents underscores the need for more effective reproductive- health, family-planning, and social-protection policies. 20. Although abortion is prohibited by law in the DR, 9.8 percent of adolescent girls who have had sexual relations reported having lost a pregnancy or having had an abortion. The incidence of lost or aborted pregnancies is directly correlated with maternal age and inversely correlated with socioeconomic status. Women ages 18 and 19 are more likely to report both having become pregnant at least once (32 percent) and having had an abortion or miscarriage (11.3 percent), whereas girls ages 15 to 17 are less likely to report having been pregnant (11.2 percent) or having had an abortion or miscarriage (7.8 percent). Among adolescents who reported having had sexual intercourse, lost or aborted pregnancies are more frequent in rural areas (8.6 percent) than in urban centers (7.7 percent). The incidence of miscarriage also rises among adolescents from poorer households. Only among young women ages 18 and 19 is the rate of lost or aborted pregnancies in urban areas (11.5 percent) higher than the rate in rural areas (10.2 percent), while the inverse correlation with socioeconomic status is consistent across all groups (Table 2). Table 2. Incidence of Lost or Aborted Pregnancies among Late Adolescents by Rural/Urban Area and Socioeconomic Status, 2018 (%) % Have had % Have had sexual Total sexual intercourse intercourse 15-17 18-19 Have ever lost a pregnancy 4.0 9.8 7.8 11.3 Number of miscarriages: 1 3.5 8.7 7.2 9.7 2 0.4 1.0 0.5 1.4 >2 0.1 0.2 0.2 0.1 Area of residence Rural 4.2 9.5 8.6 10.2 Urban 3.9 9.9 7.7 11.5 Quality of life index ICV1 7.3 11.3 10.2 12.0 ICV2 6.9 12.6 10.0 14.8 ICV3 3.0 7.5 6.7 8.1 ICV4 2.2 9.1 3.4 11.8 Source: Author’s elaboration based on ENHOGAR-2018 data. Note: Adolescents who did not respond if they have had sexual intercourse and/or lost a pregnancy were excluded. 15 Adolescent Pregnancy, Education, and Job Placement 21. The incidence of adolescent pregnancy is strongly linked with school absenteeism and dropout rates. Among adolescent girls who reported never having been pregnant, 83.2 percent are currently in school or college, compared to just 45.1 percent of those who reported having been pregnant (Figure 17). A full 33.2 percent of teenagers who have been pregnant did not attend school or university during their pregnancy with their first child. Among adolescents who have been pregnant, 78 percent cited pregnancy as the main reason that they did not continue their studies while pregnant with their first child. Though barring attendance by pregnant students is contrary to both the regulations of the Ministry of Education and Dominican law, another 4.5 percent of adolescents who reported having been pregnant stated that they did not continue their studies because the school did not allow it (Figure 18). Moreover, 48.7 percent of pregnant adolescents stopped attending school or university after the birth of their first child. Among adolescents who stopped attending school or university after the birth of their first child, 35.1 percent stated that a lack of childcare options was the main factor in their decision, and inadequate childcare was the most commonly cited reason for not continuing school or university across all socioeconomic groups (Figure 19). Figure 17. Share of Late Adolescents Who Report Having Been Pregnant by Current Educational and Employment Status, 2018 76.8 % of women aged 15-19 44.5 36.5 8.6 6.4 10.3 4.4 12.5 Studying and Studying and not Not studying and Not working and not working working working studying Ever pregnant Never pregnant Source: Author’s elaboration based on ENHOGAR-2018 data. Figure 18. Reasons for Leaving School or University among Late Adolescents Who Left during Their First Pregnancy, 2018 4.5% 7.8% 4.8% 5.0% 78.0% Due to pregnancy Felt ashamed School didn't accept her Moved to another house Other Source: Author’s elaboration based on ENHOGAR-2018 data. 16 Figure 19. Reasons for Leaving School or University among Late Adolescents Who Left after the Birth of Their First Child by Household Socioeconomic Status, 2018 Health issues Had no one to leave the child with It's her first pregnancy Didn't want to be separated from her child % of women aged 15-19 Had to work Completed studies Other 39.2 38.1 36.9 35.6 35.1 32.5 28.6 26.4 20.5 19.0 18.7 18.4 17.7 16.3 13.9 12.8 9.7 7.1 7.0 6.5 6.3 6.3 6.3 6.0 5.6 4.5 4.5 4.5 4.2 3.9 3.0 2.8 2.2 0.0 0.0 ICV1 ICV2 ICV3 ICV4 TOTAL Source: Author’s elaboration based on ENHOGAR-2018 data. 22. Since pregnancy itself constitutes a barrier to continuing education, adolescent girls who reported having been pregnant also tended to experience worse educational outcomes. Among young women aged 15-19 who reported having been pregnant, 69.5 percent completed fewer years of study than would be expected for their age, versus just 30 percent of adolescent girls who have never been pregnant. Educational attainment gaps among adolescents who have been pregnant are most common in households at the highest socioeconomic level (Figure 20). Figure 20. Share of Late Adolescents with Fewer Years of Schooling than Would Be Expected for Their Age by Pregnancy History and Socioeconomic Status, 2018 Incidence of women aged 96.4 74.1 67.0 62.3 15-19 (%) 47.7 47.5 27.3 17.2 ICV1 ICV2 ICV3 ICV4 Never pregnant Ever pregnant Source: Author’s elaboration based on ENHOGAR-2018 data. Note: The educational deficit is the difference between the expected number of years of schooling for a given age and the actual number of years of education completed by a given student. 23. Adolescent pregnancy is associated with lower rates of labor participation and school attendance. 44.5 percent of adolescents who reported having been pregnant were neither employed nor attend school or university, compared to just 12.5 of their adolescents who reported never having been pregnant. Among adolescents who are neither employed nor enrolled in education, those who reported having been pregnant tended to have education levels below the group average (Table 3). 17 Table 3. Educational Attainment among Late Adolescents by Pregnancy History and Education Enrollment or Employment Status, 2018 15-17 18-19 15-19 Level of education Pregnant Never Pregnant Never Pregnant Never pregnant pregnant pregnant Engaged in the Labor Force or Enrolled in School or University None 0.2 0.1 Incompleted primary 14.0 1.1 10.2 1.2 11.6 1.1 Completed primary 8.7 0.8 5.9 0.8 6.9 0.8 Incompleted high school 12.6 1.0 12.0 2.2 12.2 1.4 Completed high school 4.6 4.7 16.3 16.6 12.0 8.5 Attending higher education 0.0 0.0 1.2 0.6 0.7 0.2 No answer 1.0 0.4 1.1 0.2 1.1 0.3 Subtotal 40.9 8.2 46.7 21.6 44.5 12.4 Engaged in the Labor Force or Enrolled in School or University Incompleted primary 10.4 6.8 5.6 1.8 7.4 5.2 Completed primary 10.0 8.2 6.4 2.5 7.7 6.4 Incompleted high school 36.7 68.4 24.8 21.0 29.1 53.3 Completed high school 0.9 6.2 11.4 28.4 7.6 13.3 Attending higher education 0.2 1.7 4.7 24.7 3.0 9.0 No answer 0.8 0.4 0.5 0.1 0.6 0.3 Subtotal 59.0 91.7 53.4 78.5 55.4 87.5 Total None 0.0 0.2 0.0 0.0 0.0 0.1 Incompleted primary 24.4 7.9 15.8 3.0 19.0 6.3 Completed primary 18.7 9.0 12.3 3.3 14.6 7.2 Incompleted high school 49.3 69.4 36.8 23.2 41.3 54.7 Completed high school 5.5 10.9 27.7 45.0 19.6 21.8 Attending higher education 0.2 1.7 5.9 25.3 3.7 9.2 No answer 1.8 0.8 1.6 0.3 1.7 0.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source: Author’s elaboration based on ENHOGAR-2018 data. Sexual Activity, Marriage, and Perceptions of Adolescent Pregnancy 24. In the most vulnerable households, adolescent girls tend to start having sex at an earlier age. Among adolescents who have never become pregnant and who live in extreme poverty, 37 percent reported have participated in a sexual act. This share falls as household socioeconomic status rises, reaching 18.4 percent among adolescents from households at the highest socioeconomic level (Figure 21). Among those who reported never having had sexual intercourse, 75.5 percent of late adolescents stated that they did not feel ready for sexual activity. Other important reasons for delaying sexual activity included waiting for marriage (15.1 percent) and fear of getting pregnant (8.8 percent). Meanwhile, among late adolescents who reported having been pregnant, 41 percent had their first sexual experience before the age of 15. This share is highest among adolescents from households in extreme poverty (49.7 percent) and decreases at higher socioeconomic levels (Figure 22). 18 Figure 21. Age of First Sexual Intercourse among Late Adolescents by Socioeconomic Status and Pregnancy History, 2018 3.7 2.8 3.4 2.9 9.5 55.2 47.0 55.4 62.3 57.5 65.8 72.1 52.0 72.7 80.0 5.5 2.7 3.4 49.7 18.7 3.4 41.0 39.1 22.2 41.7 38.6 3.5 17.0 18.2 12.7 9.2 11.1 6.3 6.0 3.7 Ever Never Ever Never Ever Never Ever Never Ever Never pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant Total ICV1 ICV2 ICV3 ICV4 <15 15-17 18-19 Never have had sexual relation NA Source: Author’s elaboration based on ENHOGAR-2018 data. Figure 22. Reasons for Postponing Sexual Intercourse among Late Adolescents Who Have Never Had Sexual Intercourse, 2018 Does not feel ready 75.5 % of women aged 15-19 Wait until marriage 15.1 Sex before marriage is not correct 12.5 Fear of getting pregnant 8.8 Wait until to be engaged for marriage 4.9 Wait for someone who loves her 4.4 Has not had a chance 2.5 Fear of contracting AIDS or another STD 2.2 Other 15.2 Source: Author’s elaboration based on ENHOGAR-2018 data. Note: Adolescents who did not indicate whether they had had sexual intercourse were excluded. 25. Prioritizing education and avoiding exposure to verbal violence are the main reasons why adolescents do not want to be married or involved in a sexual union before reaching the age of majority, while wanting to start a family is the most frequent justification for early marriage or sexual union. The most common negative consequences of early marriage or sexual union cited by late adolescents were: inability to finish school (52.6 percent), experiencing verbal violence (26.7 percent) and having a premature pregnancy (19.1 percent) (Figure 23). The most common positive associations with early marriage or sexual union among late adolescents were: starting a family (20.2 percent), having children early in life (7.7 percent), and increasing their personal independence (6.3 percent) (Figure 24). 19 Figure 23. Negative Consequences of Early Marriage or Sexual Union Cited by Late Adolescents, 2018 Not being able to finish their studies 52.6 Living verbal violence 26.7 Early pregnancy 19.1 % of women aged 15-19 Health risk 16.8 Will not enjoy your youth 16.7 Emotional violence 15.3 Physical violence 14.0 Fewer future job opportunities 12.9 Poverty 12.1 Being abandoned after being used or made pregnant 10.1 Does not know 4.6 Other 53.7 Source: Author’s elaboration based on ENHOGAR-2018 data. Figure 24. Positive Aspects of Early Marriage or Sexual Union Cited by Late Adolescents, 2018 Have your own family 20.2 % of women aged 15-19 Does not know 18.7 Have children early 7.7 Make your own life 6.3 Be more responsible 6.1 Reach maturity 6.0 Have more freedom 4.8 Be happy 4.1 Other 48.7 Source: Author’s elaboration based on ENHOGAR-2018 data. Risk Factors for Adolescent Pregnancy 26. Econometric models can identify the main factors that affect the probability of late adolescent pregnancy. The models developed for this report control for location, household characteristics, personal life, and educational and labor status. The dependent variable measures the occurrence of pregnancy in adolescence and is equal to one if the adolescent has ever been pregnant and zero otherwise. The variables associated with location include: area (urban = 1, rural = 0); adolescent pregnancy rate at the municipal level; and regional development status (Ozama = 0). The variables linked to household characteristics include: socioeconomic stratum;12 whether the adolescent reports living with her father (yes = 1, no = 0); the number of household members; and whether her place of birth was in Haiti (yes = 1, no = 0). The variables related to the adolescent's personal life include: age; whether she has ever been married or in a sexual union (yes = 1, no = 0); the number of years since her first sexual encounter; and whether she reports that having a child during adolescence is “good” (yes = 1, no = 0). Finally, two subsets of variables related to educational and labor status were tested in separate models. The first group of variables includes years of educational deficit and whether the adolescent is neither currently employed 12 Socioeconomic strata are measured according to the SIUBEN ICV system. See supra note 12. 