Volume 91, Issue 9 p. 1114-1118
Free Access

Childbearing in adolescents aged 12–15 years in low resource countries: a neglected issue. New estimates from demographic and household surveys in 42 countries



Centre for Global Health Population Poverty and Policy, University of Southampton, Southampton, UK

Search for more papers by this author


Centre for Global Health Population Poverty and Policy, University of Southampton, Southampton, UK

Search for more papers by this author


Department of International Development, University of Oxford, Oxford, UK

Search for more papers by this author


Department of Health, Education and Research, NORAD, Oslo, Norway

Search for more papers by this author


Department of Child and Adolescent Health and Development, WHO, Geneva, Switzerland

Search for more papers by this author


Sexual and Reproductive Health Branch, Technical Division, UNFPA, New York, NY, USA

Search for more papers by this author
First published: 23 May 2012
Citations: 149
Sarah Neal, Centre for Global Health, Population, Poverty and Policy, School of Social Science, University of Southampton, Southampton, UK SO17 1BJ. E-mail: [email protected]

Conflict of interest:
Sarah Neal, Zoë Matthews, Melanie Frost, Alma Virginia Camacho and Laura Laski have have stated explicitly that there are no conflicts of interest in connection with this article. Helga Fogstad has no conflict of interests but is employed by the Norwegian Agency for Development Cooperation (NORAD). Alma V Camacho is a staff member of the World Health Organization. The author alone is responsible for the views expressed in this publication and they do not necessarily represent the decisions or policies of the World Health Organization.

Please cite this article as: Neal S, Matthews Z, Frost M, Fogstad H, Camacho AV, Laski L. Childbearing in adolescents aged 12–15 in low resource countries: a neglected issue. New estimates from demographic and household surveys in 42 countries. Acta Obstet Gynecol Scand. 2012; 91:1114–1118.


There is strong evidence that the health risks associated with adolescent pregnancy are concentrated among the youngest girls (e.g. those under 16 years). Fertility rates in this age group have not previously been comprehensively estimated and published. By drawing data from 42 large, nationally representative household surveys in low resource countries carried out since 2003 this article presents estimates of age-specific birth rates for girls aged 12–15, and the percentage of girls who give birth at age 15 or younger. From these we estimate that approximately 2.5 million births occur to girls aged under 16 in low resource countries each year. The highest rates are found in Sub-Saharan Africa, where in Chad, Guinea, Mali, Mozambique, Niger and Sierra Leone more than 10% of girls become mothers before they are 16. Strategies to reduce these high levels are vital if we are to alleviate poor reproductive health.


  • ASBR
  • age-specific birth rate
  • DHS
  • Demographic and Household Survey
  • Introduction

    It is frequently cited that girls who give birth aged 15–19 are more than twice as likely to die as those in their 20s (1,2). However, this fails to capture the fact that risk increases with decreasing age. Few studies have disaggregated outcomes for adolescent mothers by age, but one Latin American study (3) found that after adjusting for confounding factors such as economic status, parity and maternal health care, girls aged 15 or under had an odds ratio for maternal death four times higher than women aged 20–24. Girls aged 16–19, however, did not experience increased risk. An earlier study from Bangladesh found that girls aged 15–19 had a maternal mortality rate nearly twice that of women aged 20–24 and the rate for girls aged 10–14 was nearly five times higher (4). These findings are broadly supported by data from more wealthy countries (5). Morbidity is also greater in this age group: Conde-Agudelo et al. (3) found that girls aged 15 or less had markedly higher odds ratios for conditions such as eclampsia, anemia, postpartum hemorrhage and puerperal endometritis than older adolescents. There is also evidence that younger adolescents may be more at risk of obstructed labor as their pelvic bones are still forming (6). Risk of obstetric fistula is also greatly increased in this group, which can lead to a lifetime of social and economic disadvantage: one study in Nigeria found that over 80% of cases first developed before the age of 15 (7). Evidence also suggests that the adverse neonatal outcomes associated with adolescent pregnancies are greater for those below 16 years (3).

    This article presents estimates of birth rates for the 12–15-year-old age group in 42 low income countries. These rates have not been calculated and published before, although nationally representative survey data with full birth histories for women of all reproductive ages have been available for many years. Also presented here are estimates of the percentage of girls giving birth before the age of 16 as well as estimates of the actual numbers of girls who give birth in this age group annually for each country; this provides a clearer picture of the scale of the problem.

    Material and methods

    The data used in this article were drawn from 42 Demographic and Health Surveys (DHS) carried out since 2003. These are large, nationally representative surveys providing data on a wide range of indicators in the areas of population, health and nutrition. Full birth histories are collected from women aged 15–49 years in sampled households, and data is comparable both over time and between countries. Twenty-eight of the countries are in Africa, six in Asia, six in Latin America and two in Eastern Europe. Surveys where the sample was restricted to married women were excluded as this would bias estimates.