20 nor enrolled in school or university (yes = 1, no = 0). The second group of variables includes: whether the adolescent has completed primary education (yes = 1, no = 0); whether she has completed secondary education (yes = 1, no = 0); whether she is enrolled in higher education (yes = 1, no = 0); and whether she is or has been enrolled in technical training in the last 12 months (yes = 1, no = 0) (Box 1). The models were estimated using weighted least squares (WLS) with robust standard errors, the expansion factor for women ages 15-19, and data from ENHOGAR 2018 as a weight. 27. Adolescents who live in urban areas and in municipalities with high rates of adolescent pregnancy, as well as those who live in the Higuamo and Valdesia regions, are at especially high risk of becoming pregnant. Although descriptive statistics show a higher incidence of adolescent pregnancy in rural areas, after controlling for variables associated with personal life and educational and labor status, urban adolescents face a higher probability of becoming pregnant. This result implies that efforts to reduce adolescent pregnancy should be targeted at the regional and community levels as well as according to personal and household characteristics. 28. Household poverty does not, per se, increase the likelihood of adolescent pregnancy, but it is closely correlated with other explanatory variables such as educational and labor status and level of educational attainment. Indeed, after controlling for the adolescent neither being employed nor enrolled in education and for years of educational deficit, the probability of becoming pregnant may be lower among adolescents from poorer households. Moreover, higher levels of educational attainment are associated with a lower probability of adolescent pregnancy after controlling for other factors. The impact of educational and labor variables underscores the extent to which disengagement from the education system and the labor market increases the likelihood of adolescent pregnancy. 29. The presence of the father in the adolescent's household is associated with a lower probability of adolescent pregnancy, while a greater number of household members is associated with a higher probability. The presence of the father in the home may contribute to lower rates of adolescent pregnancy by offering a traditional family model for the adolescent to follow or by establishing an authority relationship that requires the adolescent to comply with certain rules of conduct. Conversely, the fact that a greater the number of household members is associated with a higher probability of teenage pregnancy could be due to less parental attention being available to each household member or to competition for limited space and resources, which incentivizes early marriage or sexual union as a means to escape an overcrowded household. 30. Early marriage or sexual union, the number of years since the first sexual experience, and positive perceptions of adolescent pregnancy are all positively associated with the likelihood of adolescent pregnancy. Among the variables considered, involvement in an early marriage or sexual union is most strongly linked with the probability of adolescent pregnancy. The number of years since the first sexual experience is correlated with both early marriage and sexual unions and with the likelihood of adolescent pregnancy. Meanwhile, variables linked to educational attainment are most negatively associated with adolescent pregnancy after controlling for other factors. Finally, after controlling for household poverty, the probability of adolescent pregnancy is lower among girls born in Haiti (Box 1). 21 Box 1. Risk Factors Associated with Adolescent Pregnancy To assess the significance of key risk factors in the likelihood of late adolescent pregnancy, the following weighted probabilistic linear models were estimated taking into consideration variables associated with four groups of factors: place of residence, household characteristics, personal life, and educational and labor status. Table 4. Probability of Late Adolescent Pregnancy (1) (2) (3) (4) (5) Localidad Localidad Localidad Hogar Hogar VARIABLES Localidad Localidad y Hogar Hogar y Comportamiento y Comportamiento y Nivel Comportamiento Educativo-Laboral Educativo Urbana -0.0439*** -0.0147*** 0.00917*** 0.0144*** 0.00905*** (0.00164) (0.00158) (0.00109) (0.00107) (0.00109) Porcentaje adolescentes embarazadas en el municipio 0.996*** 0.802*** 0.362*** 0.338*** 0.366*** (0.00880) (0.00824) (0.00571) (0.00567) (0.00570) ICV1 0.257*** 0.0197*** -0.0144*** 0.0122*** (0.00340) (0.00250) (0.00253) (0.00251) ICV2 0.200*** 0.0290*** 0.0178*** 0.0229*** (0.00151) (0.00110) (0.00109) (0.00111) ICV3 0.0719*** 0.00358*** -0.00348*** 0.000999 (0.00113) (0.000818) (0.000812) (0.000827) Padre presente -0.167*** -0.0264*** -0.0253*** -0.0258*** (0.00105) (0.000696) (0.000687) (0.000701) Total miembros en el hogar -0.0032*** 0.0128*** 0.0130*** 0.0126*** (0.000378) (0.000254) (0.000254) (0.000254) Nacida en Haití 0.0589*** -0.0179*** -0.0708*** -0.0212*** (0.00434) (0.00318) (0.00320) (0.00324) Edad 0.00719*** 0.00226*** 0.0147*** (0.000284) (0.000284) (0.000399) Alguna vez unida 0.482*** 0.466*** 0.477*** (0.00208) (0.00211) (0.00209) Tiempo exposición 0.0522*** 0.0489*** 0.0509*** (0.000426) (0.000421) (0.000426) Percepción positiva de la maternidad adolescente 0.138*** 0.130*** 0.136*** (0.00198) (0.00199) (0.00198) NiNi 0.0968*** 22 (0.00127) Déficit educativo 0.0101*** (0.000332) Primaria completa 0.00177 (0.00178) Secundaria completa -0.0298*** (0.00211) Inscrita en educación superior -0.0387*** (0.00238) Cursa o completó técnico vocacional los últimos 12 meses -0.0145*** (0.00103) Regiones de desarrollo Cibao Nordeste -0.0108*** -0.0113*** -0.0219*** -0.0102*** -0.0221*** (0.00248) (0.00242) (0.00174) (0.00168) (0.00173) Cibao Noroeste -0.00785** -0.0518*** -0.0400*** -0.0321*** -0.0419*** (0.00323) (0.00307) (0.00212) (0.00210) (0.00212) Cibao Norte -0.00719** -0.0162*** -0.0148*** -0.0126*** -0.0156*** (0.00179) (0.00169) (0.00120) (0.00119) (0.00121) Cibao Sur -0.0130*** 0.00133 -0.0182*** -0.0110*** -0.0210*** (0.00245) (0.00234) (0.00163) (0.00163) (0.00163) El Valle -0.0114*** -0.0445*** -0.0143*** -0.0116*** -0.0155*** (0.00381) (0.00372) (0.00220) (0.00215) (0.00220) Enriquillo -0.00313 -0.0367*** -0.0179*** -0.0171*** -0.0209*** (0.00336) (0.00316) (0.00197) (0.00202) (0.00197) Higuamo -0.00656*** -0.0400*** 0.0102*** 0.00989*** 0.00677*** (0.00247) (0.00242) (0.00151) (0.00149) (0.00151) Valdesia -0.00987*** -0.00508*** 0.00591*** 0.00621*** 0.00380*** (0.00193) (0.00184) (0.00128) (0.00125) (0.00129) Yuma -0.00184 -0.00828*** -0.00815*** -0.0108*** -0.0115*** (0.00253) (0.00242) (0.00168) (0.00169) (0.00169) Constante 0.0419*** 0.0443*** -0.250*** -0.179*** -0.360*** (0.00222) (0.00270) (0.00515) (0.00513) (0.00671) Observaciones 471,903 471,903 464,089 460,501 464,089 R-cuadrado 0.038 0.135 0.599 0.610 0.600 Source: Authors’ estimates based on ENHOGAR.2018. 23 Note: Weighted Least Square Models (WLS), robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1, Primary education completion refers to eighth grade completion. Exposure refers to age minus age at first sex. Dependent variable equal to 1 if the adolescent has been pregnant and zero if otherwise. 24 II.4 Effects of Adolescent Pregnancy 31. Various studies have found that pregnant adolescents are especially likely to experience health problems before, during, and after delivery, and adolescent pregnancy is strongly associated with negative social and economic outcomes. Adolescent pregnancy has associated risks of low birth weight infants, neonatal death, maternal mortality, medical complication associated to anemia, sexually transmitted diseases, post-partum hemorrhage, mental disorder (mainly depression)13, poor maternal weight gain, prematurity (births at less than 37 weeks) and pregnancy-induced hypertension14, 15, 16, as well as domestic violence and trauma17, Moreover, an early marriage or sexual union significantly diminishes the probability that a girl will complete school, permanently reducing her income potential.18 Adolescent pregnancy is also linked with lower rates of labor-force participation, lower wages, less educational attainment, a higher total number of children, and an inability to escape poverty. To examine how the dynamic associated to family size, educational attainment, labor force participation and public aid dependency operate in the DR, a probabilistic linear model of the effects of late adolescent pregnancy on key socioeconomic indicators was estimated for all female heads of household between the ages of 20 and 30. The model focuses exclusively on female-headed households because the data enable the identification of the female head’s children within the household (Box 2). For each estimated model, the hypothesis of the equality of the regression coefficients was tested for the variables when the mother was between the ages of 10 and 14 and between 15 and 19. 32. In the DR, adolescent pregnancy is associated with a greater number of children in adulthood, and this association is statistically larger for early adolescent pregnancy than it is for late adolescent pregnancy. Female heads of household between the ages of 20 and 30 who were pregnant during their adolescence are more likely to have a greater number of children as adults, and those who became pregnant before the age of 15 were especially likely to have a large number of children. Similarly, female heads of households that are poor, urban, and/or located in regions other than Ozama and South Cibao are more likely to have more children. The probability of having a large number of children is reduced if the female head of household has a partner and/or was not pregnant during adolescence. 33. Female heads of household who were pregnant as adolescents tend to have lower levels of educational attainment than do those who had children after age 19. Moreover, the likelihood that a female head of household has completed primary or secondary school drop significantly if she became pregnant in early adolescence. Similarly, having a greater number of children, being unmarried, and living in a poor household are all correlated with lower rates of primary and secondary school completion. Conversely, female heads of households who live in urban areas are more likely to complete their education at all school levels. 34. Female heads of households who were pregnant as adolescents are more likely to participate in the workforce, be employed, and earn more labor income than those who were not pregnant as adolescents, and those differences are larger for all variables in the case of early adolescent pregnancy. 13 Wilson-Mitchell et al (2014) 14 Holness (2015). Pag. 678 15 Conde-Agudelo et al (2005). 16 Chen et al. (2007) 17 Harner (2004). Pag 313 18 Vicepresidencia de la República Dominicana (2019). “El matrimonio infantil y las uniones tempranas: Resumen del estudio de conocimientos, actitudes y prácticas en seis municipios de la República Dominicana” 25 Early childbirth appears to exert pressure on women to enter the labor force, obtain employment, and spend more time working. However, labor-force participation and employment rates decrease among women with a greater number of children and among those who are married or in a sexual union. Female heads of household who live in urban areas are also more likely to participate in the labor force and to be employed. Poverty is associated with a greater probability of participating in the labor force, but the probability of being employed is lower among female heads of household in extreme poverty. Women who became pregnant as adolescents tended to earn more labor income than those who became pregnant after age 19, but due to their lower rates of educational attainment, the greater amount of income earned by adolescent mothers likely reflects longer working hours rather than higher wages. 35. Female heads of household who became pregnant as adolescents are more likely to participate in social protection programs and subsidized insurance schemes. Reliance on social programs is positively correlated with total number of children, age, and poverty status. Female-headed households located in the Cibao Noroeste, El Valle, Enriquillo, and Higuamo regions are also more likely to be enrolled in the nutritional support program Eating is First (Comer es Primero), while female-headed households in the Higuamo region are more likely to be enrolled in the subsidized Family Health Insurance (Seguro Familiar de Salud en el Regimen Subsidiado) scheme. Participation in social protection programs is less likely among female heads of household who are cohabiting and among those living in urban areas. 26 Box 2. Correlations between Adolescent Pregnancy and the Socioeconomic Conditions of Female-Headed Households Table 5. Effect early and late adolescent pregnancy on labor-market participation, educational attainment, and reliance on social programs among female heads of household ages 20-30 (1) (2) (3) (4) (5) (6) (7) (8) (9) VARIABLES Total de Participación Ocupada Ln (Ingreso Al menos Al menos Universidad Beneficiaria Afiliada hijos laboral Mensual) primaria secundaria completa Comer es SENASA completa completa Primero Madre antes de los 15 años 0.770*** 0.288*** 0.372*** 6,248*** -0.224*** -0.235*** -0.00967*** 0.0103*** -0.0922*** (0.00756) (0.00488) (0.00515) (83.23) (0.00327) (0.00428) (0.00240) (0.00277) (0.00459) Madre entre 15-19 años 0.592*** 0.134*** 0.108*** 1,083*** -0.0163*** -0.0959*** -0.0112*** 0.000396 0.0284*** (0.00218) (0.00150) (0.00159) (25.64) (0.00101) (0.00132) (0.000740) (0.000853) (0.00141) Casada o unida 0.190*** -0.126*** -0.0645*** 1,483*** -0.0225*** -0.0454*** 0.0912*** -0.0393*** -0.0573*** (0.00251) (0.00161) (0.00170) (27.51) (0.00108) (0.00141) (0.000794) (0.000916) (0.00152) Total de hijos -0.146*** -0.0770*** -3,266*** -0.0425*** -0.0213*** -0.0920*** 0.0283*** 0.294*** (0.00382) (0.00404) (65.24) (0.00256) (0.00335) (0.00188) (0.00217) (0.00360) Total hijos al cuadrado -0.00283*** -0.0161*** 269.2*** -0.000701 -0.0137*** 0.0117*** -0.00301*** -0.0501*** (0.000868) (0.000916) (14.81) (0.000581) (0.000761) (0.000427) (0.000493) (0.000817) Edad 0.115*** 0.0331*** 0.0362*** 420.0*** -0.00116*** 0.0175*** 0.00652*** 0.0158*** 0.00677*** (0.000391) (0.000273) (0.000288) (4.655) (0.000183) (0.000239) (0.000134) (0.000155) (0.000257) Urbano 0.147*** 0.105*** 0.0690*** 1,842*** 0.0206*** 0.00869*** 0.0580*** -0.00803*** -0.0171*** (0.00303) (0.00194) (0.00205) (33.12) (0.00130) (0.00170) (0.000956) (0.00110) (0.00183) ICV1 0.525*** 0.0655*** -0.0193*** -3,584*** -0.596*** -0.824*** -0.142*** 0.0776*** 0.114*** (0.00435) (0.00285) (0.00301) (48.63) (0.00191) (0.00250) (0.00140) (0.00162) (0.00268) ICV2 0.336*** -0.00392* 0.00664*** -3,936*** -0.206*** -0.566*** -0.163*** 0.0804*** 0.138*** (0.00311) (0.00203) (0.00215) (34.71) (0.00136) (0.00178) (0.00100) (0.00116) (0.00191) ICV3 0.0950*** 0.0120*** 0.00634*** -2,968*** -0.0500*** -0.248*** -0.145*** 0.0125*** 0.110*** (0.00291) (0.00187) (0.00197) (31.87) (0.00125) (0.00164) (0.000920) (0.00106) (0.00176) Regiones de desarrollo Cibao Nordeste 0.205*** 0.00167 0.00396 -751.8*** -0.0671*** -0.220*** 0.0429*** -0.0590*** -0.0241*** (0.00462) (0.00295) (0.00311) (50.35) (0.00198) (0.00259) (0.00145) (0.00168) (0.00278) Cibao Noroeste 0.0537*** 0.0141*** 0.0456*** -1,707*** -0.191*** -0.109*** -0.0179*** 0.0299*** -0.323*** (0.00641) (0.00409) (0.00432) (69.78) (0.00274) (0.00358) (0.00201) (0.00232) (0.00385) Cibao Norte 0.0605*** 0.0458*** 0.130*** 1,211*** -0.120*** -0.196*** 0.0574*** -0.0308*** -0.191*** (0.00367) (0.00234) (0.00247) (39.94) (0.00157) (0.00205) (0.00115) (0.00133) (0.00220) Cibao Sur 0.0378*** -0.0685*** 0.00485 1,211*** -0.152*** -0.202*** 0.0813*** -0.0454*** -0.0497*** (0.00481) (0.00307) (0.00324) (52.37) (0.00206) (0.00269) (0.00151) (0.00174) (0.00289) El Valle 0.211*** 0.116*** 0.242*** 3,577*** 0.0276*** -0.0309*** 0.102*** 0.0223*** -0.159*** (0.00833) (0.00532) (0.00562) (90.82) (0.00356) (0.00467) (0.00262) (0.00302) (0.00501) Enriquillo 0.262*** 0.0237*** 0.0181*** -1,177*** -0.0724*** 0.0414*** 0.0487*** 0.130*** -0.0270*** 27 (0.00597) (0.00382) (0.00403) (65.13) (0.00256) (0.00335) (0.00188) (0.00217) (0.00359) Higuamo 0.0672*** 0.193*** 0.246*** 3,033*** 0.0694*** 0.0837*** 0.145*** 0.0560*** 0.109*** (0.00435) (0.00277) (0.00293) (47.30) (0.00186) (0.00243) (0.00136) (0.00157) (0.00261) Valdesia 0.259*** 0.0329*** 0.0155*** -1,361*** -0.109*** 0.0133*** -0.0161*** -0.0943*** -0.0713*** (0.00347) (0.00223) (0.00235) (38.00) (0.00149) (0.00195) (0.00110) (0.00126) (0.00210) Yuma -0.00155 0.0918*** 0.0665*** -293.2*** -0.142*** -0.188*** 0.0594*** -0.0527*** -0.214*** (0.00394) (0.00252) (0.00266) (42.99) (0.00169) (0.00221) (0.00124) (0.00143) (0.00237) Constante -1.952*** -0.0945*** -0.324*** -533.0*** 1.156*** 0.687*** 0.0322*** -0.380*** -0.256*** (0.0113) (0.00784) (0.00827) (133.7) (0.00525) (0.00687) (0.00386) (0.00445) (0.00738) Observaciones 449,380 449,380 449,380 449,380 449,380 449,380 449,380 449,380 449,380 R-cuadrado 0.282 0.107 0.094 0.131 0.293 0.366 0.184 0.093 0.113 Prueba: El coeficiente de “Madre antes de los 15 años” es igual al coeficiente de “ Madre entre 15-19 años” Chi-cuadrado 450.92 1128.85 3868.87 1560.14 1583.45 1112.05 2.33 10.93 992.33 Valor P 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.1273 0.0009 0.0000 Resultado prueba al 95% de Rechazada Rechazada Rechazada Rechazada Rechazada Rechazada No rechazada Rechazada Rechazada nivel de significancia Fuente: Elaboración propia con datos de la ENFCT 2018. Nota: Variable dependiente es igual a 1 si la jefa de hogar cumple con la condición indicada en el nombre de la columna y cero en caso contrario. Primaria completa se refiere a completar al menos ocho años de estudios. Modelos de Mínimos Cuadrados Ponderados (WLS) usando como ponderado el factor de expansión de las mujeres de 15-19 años, errores estándares robustos en paréntesis, *** p<0.01, ** p<0.05, * p<0.1. 28 III. Policies, Programs, and Public Spending on Adolescent Pregnancy in the DR III.1 Policies and Plans 36. Since the early 1990s, the DR has adopted a policy framework composed of laws, strategic plans, government initiatives, and international conventions designed to address adolescent pregnancy; however, policymakers have refrained from banning early marriage or expanding access to modern contraceptive methods. Article 144 of the Dominican Penal Code establishes that the minimum age for marriage is 18 for boys and 15 for girls, and exceptions may be conferred by a judge or by parental consent under the provisions of Law Nº 49-99. The Dominican Constitution mandates that the state protect children and adolescents from all forms of abuse or violence, including sexual violence, and Law Nº 136- 03 criminalizes the sexual activity between adolescents and adults with an age gap of five years or more. However, the 2014 Multiple-Indicator Cluster Survey (MICS) found that 35 percent of late adolescents who were married or in a sexual union had a husband or partner with an age gap of five to nine years, while 23.2 percent had a husband or partner with an age gap of 10 years or more.19 In addition, the Criminal Code (Act Nº 550-14, Articles 107 and 108) bans abortion except in cases when the pregnancy imperils the life of the mother, which has created incentives to engage in illegal and unsafe abortions, while the inadequate provision of healthcare services violates the right to holistic health enshrined in Article 61 of the Dominican Constitution. 37. The DR’s National Development Strategy 2030 (Estrategia Nacional de Desarrollo, END) recognizes adolescent pregnancy as a major challenge and highlights the role of education, equal rights, and expanded opportunities for youth and women in preventing adolescent pregnancy and countering the economic marginalization of pregnant adolescents. The END 2030 calls for strengthening sexual and reproductive health education; promoting responsible parenthood and family planning; establishing procedures to prevent, report, and prosecute sexual harassment, violence, and exploitation against children, adolescents, and women; and helping pregnant adolescents remain in the school system. As part of the National Early Childhood Development Policy, Comprehensive Early Childhood Care Centers and Child and Family Care Centers have been set up to provide care services for children under five, and the 2014 National Pact for Education Reform calls for establishing early childhood development centers near colleges and vocational training centers to ease the burden on students with young children. The pact also supports incorporating sexual and reproductive health into the school curriculum, and the Ministry of Education has proposed the implementation of sexual and reproductive health education at all pre- university education levels, though this policy has not yet been officially endorsed. 38. The government adopted the first Adolescent Pregnancy Prevention Plan (Plan EA)20 in 2010, but budgetary and governance challenges undermined its effectiveness. The plan was initially coordinated by the Inter-Institutional Technical Committee for the Prevention of Adolescent Pregnancy, which was composed of representatives from the Ministries of Education, Public Health, Women, and Youth, as well as the National Council for Childhood and Adolescence ( Consejo Nacional Para la Niñez y la Adolescencia, CONANI) and the National Council on HIV and AIDS (Consejo Nacional para el VIH y el SIDA, CONAVIHSIDA). However, coordination responsibilities were transferred to the Ministry of Women and then to the Social Policy Coordination Cabinet (Gabinete de Coordinación de Políticas Sociales, GCPS). 21 An appraisal of Plan EA revealed key challenges, including insufficient financial resources to implement 19 See ONE-UNICEF (2016): ENHOGAR-MICS 2014. Table CP.9. Page 220 20 https://dominicanrepublic.unfpa.org/sites/default/files/pub-pdf/PlanPrevencionCompleto.pdf. 21 National Adolescent Pregnancy Reduction Plan (PREA) 2019-2023 29 social programs and address the causes of adolescent pregnancy; a failure to provide adolescent parents with timely support when returning to school; a failure to integrate the children of adolescents into the social protection system; a failure to address adolescent populations with specific needs; inadequate integration of the male population into the plan’s operational framework; and the absence of a comprehensive monitoring and evaluation system for all initiatives implemented as part of the plan. In addition, most of the social programs implemented by the government focused on expanding sexual and reproductive health education among adolescent girls, and the plan did not adequately address the full range of socioeconomic and cultural factors that influence adolescent pregnancy.22 39. In 2019, the government adopted the National Plan for Reducing Adolescent Pregnancy (Plan de la Respuesta Nacional para la Reducción del Embarazo en Adolescente , PREA) for 2019-2023. The PREA built on the mandates established in the END 2030, the National Pact for Education Reform, and the Sustainable Development Goals (SDGs).23 The PREA’s implementation involves 27 government agencies, 10 nongovernmental organizations (NGOs), and 11 specialized UN agencies and international cooperation institutions.24 Meanwhile, the National Pact for Education Reform calls for incorporating sex education in the school curriculum from early childhood (Agreement 4.2.4), expanding sex education programs, and strengthening parental support schemes to help adolescent parents remain in the school system (Agreement 3.1.6).25 40. By 2023, the PREA targets a reduction in the early and late adolescent pregnancy rates by at least 5 percentage points below the 2014 MICS baseline.26 The plan aims to reduce the adolescent pregnancy rates on a progressive basis over the five years of its implementation (Table 6). The PREA is designed around four strategic pillars: (i) preventing adolescent pregnancy; (ii) providing adequate care in response to adolescent pregnancy; (iii) protecting the rights of children and adolescents; and (iv) promoting political advocacy on behalf of children and adolescents. Table 6. Expected Decrease in the Overall Adolescent Pregnancy Rate under the PREA, 2019-2023 Year Percentage points 2019 0.0 2020 -1.0 2021 -1.5 2022 -1.5 2023 -1.0 2019-2023 -5.0 Source: Author’s elaboration based on the PREA 2019-23. 41. Although the PREA includes more detailed activities than did the Plan EA, and it is accompanied by a monitoring and follow-up mechanism, its governance framework and budgetary arrangements have yet to be defined. Each strategic theme under the PREA 2019-2023 encompasses specific lines of action that directly address the multiple causes of adolescent pregnancy, and its implementation will be assessed via a 36-indicator monitoring and evaluation system. However, a Presidential Decree appointing 22 GCPS, 2019. 23 The SDGs linked to reducing adolescent pregnancy include No Poverty (SDG 1), Good Health and Wellbeing (SDG 3), Quality Education (SDG 4), and Gender Equality (SDG 5). 24 GCPS, 2019. 25 National Pact for Education Reform, 2014. 26 The 2014 ENHOGAR- MICS Survey Report found that 21.4 percent of women ages 20 to 24 had had at least one live birth before age 18. 