    Three measures of early adolescent childbearing were estimated: first the age-specific birth rate (ASBR) for 12–15-year-olds, second the percentage of girls who give birth at age 15 or younger, and third the estimated actual number of births occurring each year to girls aged 12–15. We also calculated the percentage of births occurring at each age within this age group. Analysis was carried out using SPSS (SPSS Inc., Chicago, IL, USA).

    The lower age limit of 12 for the ASBR was established as a result of preliminary analysis showing that very few (less than 1% in most countries) births under 16 were to girls less than 12 years of age. Fifteen was used as an upper cut-off as there is some indication based on analysis in wealthier countries that this may be the age below which neonatal outcomes are most adversely affected (8).

    The DHS does not normally interview girls below the age of 15, but asks all respondents about their childbearing history. From these data we were able to calculate the number of births reported by each participant in the years prior to the survey that occurred when they were between the ages of 1 and 15. For the purposes of this study we only included participants who were interviewed at the age of 16–19 in order to reduce recall bias; therefore the births that inform our estimates occurred up to seven years prior to the date of the survey. A denominator was then established by calculating the number of years each girl contributed to this period for this age group. The number of births was then divided by the number of years and multiplied by 1000 to get a rate per 1000. In keeping with normal DHS methodology, no adjustment was made for possible age heaping.

    It is not likely that this methodology will introduce greater error than that used for other age groups. The main issues with calculating fertility rates for any age group are recall bias and age misreporting. The issue of ages being misreported should cause no greater problem for the 12–15-year age group than for any other. While recall bias may be slightly increased, but this is unlikely to be a significant problem as the recall period for fertility rates calculated from most DHSs is five years and the recall period used here is three to seven years.

    While the ASBR is the most commonly used indicator for adolescent fertility, the percentage of girls giving birth at age 15 or younger is a more easily and intuitively understood measure. To ensure that the estimate was as current as possible, only respondents aged 16–19 were included in the analysis.

    In addition, an estimate for the absolute number of girls giving birth aged 12–15 each year within each country was calculated using United Nations population estimates for 2010 as a denominator (9). These estimates are presented by five-year age groups (10–14 and 15–19), so estimates for the four-year age group 12–15 were calculated as; (age 10–14 population estimate × 0.6) + (age 15–19 population estimate × 0.2).

    In many low resource countries a very high proportion of the population is aged 15 or younger, and therefore the actual scale of the problem is much more significant than might be expected. Estimates of the actual numbers of girls affected per country also gives a clearly understood measure of the problem, and allows a very rough initial estimate to be made of the incidence of very early adolescent births in low resource countries as a whole. This was calculated by extrapolating the total number of births for countries in the study to the whole region based on the proportion of all girls aged 12–15 in each region that were included in the study countries. This methodology obviously has limitations as it assumes countries not included in the study have a similar rate of ASBR for 12–15-year-olds as those in the same regions that are included in the study. As there is much heterogeneity within regions this is obviously problematic, and the estimate for Asia may be particularly flawed as the estimate is based on a very small number of countries. These regional and global estimates need to be interpreted with caution, but they are still an important first step in quantifying the issue of very early adolescent births.


    Table 1 clearly illustrates that birth rates among 12–15-year-old adolescents in many countries are high; it also shows that there is wide variation between countries. The highest rates are found in sub-Saharan Africa, and are particularly high in a number of countries in West and Central Africa. Mali has the highest rate at 60/1000, and the rate is also very high in Guinea (49/1000) and Niger (48/1000). While most countries in Eastern Africa have relatively moderate or low percentages of very early pregnancy, Mozambique and Madagascar have rates of 24 and 27/1000, respectively. Conversely, some countries within these sub-regions have very low levels: in Rwanda only 2/1000 girls give birth before the age of 16.