30 the coordinating institution and stipulating which public agencies would form the new Inter-Institutional Commission is still being prepared. Furthermore, the Inter-Institutional Agreement should describe the role played by the various stakeholders and define their respective responsibilities for the plan’s outputs. Governance and budgetary issues undermined the effectiveness of the Plan EA, and to avoid similar challenges, the government must define the amount of resources that the PREA will allocate to each agency, including financing for monitoring and evaluation. 42. In recognition of the relevance of early unions and child marriage as a factor highly associated with adolescent pregnancy, there is a proposal for a National Plan for the Elimination of Child Marriage and Reduction of Early Unions in the Dominican Republic (MIUT Plan) 2030. The Cabinet for Coordination of Social Policy (GCPS), with the support of UNICEF, presented in July 2020 the final draft of the MIUT 2030 Plan, which aims to eradicate child marriage by 2025 and reduce at least in 5% early unions in relation to the line baseline (ENHOGAR 2014) by 2030. As a result, schooling is expected to increase and reduce adolescent pregnancy, unmet need for contraception, maternal and neonatal mortality, and gender- based violence. The MIUT Plan outlines its intervention mechanisms in coordination with the PREA 2030 Plan. Its approval is pending by the Executive Power 27. III.2 Adolescent Pregnancy Prevention Programs and Support for Adolescent Mothers 43. The first comprehensive programs oriented to provide adolescent pregnancy care and train health personnel for such purpose were established in 1993, but adolescent pregnancy prevention and sexual and reproductive health education programs did not begin to proliferate until 2012. In 1993, the Ministry of Public Health established the National Comprehensive Adolescent Health Program and the Comprehensive Prenatal Care Program for Adolescents, and 2002 and 2014 Comprehensive Adolescent Health Care Units were established in public hospitals nationwide. These initiatives focused mainly on prenatal and postpartum care and sex education for second-pregnancy planning. From 2012 onwards, the government’s approach shifted to adolescent pregnancy prevention, sexual and reproductive health education, and train-the-trainer networks supported by NGOs. Following the launch of Plan EA, new stakeholders began implementing adolescent pregnancy prevention, including the Ministry of Education (2013), CONAVIHSIDA (2013), the Progress in Solidarity (Progresando con Solidaridad, PROSOLI) program (2013), the Ministry of Women (2015), and the Ministry of Youth (2016). Between 2012 and 2018, an average of two adolescent pregnancy prevention programs were established each year. 44. During the 2012-19 period, the government implemented 21 programs designed to directly prevent adolescent pregnancy and support adolescent mothers, as well as other programs that indirectly impact adolescent pregnancy. The programs directly related to adolescent pregnancy focused on: sexual and reproductive education (6); reproductive health (6); family counseling (2); aspirational enhancement (4); protection against sexual abuse (1); and multidimensional interventions (2). However, deficiencies in the registry of beneficiaries for these interventions prevent a full assessment of their coverage. Other programs that indirectly impacted adolescent pregnancy include the Extended School Day (Jornada Escolar Extendido) and the Support for Educational Progress (Bono Escolar Estudiando Progreso) program, which provides a conditional cash transfer for eligible high school students (Table 7). 27 Final Draft Plan Nacional para la eliminación del matrimonio infantil y reducción de las uniones tempranas en la República Dominicana (Plan MIUT) 2030. 31 Table 7. Government Programs Designed to Prevent Adolescent Pregnancy and Support Adolescent Mothers, 2012-2018 Area of Start Program Title Intervention Mechanism Implementing Entity intervention Year Adolescent Pregnancy Project Health counseling on pregnancy and 2017 PROSOLI Care for childbirth for adolescent mothers adolescent Por un Comienzo Positivo Education on pregnancy, delivery, and 2013 Office of the First Lady mothers (Intelligent Communities postpartum issues for low-income Program) pregnant women, including adolescents Train the Trainer Program Sex education 2012 Adelante Jóvenes Adelante (AJA), in coordination with Ministry of Public Health Life skills-based education on Sex education 2013 Ministry of Education and sexually transmitted infections, CONAVIHSIDA including HIV/AIDS Strengthening the Integrated Sex education 2014 Ministry of Education, European Sexuality Education Strategy Union and AECID Education on Center for Integrated Sex education and TTT Network 2015 Ministry of Women sexual and Adolescent Health Promotion reproductive health Project for Strengthening Comprehensive adolescent health services 2017 Ministry of Public Health y Comprehensive Adolescent and training of trainers Ministry of Education Services, with an emphasis on Adolescent Pregnancy Prevention and Care Population and Family Training, awareness-raising, and National Population and Family Research, Planning and information outreach on sexual and Council Counseling Program reproductive health, domestic violence, and values-based society Programa Familias Fuertes: Training, awareness-raising, and 2012 Adelante Jóvenes Adelante «Amor y límites» counseling for families designed to (AJA), and Ministry of Public strengthen communication and build core Health Family family values Escuelas de Familia Education for parents and children on PROSOLI various topics, including sex education Jóvenes Progresando con Training of trainers on sex education and 2013 PROSOLI Solidaridad other topics Integrated Integrated Adolescent Prenatal Therapeutic and psycho-social support and 2014 Office of the First Lady programs Care Center (Intelligent counseling for pregnant adolescents Communities Program) Child abuse Tolerancia Cero al embarazo en Health and education workers sensitized 2014 AJA and the Ministry of Public prevention menores de 15 anos on understanding child sex abuse Health Yo decido esperar Sex education and training of trainers 2014 PROSOLI Bebé piénsalo Bien Parenting simulators and parenthood talks 2015 PROSOLI Life Tu no Ta' Pa’eso Sex education and training of trainers 2016 Ministry of Youth aspirations Club de Chicas Activities and education to prevent 2018 PROSOLI adolescent pregnancy Comprehensive Prenatal Care Prenatal and postpartum care visits and 1993 Ministry of Public Health Program for Adolescents sex education National Program of Training of healthcare providers and 1993 Ministry of Public Health Comprehensive Care for the trainers in postpartum care Sexual and Health of Adolescents reproductive (PRONAISA) health Comprehensive Adolescent Comprehensive adolescent health services 2002- Ministry of Public Health education Health Care Units 2014* Integrated adolescent health Counseling on adolescent pregnancy 2014 Ministry of Public Health approach emphasizing reducing prevention for youth and healthcare adolescent pregnancy and workers maternal mortality Source: Authors’ elaboration based on information from the PREA and PROSOLI. 32 45. No information is available on the adolescent population served by these programs, which complicates efforts to measure their coverage. Some programs have adequately identified their beneficiaries, including the Bebe, Piénsalo Bien; Yo Decido Esperar; and Club de Chicas initiatives developed under PROSOLI and the Integrated Adolescent Prenatal Care Center and Por un Comienzo Positivo programs developed by the Office of the First Lady. Collectively, these programs benefited 30,138 adolescents in 2019 (Table 8). However, others have failed to discriminate between the target adolescent population and the target adult population, including the sexual and reproductive health programs implemented by the Ministry of Public Health, the Ministry of Women, the Ministry of Youth, the National Population and Family Council, and the Family Schools (Escuelas de Familia) program under PROSOLI. Sexual and reproductive health programs targeting both adolescents and adults reached 929,007 beneficiaries in 2019. In 2018, the DR’s total population between the ages of 10 and 19 totaled 1,894,072, of whom 81,335 (4.3 percent) lived in extreme poverty, 499,939 (26.4 percent) lived in moderate poverty, and 870,783 (46.0 percent) lived in households vulnerable to poverty.28 28 These estimates are based on 2018 ENCFT. Extreme poverty, moderate poverty, and vulnerability correspond to the ICV-1, 2, and 3 classifications in the SIUBEN. 33 Table 8. Programs Designed to Prevent Adolescent Pregnancy, Support Adolescent Mothers, or Provide Sexual and Reproductive Education by Number of Beneficiaries, 2015-2019 2015 2016 2017 2018 2019 SOCIAL PROGRAMS DISTINGUISHING ADOLESCENT POPULATIONS SERVED PROSOLI Program Bebe piénsalo bien: Number of adolescents and youth with 90,774 81,170 51,220 29,745 25,004 SRH counseling Office of the First Lady Integrated Adolescent Prenatal Care Center* 176 210 1,137 1,668 Por un Comienzo Positivo (Adolescent population)* 2,241 3,466 SOCIAL PROGRAMS DISTINGUISHING ADULT POPULATION SERVED Office of the First Lady Por un Comienzo Positivo (Adult population)* 6,113 10,359 SOCIAL PROGRAMS FAILING TO DISCRIMINATE BETWEEN ADULT AND ADOLESCENT POPULATIONS SERVED PROSOLI Program Outcome: Family members with pregnancy prevention and 388,975 581,975 506,284 570,398 STIs counseling Ministry of Public Health Collective Health Program: Outcome: Population of child- 348,856 bearing age informed and empowered receiving full SRH promotion package Ministry of Women Promoción de los Derechos Integrales de la Mujer: Outcome: 7,517 Women and female youngsters and adolescents sensitized on sexual and reproductive health National Population and Family Council Population and Family Research, Planning and Counseling 2,236 Program: Outcome: Population provided with capacity building, awareness-raising, and information on pregnancy prevention and SSH counseling nationwide. Population and Family Research, Planning, and Counseling 1,401 Program. Outcome: Number of people trained in workshops Counseling for Family Life for a Values-based Society SOCIAL PROGRAMS FAILING TO DISCRIMINATE BETWEEN BENEFICIARIES, INTERVENTIONS OR BENEFICIARY AGE Ministry of Youth Comprehensive Youth Development Program: Outcome: 124,506 Young people access integrated training services, or incentives and national and international scholarship financing** Source: Authors’ estimates based on DIGEPRES and Institutional Annual Reports. (*) Intelligent Communities Components. (**) Number of beneficiaries does not make a distinction by type of intervention under the Comprehensive Youth Development Program. III.3 Public Spending on Adolescent Pregnancy Prevention Programs and Support for Adolescent Mothers 46. Public spending on programs designed to prevent adolescent pregnancy or support adolescent mothers is either negligible or cannot be distinguished from expenditures earmarked for other target populations and activities. Budgeted spending on programs explicitly targeting the adolescent population served is low,29 and the programs that fail to distinguish between adolescent and adult beneficiaries had a combined budget of less than US$5.4 million in 2019 (Table 9). Additionally, some programs with 29 For example, in 2018 the Baby Think about it (Bebé Piénsalo Bien) program had 29,745 adolescent beneficiaries and was allocated US$180 thousand, or just US$6.17 per beneficiary. 34 adolescent-centered activities do not report their expenditures on a disaggregated basis, such as the Escuelas de Familia program under PROSOLI. Table 9. Public Spending on Adolescent Pregnancy Prevention Programs, Support for Adolescent Mothers, and Sexual and Reproductive Health Education, 2018-19 2018 2019 Expenditure Expenditure Expenditure budgeted budgeted executed Reporting discriminating and identifying expenses in serving the target adolescent population Club de Chicas 31,245.0 Bebé, Piénsalo Bien 183,621.6 Yo decido esperar 90,881.4 Reporting fails to discriminate expenses by type of intervention and age of target population MSP 584,717.8 35,712.4 Ministry of Women 59,446.3 36,528.2 National Population and Family Council 47,767.2 39,109.3 Reporting fails to discriminate expenses by type of intervention and age of target population Ministry of Youth 4,737,880.5 4,138,073.8 Office of the First Lady (Intelligent Communities Program) 961,468.6 Source: Authors’ estimates based on DIGEPRES and Institutional Annual Reports . Note: Dominican Pesos exchange rate per USD: 2018: DOP 49.5151, 2019: DOP 51.3068. 47. The lack of consistent information on the expenditures and beneficiaries of adolescent pregnancy prevention programs, sexual and reproductive health education, and support for adolescent mothers prevents a meaningful analysis of their funding levels. The available data on program expenditures and beneficiaries is fragmentary. For example, PROSOLI reduced spending on adolescent pregnancy prevention programs (Figure 25), while the number of beneficiaries of programs implemented by the Office of the First Lady increased over time (Table 9). However, the absence of systematic information precludes an assessment of overall trends in spending and coverage. This critical information gap extends to the PREA, which still does not publish estimates of the total expenditures required for its implementation. While the PREA includes strategies, lines of action, and activities for each objective, the budget required to implement each activity is not defined and neither is the agency responsible for its implementation. 35 Figure 25. The Annual Budget for Adolescent Pregnancy Prevention under PROSOLI by Program, 2015- 18 Source: Authors’ estimates based on PROSOLI Annual Reports. 48. Clearly defining and disaggregating the budgetary resources and coverage levels of sexual and reproductive health education and adolescent pregnancy prevention programs is a prerequisite for evaluating the efficiency of public spending. The total cost of the Plan EA is estimated at RD$522.2 million in 2019 prices, or RD$87.0 million (about US$1.70 million) in average annual spending. Although the PREA has not been budgeted, it envisages a larger set of activities than the Plan EA and may require more resources. Meanwhile, the 2019 budget includes four initiatives that are partially or fully focused on adolescent pregnancy prevention or sexual and reproductive health education: the Ministry of Health’s Integrated Adolescent Health Approach, which emphasizes reducing adolescent pregnancy and maternal mortality; the Center for the Promotion of Comprehensive Health at the Ministry of Women; the Integrated Adolescent Prenatal Care Center founded by the Office of the First Lady; and the Strengthening Integrated Adolescent Health Project. 49. From a medium-term perspective, financing adolescent pregnancy prevention and sexual and reproductive health education programs can be more cost effective than addressing adolescent pregnancy and motherhood. UNFPA-INTEC (2013) 30 estimated that the annual cost of adolescent pregnancy and motherhood care in 2013 was RD$2.1 billion in 2013, equivalent to US$50.3 million or 0.08 percent of GDP. At that time, the annual cost of adolescent pregnancy and motherhood care was 33 times higher than the annual cost of implementing the Plan EA. This finding is consistent with the demand for reproductive health services among late adolescents. Over a 12-month period, adolescents who report having been pregnant a three times more likely to require reproductive health services than adolescents who have never been pregnant (53.3 percent versus 15.5 percent), and both groups primarily rely on hospitals or other public health facilities.31 Measures that reduce adolescent pregnancy could significantly reduce costs in the public healthcare system (Table 10). 30 UNFPA-INTEC, 2013. 31 ENHOGAR, 2018. 36 Table 10. Share of Late Adolescents Who Have Visited a Health Center in the Last 12 Months by Type of Health Service Requested, Type of Facility, and Pregnancy History, 2018 Total Have Been Pregnant Never Pregnant Total 22.68 53.25 15.48 Type of health service Gynecological consultation 56.3 65.5 48.8 Pregnancy consultation 9.6 20.8 0.5 Search for contraceptive methods 5.2 4.5 5.7 Service information 4.5 1.2 7.1 Other 24.2 7.8 37.5 No information 0.3 0.1 0.4 Total 100.0 100.0 100.0 Place of health service Hospital or other public place 69.6 78.7 62.1 Private sector clinic or center 25.9 17.3 32.9 PROFAMILIA Clinic 2.8 3.2 2.5 Another private sector or NGO site 0.3 0.3 0.3 Other 1.4 0.4 2.2 No information 0.0 0.1 0.0 Total 100.0 100.0 100.0 Source: Authors’ estimates based on ENHOGAR.2018. 37 IV. The Effectiveness of Adolescent Pregnancy Prevention and Protection Programs IV.1 Program Coverage among Late Adolescents 50. As noted above, most government programs aimed at preventing adolescent pregnancy and supporting adolescent mothers do not publish coverage data, and the regular household surveys do not include them in the list of social protection programs. Inadequate information prevents a thorough evaluation of these programs, but a review of the evaluations of causal factors that impact the likelihood of pregnancy among Dominican adolescents may shed light on their effectiveness. Moreover, the involvement of adolescents and households in certain programs can be identified from the 2018 ENHOGAR and compared against the observed probability of adolescent pregnancy. 51. Late adolescents who reported having been pregnant were less likely to have received reproductive education than those who had never been pregnant. In the 2018 ENHOGAR, 78.7 percent of late adolescents reported having attended a class at school, college, or university in which human anatomy and reproductive physiology were taught. Among late adolescents who had never been pregnant, 81.3 percent had received education in human anatomy and reproductive physiology, versus just 67.8 percent of adolescents who had been pregnant. This gap in access to reproductive education is wider in rural areas, among extremely poor households, and among adolescents with low educational levels (Table 11). Table 11. Share of Late Adolescents Who Report Having Attended a School or University Class in Which Human Anatomy and Reproductive Physiology were Taught by Socioeconomic Characteristics and Pregnancy History, 2018 Have Been Pregnant Never Pregnant Total 67.8 81.3 Area Rural 65.2 82.0 Urban 68.6 81.2 Quality-of-Life Score (SIUBEN) Extreme Poverty (ICV-1) 38.7 61.4 Moderate Poverty (ICV-2) 67.5 77.6 Vulnerable to Poverty (ICV-3) 71.4 80.7 Economically Secure (ICV-4) 85.4 87.3 Highest level of educational attainment None or preprimary only 0.0 35.3 Incomplete primary 39.3 54.2 Complete primary 63.7 74.2 Incomplete secondary 75.0 82.1 Complete secondary 84.1 87.4 Enrolled in higher education 80.2 89.7 Source: Authors’ estimates based on ENHOGAR 2018. 52. Late adolescents who reported having been pregnant were slightly less likely to use modern contraception methods and more likely to obtain contraceptives in a public hospital. About six out of 10 adolescents who reported having had sexual intercourse reported using a modern contraceptive 38 method.32 Adolescents who had never been pregnant were slightly more likely to use modern contraception, except for those living in extreme poverty (Figure 26). At least four out of 10 adolescents who reported having sexual intercourse obtained contraception from a public hospital (Figure 27). Figure 26. Share of Late Adolescents Who Have Had Sexual Intercourse Who Report Using a Modern Contraceptive Method by Socioeconomic Characteristics and Pregnancy History, 2018 Source: Authors’ estimates based on ENHOGAR 2018. Note: Modern contraceptive methods include male and female sterilization, oral and intrauterine contraceptives, and condoms. Figure 27. Sources of Contraception among Late Adolescents Who Have Had Sexual Intercourse by Socioeconomic Characteristics and Pregnancy History, 2018 Source: Authors’ estimates based on ENHOGAR 2018 53. Access to sexual and reproductive health education is critical to prevent adolescent pregnancy, and teachers, family, and friends are the main sources of reproductive information for adolescents. When asked who provided them with information about sexual and reproductive health, 69.5 percent of late adolescents reported receiving information from school teachers, while 24.9 percent reported receiving information from a relative (Table 12). Regardless of the source, adolescents who reported 32 Modern contraceptive methods include male and female sterilization, oral and intrauterine contraceptives and condoms. Traditional contraceptive methods include periodic abstinence (rhythm) and withdrawal. 39 having been pregnant were less likely to have received information on reproductive health. Table 12. Sources of Sexual and Reproductive Health Information among Late Adolescents by Pregnancy History, 2018 (%) Attended human anatomy and reproductive physiology class at school or college Total Ever pregnant Never pregnant Yes No Yes No Yes No Total 78.7 21.3 67.8 32.2 81.3 18.7 Non-school sources of information about human anatomy and reproductive physiology School teachers 69.5 6.3 56.6 5.8 72.5 6.5 Some relative 24.9 4.5 17.9 5.0 26.5 4.4 Friend 3.0 1.4 1.2 2.1 3.4 1.2 Physicians 1.0 0.4 2.9 1.4 0.5 0.1 PROFAMILIA Health Educator 1.1 0.3 1.9 1.1 0.8 0.1 Books, films and magazines 0.7 0.2 0.8 0.0 0.7 0.3 Other 1.1 0.4 0.6 0.6 1.3 0.3 N 4,946 4,946 1,059 1,059 3,887 3,887 N (expansion factor) 472,022 472,022 89,969 89,969 382,054 382,054 Source: Authors’ estimates based on ENHOGAR 2018. 54. Access to public sexual health services is especially crucial for adolescent girls from disadvantaged households. According to ENHOGAR 2018, most late adolescents with concerns about pregnancy, abortion, vaginal infections, or other aspects of sexual health and reproduction indicated that they preferred to seek assistance at a clinic or hospital. This preference was stronger among adolescents from extremely poor and moderately poor households, and it was stronger among adolescents who had already been pregnant (93.3 percent) than among those who had never been pregnant (82.4 percent). Family and friends are the second most common choice for adolescents seeking assistance with issues related to sexual health, and adolescents who have never been pregnant are more likely to express a preference for seeking help from family and friends (33.5.6 percent) than in those who have (14.4 percent). In addition, the preference for seeking help from family and friends also tends to be stronger in among wealthier households (Figure 28). 40 Figure 28. Preferred Source of Assistance with Reproductive-Health-Related Concerns by Socioeconomic Characteristics and Pregnancy History, 2018 (%) 94.6% 90.8% 94.9% 91.2% 92.9% 93.3% 85.7% 83.3% 82.4% % of women aged 15-19 77.9% 35.0% 37.7% 33.5% 26.8% 19.0% 16.4% 18.5% 14.4% 10.9% 12.6% Ever Never Ever Never Ever Never Ever Never Ever Never pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant pregnant ICV1 ICV2 ICV3 ICV4 Total Clinic or hospital Relative or friend Other Schoool teacher Health worker, PROFAMILIA educator or youth orientation center Source: Authors’ calculation based on ENHOGAR 2018. IV.2 Participation in Social Programs and Other Interventions and the Incidence of Adolescent Pregnancy 55. The complex set of circumstances and decisions that contribute to adolescent pregnancy highlights the importance of inclusive public policies that address a wide range of causal factors. However, not all programs are equally effective in reducing adolescent pregnancy. To determine which programs are having the greatest impact in the DR, the following section examines the correlation between the incidence of adolescent pregnancy and participation in programs designed to address it. These programs include reproductive health education, contraceptive supply, the Comer es Primero program, the Jornada Escolar Extendido program, and various public outreach initiatives related to adolescent pregnancy. 56. Adolescents who are currently using modern contraceptive methods, are enrolled in reproductive health education, or who report using condoms during their first act of sexual intercourse are less likely to become pregnant.33 Previous studies have found that sex education and access to contraception are critical to prevent adolescent pregnancy.34 The use of contraception may also indicate a higher degree of female empowerment within the relationship, as men are more likely to eschew condom use. While access to modern contraceptive methods is not necessarily the result of a specific public policy, programs that increase the availability or uptake of contraception are likely to reduce the rate of adolescent pregnancy. 57. According to the Berkeley Impact Evaluation (2019), young women and girls from households that participate in PROSOLI35 are more likely to use long-term temporary contraceptive methods and 33 Modern contraceptive methods include both permanent surgical procedures, such as vasectomies and tubal ligations, as well as temporary physical contraceptives, such as condoms. 34 Azevedo et al., 2012. An analysis of the contraception-focused Profamilia program in Colombia found that it did not seem to have a significant impact on total fertility, but it did raise the average age of first pregnancy. 