    Table 1. Estimates for age-specific birth rate 12–15,% girls giving birth at age 15 or under and total annual number of births to girls aged 12–15 years by country.
    Country Age-specific birth rate/1000 12–15 years Percentage of girls giving birth aged 15 or under Total annual number of births to girls aged 12–15 years
    Benin 2006 26.6 5.4 11 400
    Burkina Faso 2003 18.0 3.7 13 700
    Cameroon 2004 36.8 7.6 33 600
    Chad 2004 47.8 14.2 26 800
    Congo Brazzaville 2005 22.9 6.5 4 000
    Congo DRC 2007 26.0 7.2 88 800
    Ethiopia 2005 25.4 5.5 106 800
    Ghana 2008 11.1 3.0 12 000
    Guinea 2005 48.9 12.2 23 400
    Kenya 2003 14.6 4.2 27 600
    Lesotho 2004 7.8 1.8 800
    Liberia 2007 29.7 8.5 5 800
    Madagascar 2003/4 27.2 9.6 27 100
    Malawi 2004 34.2 4.7 27 600
    Mali 2006 60.0 12.7 38 600
    Morocco 2003/4 3.8 0.6 4 600
    Mozambique 2003 24.2 10.5 27 200
    Namibia 2006/7 12.2 3.1 1 200
    Niger 2006 47.9 12.8 36 400
    Nigeria 2008 36.8 7.9 271 100
    Rwanda 2005 2.4 0.5 1 100
    Senegal 2006 24.2 5.7 15 200
    Sierra Leone 2008 42.8 10.1 11 800
    Swaziland 2006/7 16.4 4.2 1000
    Tanzania 2004/5 10.0 3.6 21 200
    Uganda 2006 22.8 4.7 39 400
    Zambia 2007 21.7 4.7 14 400
    Zimbabwe 2005/6 11.4 2.9 7 500
    Armenia 2005 0.3 0.1 <30
    Azerbaijan 2006 1.0 0.2 300
    Cambodia 2005 4.6 0.7 680
    India 2005/6 12.8 3.6 602 800
    Nepal 2006 11.5 2.1 15 900
    Philippines 2003 3.2 0.9 12 600
    Latin America and the Caribbean
    Bolivia 2003 12.8 4.2 5 600
    Colombia 2005 16.6 5.6 28 600
    Dominican Republic 2007 23.3 6.5 9 400
    Haiti 2005/6 9.8 2.3 4 500
    2005/6 20.7 5.4 7 300
    Peru 2004 10.1 3.4 11 600
    Eastern Europe
    Moldova 2005 1.6 0.4 140
    Ukraine 2007 0.4 0.5 310
    • For total annual numbers, figures are rounded to nearest 100 for estimates over 1000, and to the nearest 10 for estimates less than 1000.

    The analysis of the percentage of girls aged 16–19 who have given birth by the age of 16 shows that in six sub-Saharan African countries more than 10% had become mothers before this age: Chad, Guinea, Mali, Mozambique, Niger and Sierra Leone. In Chad 14% of girls (or one in seven) had become mothers at age 15 or younger. As different sample groups were used for analyzing the ASBR and the percentage of girls giving birth before the age of 16, the figures are not directly comparable.

    Several countries in Latin America have significant rates of adolescent births aged 12–15 (e.g. Dominican Republic 23/1000, Honduras 21/1000). The six countries in Asia had quite low rates and the rates for the two Western Asian countries (Azerbaijan and Armenia) are particularly low. Available data for the two countries in Eastern Europe also showed very low ASBRs for 12–15 (below 2/1000).

    As can be seen from Table 1 the countries with the greatest numbers do not necessarily have the highest rates: India has a fairly moderate rate of only 13/1000, but a total number of more than 600 000 due to its large population. Nigeria also has over a quarter of a million girls giving birth before the age of 16 each year.

    If we extrapolate our estimates to the whole region, it indicates that just over a million births may occur to girls younger than 16 in Africa each year. In Asia (not including Japan and China) the rough estimate also stands at just over a million, although this is based on a relatively small number of countries. A markedly lower figure of just over 300 000 was estimated for Latin America and the Caribbean. Numbers were not estimated for low resource countries in Eastern Europe as data from only two countries are available. Therefore we estimate that there may be as many as 2.5 million births to girls under 16 each year in these regions alone.


    This article estimates that around 2.5 million births occur to girls aged under 16 in low resource countries each year. On average, around 50–65% of these are to girls aged 15 in the study countries, and the remaining 35–50% are to younger girls. As statistics for adolescent births are only usually calculated for girls aged 15–19 this means that there may be around a million births each year to girls under 15 that have come to light only from the analysis in this article.

    Generally, the pattern of very early pregnancy is one of heterogeneity between, and in many cases within, regions. However, this analysis does point to a very high rate of births to girls under 16 in Western, Central and parts of Eastern Africa, as well as to a lesser extent Latin America. The patterns of variation tend to strongly reflect the levels of overall adolescent fertility. For instance, despite their geographical proximity, Mali, Guinea and Senegal have marked differences between both their birth rate for 12–15-year-olds (60/1000, 49/1000 and 24/1000, respectively) and their birth rate for 15–19-year-olds (190/1000, 153/1000 and 96/1000) (10). Further research is needed to understand more fully the reasons for these differences, but it is likely to be at least partly the result of differing cultural patterns and practices related to factors such as age at marriage.

    It is evident that many of the countries with particularly high levels of very early adolescent motherhood are also those with very high maternal mortality ratios. Undoubtedly the high proportion of early adolescent pregnancies will be contributing to this, and the risk of mortality in girls under the age of 16 in these countries will be extremely high.

    Greater disaggregation by age is essential if the risks and potentially adverse consequences of adolescent fertility are to be assessed accurately. A greater understanding of the scale of pregnancy in early adolescence, as well as a focus on the potential health risks faced by very young mothers will be valuable in supporting advocacy efforts to address this issue. Further research should also be carried out to more accurately understand trends and determinants of very early childbearing in order to inform the development of context-appropriate, multi-sectoral interventions to reduce pregnancies in this very vulnerable group.


    This study was funded by the Norwegian Agency for Development Cooperation (NORAD).