35 PROSOLI also encompasses a subset of beneficiaries of the Comer es Primero program. 41 are less likely to become pregnant in late adolescence. The descriptive data are consistent with a slight decrease in the percentage of pregnant adolescents from households participating in PROSOLI. The share of women and girls ages 15-20 who report having been pregnant was 16.7 percent in PROSOLI households and 19.1 percent in control-group households. 58. Household participation in the Comer es Primero program also appears to decrease the odds of late adolescent pregnancy. This program is designed to provide basic food support to households in moderate or extreme poverty. To be eligible for the program, an expectant mother must attend antenatal visits, and all children in the nuclear family must have their weight and height measured regularly by a physician.36 59. Municipal-level participation in the Jornada Escolar Extendido program is associated with a lower probability of late adolescent pregnancy. This program doubles the amount of primary and secondary school hours and is being gradually phased in across municipalities. During the 2017/18 school year, Jornada Escolar Extendido covered 62 percent of public and semi-official school enrollment in the DR,37 and households in participating municipalities experienced lower rates of adolescent pregnancy. An analysis of a similar policy in Chile concluded that young women with greater access to afterschool programs at the secondary school level were less likely to become pregnant.38 60. Although the DR has launched media campaigns designed to reduce adolescent pregnancy, as well as outreach programs aimed at raising awareness among young women of the changes and responsibilities motherhood implies, these programs have been limited in scope, and not enough information is available to evaluate their effectiveness. A recent review of the international literature found that adolescent pregnancy prevention programs tend to be effective when they focus on promoting confidence, building self-esteem, enhancing negotiation skills, and fostering strong socio-emotional development, while also providing more traditional education and building cognitive skills.39 Other research has found that programs focusing on socio-emotional skills were more effective in changing behavior than those based on knowledge and information,40 and there is evidence that media campaigns can effectively influence adolescent behavior. 41 36 ADESS, 2017 37 MINERD, 2018. 38 Berthelon and Kruger, 2012. 39 Inter-American Development Bank, 2012. 40 Kirby et al., 2007. 41 Kearney and Levine, 2015. 42 Table 13. Statistical Correlations between Participation in Social Programs and Other Interventions and the Probability of Late Adolescent Pregnancy Variables Coefficient Statistical Standard robust significance error Urban 0.00684 *** (0.00108) Ever cohabited with an intimate partner 0.480 *** (0.00207) Household in extreme poverty (ICV-1) 0.0110 *** (0.00254) Household in moderate poverty (ICV-2) 0.0266 *** (0.00117) Household vulnerable to poverty (ICV-3) 0.00488 *** (0.000859) Primary education completed 0.00983 *** (0.00181) Secondary education completed -0.0181 *** (0.00214) Post-secondary education underway or completed -0.0294 *** (0.00242) Technical or vocational course taken in previous 12 months -0.0110 *** (0.00104) Having a father figure around the house -0.0271 *** (0.000695) Total household size 0.0132 *** (0.000249) Born in Haiti -0.0280 *** (0.00321) Age 0.0139 *** (0.000402) Exposure 0.0548 *** (0.000518) Adolescent pregnancy perceived as negative 0.135 *** (0.00199) Municipal adolescent pregnancy rate 0.369 *** (0.00579) Knowledge and agency Reproductive health education classes 0.0123 *** (0.000907) Used modern contraceptive methods -0.0118 *** (0.00307) Reproductive health education classes & used modern -0.0313 *** (0.00329) contraceptive methods Used condom during first sexual intercourse 0.0497 *** (0.00317) Reproductive health education classes x Used condom -0.0656 *** (0.00360) during first sexual intercourse Seek assistance for sexuality concerns Clinic or hospital 0.0238 *** (0.00254) Health worker, educator, or youth counseling center 0.0149 *** (0.00289) School teacher 0.0859 *** (0.00434) Relative or friend 0.00294 (0.00256) Social Programs Comer es Primero -0.0102 *** (0.000848) Jornada Escolar Extendido -0.0229 *** (0.00271) Region Cibao Nordeste -0.0171 *** (0.00198) Cibao Noroeste -0.0321 *** (0.00245) Cibao Norte -0.0144 *** (0.00128) Cibao Sur -0.0184 *** (0.00182) El Valle -0.0123 *** (0.00242) Enriquillo -0.0137 *** (0.00224) Higuamo 0.0102 *** (0.00170) Valdesia 0.00369 *** (0.00134) Yuma -0.0144 *** (0.00168) Constant -0.370 *** (0.00704) Observations 463,385 R-squared 0.604 Source: Authors’ estimates based on ENHOGAR 2018. 43 Note: Weighted Least Square Model (WLS), robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1, Primary education completion refers to eighth grade completion, Exposure refers to age minus age at first sex. Modern Contraceptive Methods include male and female sterilization, oral and intrauterine contraceptives, and condoms. 44 V. Conclusions and Recommendations 61. Despite decades of robust growth, the DR’s adolescent pregnancy rate has remained stubbornly high and is now well above the average for both the LAC region and upper-middle-income countries worldwide. At 94.3 births per 1,000 women aged 15-19 in 2017, the DR’s adolescent pregnancy rate is almost 50 percent higher than the LAC average (63.0), more than three times the average for upper- middle-income countries (30.7), and approaching the average for SSA (102.8). After falling steadily from the 1960s through the early 1980s, the DR’s adolescent pregnancy rate rose in the early 1990s and then declined at a slower rate than the averages for SSA or LAC. On its current trajectory, the DR’s adolescent pregnancy rate will soon surpass the SSA average. 62. While the DR has made significant progress in economic growth and poverty reduction, achieving shared prosperity and eliminating poverty will require an inclusive and concerted policy response to the persistent challenge of adolescent pregnancy. National42 and international 43 studies find that women who were pregnant during adolescence are at higher risk for maternal and child morbidity and mortality, sexual abuse, and domestic violence and that they experience worse educational, economic, and social outcomes than their peers. These consequences are especially severe among women who became pregnant before the age of 15. In 2013, 3.1 percent of women in the DR reported having been pregnant in between the ages of 10 and 14, 44 and the country’s fertility curve begins rising at age 12, significantly younger than the fertility curves of Colombia, Guatemala, Haiti, Honduras, and Peru. 45 63. In 2018, 19.1 percent of late adolescents in the DR reported having been pregnant; 13.8 percent were already mothers, 4.8 percent had two or more pregnancies, and 3.97 percent had a miscarriage or abortion. Assuming that these rates have remained unchanged since 2018, recent demographic information implies that 90,092 girls and young women between the ages of 15 and 19 have been pregnant at least once, of which 65,230 are now mothers, while 18,765 have suffered a miscarriage or received an abortion. In addition, an estimated 7,400 girls between the ages of 10 and 14 have been pregnant at least once. 46 In 2020, a total of about 100,000 girls and young women in the DR had experienced an early or late adolescent pregnancy. 64. The incidence of late adolescent pregnancy is higher in rural areas, border regions, and among poor households nationwide. In 2018, 23.7 percent of late adolescents in the DR reported having been pregnant, but in urban areas this rate was just 18.1 percent. The regions with the highest rates of late adolescent pregnancy included Enriquillo (27.2 percent), Cibao Noroeste (25.6 percent) and El Valle (23.1 percent), all of which are along the Haitian border, while the lowest rates were in Ozama (16.1 percent) and Cibao Nordeste (18.1 percent). The early adolescent pregnancy rate was 4.6 percent in rural areas, compared with just 2.6 percent in urban areas.47 In 2018, the late adolescent pregnancy was 41.4 percent among households in extreme poverty and just 7.1 percent in the wealthiest households.48 In 2013, the 42 Vicepresidencia de la Republicana, 2017. Embarazo y adolescencia en la República Dominicana. 43 World Bank, 2012. Embarazo en adolescentes y oportunidades en América Latina y el Caribe. 44 ENHOGAR, 2018. 45 MacGaurrie et al., 2017. Fig A6.6 46 ENDESA, 2013. This estimate is based on the percentage of women ages 20-24 who report having been pregnant between the ages of 10 and 14. 47 Ibid. As above, this is the share of women ages 20-24 who reported having been pregnant between the ages of 10 and 14 in the 2013 ENDESA. 48 These households correspond to the ICV-1 and ICV-4 SIUBEN classifications. 45 rate of early adolescent pregnancy was also five times higher among the poorest households than among the wealthiest.49 Causes and Effects of Adolescent Pregnancy 65. The most important factors that contribute to late adolescent pregnancy in the DR are early marriage or sexual unions, low levels of educational attainment, and being neither engaged in the labor market nor enrolled in school. In 2018, 71.3 percent of last adolescents who reported having been married or in a sexual union had already been pregnant at least once, versus just 3.2 percent of those who had never been in such a relationship. In addition, 69.5 percent of late adolescents who reported having been pregnant had completed fewer years of schooling than what is expected for their age, compared to just 30.1 percent of adolescents who had never been pregnant. Finally, 44.5 percent of late adolescents who reported having been pregnant were neither engaged in the labor market nor enrolled in school. 66. While individual and household characteristics are correlated with adolescent pregnancy, many of these correlations vanish after other contextual factors have been taken into account. Probability of becoming pregnant during the late adolescence is positively associated with early marriage or sexual union, lower levels of educational attainment, and a diminished likelihood of being either engaged in the labor market or enrolled in school. Location and household characteristics are also significantly correlated with adolescent pregnancy, but these relationships are more complex than they may initially appear. For example, while rural areas have higher overall adolescent pregnancy rates than urban areas, their positions reverse when other circumstantial factors have been accounted for. Similarly, the Haitian border regions have the highest absolute rates of adolescent pregnancy, but after controlling for other factors the highest rates are in Higuamo and Valdesia. Likewise, having been born in Haiti and living in a poor household are only positively correlated with adolescent pregnancy when other factors have not been accounted for; when these factors are taking into consideration, Haitian-born Dominicans and members of poor households have lower rates of adolescent pregnancy than their peers. 67. Adolescent pregnancy is associated with lower levels of educational attainment, worse employment outcomes, and increased dependence on social programs. Female heads of household in their 20s and 30s who were adolescent mothers are more likely to have more children, participate in the workforce, be employed, earn more labor income, and depend on social programs such as Comer es Primero and Seguro Familiar de Salud en el Regimen Subsidiado;50 they are also less likely to have completed primary, secondary, or tertiary education. Moreover, 31.1 percent of adolescents who report having been pregnant did not attend school or college during their first pregnancy, and 48.7 percent dropped out after the birth of their first child.51 Early adolescent pregnancy has a greater impact than late adolescent pregnancy, and these differences are statistically significant for all outcomes except the probability of completing tertiary education. Policies, Programs, and Public Spending on Adolescent Pregnancy 68. Since the early 1990s, the DR has elaborated a policy framework to address adolescent pregnancy, but this framework does not include a ban on early marriage or measures to expand access to modern contraceptive methods. Nevertheless, the government recognizes adolescent pregnancy as 49 ENDESA, 2013. 50 The exception is early adolescent mothers, who are less likely to be enrolled in Seguro Familiar de Salud. 51 ONE, 2018. 46 major development challenge, and multiple strategic documents highlight the need for an integrated approach to addressing adolescent pregnancy. The National Development Strategy, the National Pact for Education Reform, and the specific plans related to adolescent pregnancy emphasize the role of education, health, and social policies in mitigating the risk factors for adolescence pregnancy and improving outcomes for adolescent mothers. Since 2011, the DR has implemented two plans designed to reduce the rate of adolescent pregnancy, the Plan EA (2011-2016) and the PREA (2019-2023), but budgetary and governance constraints undermined the effectiveness of both plans. 69. Public spending on adolescent pregnancy prevention and protection programs, as well as on support for adolescent mothers, is either very low or indistinguishable from expenditures earmarked for other target populations and activities. Between 2012 and 2019, the government implemented 21 programs related to adolescent pregnancy prevention and support for adolescent mothers. However, these programs were small, and a lack of adequate coverage information prevents a thorough analysis of their impact. 70. Despite the limited scope for expenditure analysis in the DR, financing programs that prevent adolescent pregnancy is almost certainly more cost effective than addressing its consequences. In 2013, the annual cost of adolescent pregnancy and motherhood was estimated at RD$2.1 billion52 (about US$50 million), substantially more than the estimated annual cost of implementing the Plan EA during 2011- 2016. Estimating and comparing the impact of different interventions will require comprehensive data on the resources currently spend on sexual and reproductive health education and adolescent pregnancy health programs, disaggregated by the age of the target population, as well as detailed information on the beneficiaries covered by existing programs. 71. While data limitations prevent a thorough empirical analysis, participation in certain programs and initiatives is associated with lower rates of adolescent pregnancy. Just 67.8 percent of adolescents who report having been pregnant have received sexual and reproductive health education at school or college, versus 81.3 percent of those who have never been pregnant. The gap in adolescent pregnancy rates between girls who have and have not had access to reproductive health education widens among the poorest households. Adolescents at all income levels are most likely to seek sexual-health-related information and support from a health center or from parents and friends, which suggest that an outreach strategy aimed at addressing adolescent pregnancy should leverage these relationships. Access to sexual health services at a health center is especially crucial for adolescents from moderately and extremely poor households, who are much more likely to seek support from a health center than from friends and family. 72. Municipal-level participation in the Jornada Escolar Extendida program and household-level participation in Comer es Primero and Escuelas de Familia are associated with a lower probability of late adolescent pregnancy. The Jornada Escolar Extendida program extends the school day, which reduces the exposure of adolescent girls to sexual activity—including coercive activity—by keeping them under the supervision of tutors and teachers. Comer es Primero imposes conditions on expectant mothers that positively impact maternal and child health. Escuelas de Familia offers access to information about contraceptive methods, which has proven effective in reducing the incidence of adolescent pregnancy. By contrast, public outreach campaigns aimed at reducing adolescent pregnancy and protecting the welfare of adolescent mothers have been limited, and no information is available on their effectiveness. Similarly, programs that strengthen socio-emotional skills can help prevent adolescent pregnancy, but in the DR these programs are modest in scale and not subject to rigorous evaluation. 52 UNFPA-INTEC, 2013. 47 Recommendations To enhance the effectiveness of its efforts to reduce adolescent pregnancy, the government should: • Implement a comprehensive strategy that includes female and male adolescents, their families and friends, teachers and school administrators, social and religious organizations, and the media and leverages complementary interventions to address the full range of factors that contribute to adolescent pregnancy. Interventions implemented under this strategy should bolster sexual and reproductive health education, enhance access to family planning services, and strengthen socio-emotional skills, life skills, and the development of self-esteem and personal agency among adolescent girls. • Amend the laws governing the marriage of minors to raise the legal age of marriage and establish a broader legal framework to protect adolescent girls and prevent adolescent pregnancy. • Define the role of each public agency involved in implementing the PREA, establish monitoring and evaluation mechanisms, and allocate adequate budgets across agencies. Institutional roles should be defined via presidential decree. Implementing agencies require both targeting and monitoring and evaluating mechanisms underpinned by compatible systems for identifying the beneficiaries of adolescent pregnancy initiatives and measuring their impact. Empirical evidence will be critical to enable the government to review policies and redesign programs based to enhance their effectiveness. • Expand social programs that are associated with a lower probability of adolescent pregnancy, including Jornada Escolar Extendida, Comer es Primero, and Escuelas de Familia. In the absence of rigorous empirical evidence, policymakers should leverage correlations and intuitive reasoning to identify the programs that are most effective in reducing adolescent pregnancy. Jornada Escolar Extendida, Comer es Primero, and Escuelas de Familia appear to lower the probability of adolescent pregnancy and mitigating the risks associated with adolescent motherhood by limiting the exposure of adolescent girls to sexual activity, incentivizing pregnant adolescents to seek medical care, and expanding access to reproductive health information. The government can also build on the apparent success of these programs by creating new mechanisms to incentivize school attendance, support the reintegration of students who have dropped out of the school system, provide daycare options for adolescent mothers who wish to obtain further education, and encourage adolescent girls to seek out reproductive health information before they become pregnant. • Develop new programs to expand access to modern contraceptive methods and ensure that students receive reproductive health education in public school. Access to contraception and health education are strongly associated with lower rates of adolescent pregnancy. Reproductive health programs should stress the importance of female agency in the use of contraception. • Strengthen the provision of comprehensive reproductive health services for adolescents and adolescent mothers in public hospitals and clinics. Adolescents are most likely to seek reproductive health information and support from public health facilities, especially adolescents 48 from poorer households. Adolescent mothers need special attention in order to avoid repeated adolescent pregnancy, since the adverse effects of being pregnant are intensified. To ensure access to high-quality reproductive health services, health personnel should receive specialized training, and dedicated funding for adolescent reproductive healthcare should be earmarked for public health centers. 49 Bibliography Administradora de Subsidios Sociales (2017). Programa Comer es Primero (PCP). Santo Domingo. Azevedo, J. et al (2012). Teenage Pregnancy and Opportunities in Latin America and the Caribbean: On Teenage Fertility Decisions, Poverty and Economic Achievement, Washington DC, USA. Azevedo J.P, Favara M., (2012), “The Impact of Bolsa Familia on the Incidence of Teenage Pregnancies in Brazil”. Banco Mundial, UNICEF (2017). Impacto económico del matrimonio infantil y las uniones tempranas: resumen para la República Dominicana, Washington, DC. Baumgartner, J. N., Geary, C. W., Tucker, H., & Wedderburn, M. (2009). The Influence of Early Sexual Debut and Sexual Violence on Adolescent Pregnancy: A Matched Case-Control Study in Jamaica. International Perspectives on Sexual and Reproductive Health, 35(1), 21–28. http://www.jstor.org/stable/25472412. BID (2012). “Salud Sexual y Reproductiva para Jóvenes: Revisión de evidencia para la prevención. Washington. D.C. Carvalho, M., K. Alencar, A. Melhado, V. Coates, y H. Omar (2007) Teenage Pregnancy: Impact of the Integral Attention Given to the Pregnant Teenager and Adolescent Mother as a Protective Factor for Repeat Pregnancy TheScientificWorld JOURNAL (2007) 7, 187–194. Central Bank of the Dominican Republic (2019): Encuesta Nacional Continua de Fuerza de Trabajo 2018. Database. Charles, J., J. Rycroft-Malone, R. Aslam, M. Hendry, D. Pasterfield and R. Whitaker (2016). Reducing repeat pregnancies in adolescence: applying realist principles as part of a mixed-methods systematic review to explore what works, for whom, how and under what circumstances. BMC Pregnancy and Childbirth (2016) 16:271 Chen, X., Wu Wen S., Feming, N. Demissie, K, Rhoads, G. and Walker, M. (2007) Teenage pregnancy and adverse birth outcomes: a large population based retrospective cohort study. International Journal of Epidemiology 2007;36:368–373. Dirección General de Presupuesto: Informes de Ejecución Físico Financiera. Different years www.digepres.gob.do Dominican Government: National Pact for Education Reform in the Dominican Republic (2014-2030). ________ Constitution of the Dominican Republic. ________ Criminal Code of the Dominican Republic Act Nº 550-14. ________National Development Strategy Act Nº 1-12. 50 ________ Dominican Code for the Protection System and Basic Rights of Children and Adolescents Act Nº 136-03 Freitas Galvão, R, C. Olivares, A. Borovac-Pinheiro, D. de Morais, D. Faria-Schützer and F. Garanhani (2018). Hazards of Repeat Pregnancy during Adolescence: A Case-control Study. Revista Brasileira Ginecología Obstetricia 2018;40:437–443. Gabinete de Coordinación de la Política Social (2020). Final Draft Plan Nacional para la eliminación del matrimonio infantil y reducción de las uniones tempranas en la República Dominicana (Plan MIUT) 2030. Harner, H. (2004). Domestic Violence and Trauma Care in Teenage Pregnancy: Does Paternal Age Make a Difference?. JOGNN Clinical Research. 33 (3), 312-319. Holness, N. (2015). A Global Perspective on adolescent pregnancy International Journal of Nursing Practice; 21: 677–681 Kearney, M. & Phillip B. Levine (2015). http://ideas.repec.org/a/aea/aecrev/v105y2015i12p3597- 3632.hml. American Economic Review, American Economic Association, vol. 105(12), pages 3597-3632, December. Kirby, D. (2007). Emerging Answers, Research Findings on Programs to Reduce Teen Pregnancy and Sexually Transmitted Diseases. Washington, DC: National Campaign to Prevent Teen and Unplanned Pregnancy Kruger, D.I y Berthelon M. (2012). Education consequences of adolescent motherhood in Chile. MacQuarrie, K., Lindsay Mallick, and Courtney Allen (2017). Sexual and Reproductive Health in Early and Later Adolescence: DHS Data on Youth Age 10-19. DHS Comparative Reports 45. DHS Program. Ministry of Education: Institutional Annual Reports, different years. www.minpre.gob.do/transaprencia/servicios-al-publico National Office of Statistics (2019). Vital Statisticcs 2019. https://www.one.gob.do/demograficas/estadisticas-vitales _________ (2018). Dominican Republic National Multipurpose Household Survey ENHOGAR 2018. National Office of Statistics-UNICEF (2016). Dominican Republic National Multipurpose Household Survey ENHOGAR 2014. Neal, Harvey, Chadra-Mouli, Caffe y Camacho (2018). Trends in adolescent first births in five countries in Latin America and the Caribbean: disaggregated data from demographic and health surveys. Reproductive Health 15: 146. https://doi.org/10.1186/s12978-018-0578-4 ORC Macro/MEASURE DHS+, CESDEM (2014): Demographic and Health Surveys, Dominican Republic 2013. https://dhsprogram.com/data/available-datasets.cfm 51 __________(2008) Demographic and Health Survey, Dominican Republic 2007. https://dhsprogram.com/data/available-datasets.cfm __________(2003) Demographic and Health Survey, Dominican Republic 2002. https://dhsprogram.com/data/available-datasets.cfm __________(1997) Demographic and Health Survey, Dominican Republic 1996. https://dhsprogram.com/data/available-datasets.cfm __________(1992) Demographic and Health Survey, Dominican Republic 1992. https://dhsprogram.com/data/available-datasets.cfm Office of the First Lady: Institutional Annual Reports, different years. www.minpre.gob.do/52ransparencia/servicios-al-publico Plan Internacional (2017): Reporte Niñas Embarazadas: Desafíos de la respuesta nacional de prevención de embarazos en la adolescencia. Planteamientos. Septiembre. No. 3 PROSOLI (2020). Memoria Institucional, 2015-2019. Retrieved from: http://transparencia.progresandoconsolidaridad.gob.do/ Social Policy Coordination Cabinet (2019 a). National Adolescent Pregnancy Reduction Plan (PREA-RD). Retrieved from: https://repositorio.msp.gob.do/bitstream/handle/123456789/1218/planprea.pdf?sequence=1&isAllow ed=y _______(2019 b). Universidad de California Berkeley revela hogares Prosoli tienen mejor salud, educación y menos violencia. En PROSOLI. Retrieved from: https://progresandoconsolidaridad.gob.do/noticias/universidad-de-california-berkeley- revela-hogares- prosoli-tienen-mejor-salud-educacion-y-menos-violencia/ UNICEF, ECLAC (2007). “Teenage motherhood in Latin America and the Caribbean Trends, problems and challenges”. Recuperado de: https://repositorio.cepal.org/bitstream/handle/11362/36002/1/desafios_4_eng_Final_en.pdf UNFPA-INTEC (2013).Costos del embarazo y la maternidad en la adolescencia en la The Dominican Republic. Santo Domingo. Vicepresidencia de la Republica Dominicana, UNICEF, PROSOLI (2019). “El matrimonio infantil y las uniones tempranas Resumen del estudio de conocimientos, actitudes y prácticas en seis municipios de la República Dominicana”. Recuperado de: https://www.unicef.org/republicadominicana/CAP_MIUT.pdf WHO. 1989. The Health of Youth, Document A42/Technical Discussion/2. Geneva: World Health Organization. Wilson-Mitchell, K. Bennett, J. and Stennett, R. (2014) Psychological Health and Life Experiences of Pregnant Adolescent Mothers in Jamaica. International Journal Environmental Research and Public Health 2014, 11, 4729-4744. 52 World Bank (2019). Adolescent fertility rate (births per 1,000 women ages 15-19). Retrieved from: https://data.worldbank.org/indicator/SP.ADO.TFRT 53 Appendices Table A. 14. Legal Framework Supporting Adolescent Pregnancy Prevention and Protection Programs Legal Instrument Article Constitution of the Dominican Article 39.- Right to Equality Republic Paragraph 4.- Women and men are equal before the law. Any act which has the purpose or effect of impairing or nullifying the recognition, enjoyment or exercise equally of the fundamental rights of women and men is hereby prohibited. The necessary measures shall be promoted to ensure the eradication of gender inequalities and discrimination. Article 55.- Family Rights. Paragraph 6.- Motherhood, whatsoever the social or marital status of women, shall enjoy the protection of the public authorities and shall generate the right to official assistance in case of helplessness. Paragraph 10.- The State shall promote responsible paternity and maternity. The father and mother, even after separation and divorce, have the shared and inalienable duty to feed, raise, prepare, educate, maintain, provide security and assist their children. The law shall establish the necessary and appropriate measures to ensure the effectiveness of these obligations. Article 56.- Protection of minors. Paragraph 1.- The eradication of child labor and all forms of abuse or violence against minors is hereby proclaimed to be of the highest national interest. Children and adolescents shall be protected by the State against all forms of abandonment, kidnapping, state of vulnerability, physical, psychological, moral or sexual abuse or violence, commercial, labor, economic exploitation and hazardous tasks. Article 63.- Right to Education. All persons have the right to a comprehensive, quality, permanent education, under equal conditions and opportunities, with no other limitations than those derived from their aptitudes, vocation and aspirations. Article 63, Paragraph 3.- The State shall guarantee free public education and shall declare it mandatory at the early, elementary and secondary levels. Article 61.- Right to Health. Every person has right to the integral health. Hence: 1) The State shall protect the health of all persons, ensure access to safe drinking water, improved nutrition, health services, hygienic conditions and environmental sanitation, as well as provide the means for the prevention and treatment of all diseases, ensuring access to quality medicines and providing free medical and hospital assistance to those who require it. Dominican Code for the Article 396 - SANCTION OF ABUSE AGAINST CHILDREN AND ADOLESCENTS. Whereas: Protection System and Basic Paragraph c. - Sexual child abuse means the sexual practice with a child or adolescent Rights of Children and by an adult, or person with a 5-year age gap or over, for own sexual gratification, Adolescents Act Nº 136-03 without consideration of the child's or adolescent's psychosexual development and that said abuse can occur even without physical contact. It shall be punished with two (2) to five (5) years of imprisonment and a penalization of three ( 3) to ten (10) times the statutory minimum wage, in force at the time the offence is committed; if the offender has a relationship of authority, guardianship or supervision (teacher, guardian, civil servant, police officer, etc.) over the child or adolescent and severe injuries occur, as verified by experts on the subject, the maximum penalty indicated above shall be enforced. Should the offender be foreign or national who, while committing the offense, negotiates, traffics or has been linked to child or adolescent traffickers or traders to commit the offense, he or she will be punished by double the maximum penalty. 54 Legal Instrument Article Art. 397.- PENALTIES FOR ABUSE BY THE CHILD'S GUARDIAN OR TUTOR. If the abuse is committed by the father, mother or other relatives, tutors or guardians responsible for the child or adolescent, against his or her sons or daughters or those placed under his or her guardianship or authority, they shall be punished with two ( 2) to five (5) years of imprisonment and a penalization of one (1) to five (5) times the statutory minimum wage. In any case, the penalty must be accompanied by psychotherapeutic treatment. National Development Article 12. Gender-sensitive approach: All plans, programs, projects and public policies shall Strategy 2030 Act Nº 1-12 incorporate the gender approach in their respective fields of action, to identify situations of discrimination between men and women and take action to ensure gender equality and equity. 2.2.1.2 Strengthen collective health services related to the events in each life cycle, in collaboration with local authorities and communities, with an emphasis on sexual and reproductive health, addressing the particularities of each gender, adolescent pregnancy prevention, and prevention of communicable (tuberculosis, dengue, malaria, HIV and AIDS, among others), chronic (hypertension, diabetes, cardiovascular disease, obesity, among others) and catastrophic diseases (breast, cervical and prostate cancer, among others), as well as the promotion of healthy lifestyles by ensuring access to vulnerable and extremely poor populations. 2.3.1.7 Raise awareness of equal rights and gender equality to build a revalued image of women's contribution to the economy and society that goes beyond the stereotypes traditionally assigned to males and females, using the spaces and activities developed by schools, municipal governments and civil society organizations. 2.3.4.3 Foster a culture of respect and protection of the fundamental rights of children, adolescents and youngsters. 2.3.4.10 Incorporate into the school curriculum sexual and reproductive health education for children, adolescents, including sexually transmitted infections (STIs) and HIV, and promote responsible parenthood and family life values, within a gender equality and respect framework. 2.3.4.11 Ensure that pregnant adolescents stay in the school system throughout regular school schedule, providing it does not pose any additional risk to their overall health, and encourage the expecting couple to assume their parental responsibilities. 3.1.2.5 Promote the inclusion of gender-sensitive life-cycle perspective into the design and execution of the National Budget, with an emphasis on the allocation of resources to priority areas to overturn inequalities. 3.5.5.16 Establish mechanisms for preventing, reporting and punishing harassment, violence and sexual exploitation against children, adolescents and women. Criminal Code of the SECTION III - ABORTION Dominican Republic Act Nº 550-14 Article 107. Abortion. Unless otherwise provided in Article 110, any person who, by means of food, drink, medicine, tests, treatment or otherwise, causes the termination of a woman's pregnancy or cooperates to do so, even if the expecting mother consents, shall be penalized with two to three years in prison. Paragraph I: The same punishment shall be imposed on the woman who carries out an abortion or who consents to use substances indicated or administered thereto for such purpose, or who consents to undergo the aforesaid means of abortion, provided that such abortion has been carried out. Paragraph II: If such abortion is not carried out in full but results in serious injury or illness to the fetus that impairs its normal development or causes severe physical or psychological damage, the offender shall be sentenced to one to two years in prison. 55 Legal Instrument Article Article 108. Penalties for medical professionals or midwives. Medicine doctors, nurses, pharmacists and other medical professionals, as well as midwives, who, by misconducting their practice or profession, cause or help to cause the abortion, shall be punished with four to ten years in major imprisonment. Article 109. Penalties for the death of the expecting mother. If the events referred to in articles 107 and 108 of this Code hereto cause the death of the woman, the offender shall be punished with ten to twenty years in major imprisonment. Article 110. Exceptions. The termination of a pregnancy carried out by specialized medical staff in a public or private health facility is not punishable if all available scientific and technical means are exhausted to save both lives, to the fullest extent possible. Paragraph. The termination of a pregnancy due to rape, incest, or one resulting in clinically- proven fetal anomaly or malformation incompatible with life shall be subject to the requirements and protocols established by special law. National Pact for Education 3.1.6 Strengthen and implement affective-sex education programs, as well as adolescent Reform in the Dominican parental support and accompaniment schemes to ensure their stay in the school system. Republic (2014-2030). 3.4.6 Establish early childhood development centers in the vicinity of colleges and vocational training centers as a mechanism for promoting early childhood development, facilitating access to education for mothers and fathers, and keeping vulnerable youngsters and adults in the education system, thus helping prevent school dropouts. 4.2.4 Incorporate into the school curriculum, from early childhood and using pedagogical strategies suitable for each level, sexual and reproductive health education, knowledge and prevention of sexually transmitted infections and HIV, as well as the teaching of core values of respect, gender equality and equity, family life and responsible parenthood. Source: Authors’ compilation based on information retrieved from the National Pact for Education Reform in t he Dominican Republic (2014-2030), Criminal Code of the Dominican Republic Act Nº 550-14, the National Development Strategy 2030 Act Nº 1-12. 56 Table A. 15. Mean age at first sex among adolescents aged 15-19, by instruction in SRHE, area of residence, family socioeconomic group and years of school completed, 2018 Attended SRH education Did not attend SRH Total class education class Total 15.2 (1.7) 15.4 (1.6) 14.8 (1.6) Zone Rural 15.1 (1.7) 15.3 (1.7) 14.6 (1.7) Urban 15.3 (1.6) 15.5 (1.6) 14.8 (1.6) Socioeconomic Group Very low 14.9 (1.6) 15.2 (1.5) 14.5 (1.7) Low 15.3 (1.7) 15.4 (1.8) 15 (1.7) Lower-middle 15.3 (1.6) 15.4 (1.6) 14.8 (1.4) Middle and upper- middle 15.6 (1.6) 15.9 (1.6) 14.7 (1.5) High 15.6 (1.4) 15.7 (1.5) 15.4 (1.2) Number of years of school completed 0-8 14.4 (1.5) 14.5 (1.5) 14.3 (1.6) 9-12 15.5 (1.5) 15.5 (1.6) 15.2 (1.5) >12 16.3 (1.7) 16.5 (1.6) 15.3 (1.7) Source: Authors' compilation based on ENHOGAR-2018 data. Note: The standard deviation of the mean is shown in brackets. A sexually active adolescent refers to a woman aged 15-19 who has had sex in the last 12 months. 57