The pooled proportion and associated factors of adolescent pregnancy in 47 low- and middle-income countries: A multilevel analysis

Abstract

Background:

In low- and middle-income countries (LMICs), nearly 21 million adolescents become pregnant every year. Half of these pregnancies are unplanned, and more than half of unintended pregnancies result in unsafe abortion, which accounts for a major proportion of adolescent pregnancies worldwide. This study intended to assess the pooled proportion of adolescent pregnancy and associated factors in LMICs.

Methods:

A community-based cross-sectional study design was employed. The data were taken from 47 LMICs from 2015 to 2024. A total of 327,394 (weighted = 323,767) adolescents aged 15 to 19 years were included. The data were sourced from the demographic and health survey (DHS) datasets available online. Data were analyzed using STATA V.17. We used multivariable multilevel logit regression for the outcome variable. The p-values < 0.05 were regarded as statistically significant, and the adjusted odds ratios with 95% confidence intervals (CIs) were computed from the final model. The candidate model was evaluated using the Deviance Information Criterion.

Results:

The pooled proportion of adolescent pregnancy among all adolescent girls in 47 LMICs was 23% (95% CI: [20, 26]). It ranges from 10% (95% CI: [6, 14]) in North Africa/West Asia/Europe to 28% (95% CI: [25, 32]) in sub-Saharan Africa (SSA). No formal education (AOR = 1.10, 95% CI: [1.04, 1.16]), primary education (AOR = 1.58, 95% CI: [1.51, 1.66]), middle wealth index (AOR = 0.83, 95% CI: [0.80, 0.87]), rich wealth index (AOR = 0.58, [0.55, 0.61]), media exposure (AOR = 0.86, 95% CI: [0.83, 0.89]), aged 18 to 20 (AOR = 4.33, 95% CI: [4.17, 4.49]), female household heads (AOR = 1.40, 95% CI: [1.35, 1.46]), condom use (AOR = 0.79, 95% CI: [0.74, 0.85]), contraceptive use (AOR = 0.44, 95% CI: [0.42, 0.46]), knowledge of contraception (AOR = 0.38, 95% CI: [0.35, 0.41]), being married (AOR = 9.02, 95% CI: [8.91, 9.73]), having a bank account (AOR = 0.89, 95% CI: [0.83, 0.96]), being from SSA (AOR = 8.79, 95% CI: [6.96, 10.24]), Central Asia (AOR = 2.96, 95% CI: [2.16, 3.67]), South and Southeast Asia (AOR = 1.65, 95% CI: [1.35, 2.03]), Oceania (AOR = 4.11, 95% CI: [2.66, 5.01]), Latin America and the Caribbean (AOR = 5.83, 95% CI: [4.93, 6.49]) were significantly associated with adolescent pregnancy.

Conclusion:

The pooled proportion of adolescent pregnancy is high in the study area with significant disparity. Improving women’s education, media exposure, financial support, knowledge, and access to contraceptive and condom use among adolescents were potential modifiable factors to reduce adolescent pregnancy. The WHO regions, such as SSA and the South and Southeast Asia regions, need particular attention to lower adolescent pregnancy rates.

Keywords

low- and middle-income countries adolescent pregnancy multilevel model hierarchical model

Background

Adolescence is the stage between childhood and adulthood during which individuals undergo physical, mental, and social development.¹ During this period, they explore their environment and form new relationships. Peer pressure can influence adolescents and expose them to risky behaviors, potentially leading to frequent unprotected sex.^2,3 Adolescent pregnancy can be influenced by a complex structural, cultural, and socioeconomic factors,³ such as marriage, substance abuse, family experiences of adolescent birth, a lack of sex education, and limited health services.²

As of 2023, nearly 17% of adolescent girls worldwide gave birth before the age of 18, although this rate varies significantly across regions.⁴ According to 2020 estimation, approximately 21 million girls aged 15 to 19 become pregnant every year, demonstrating adolescent pregnancy as a persistent public health concern in low- and middle-income countries (LMICs). Notably, half of these pregnancies are unintended, and more than half (55%) of unintended pregnancies end in unsafe abortions.⁵ While early childbearing has decreased overall, the rate of change has varied across countries and regions. It is important to note that the vast majority (90%) of births to adolescent girls aged 15 to 19 occur in LMICs.⁶

In a global context, adolescent pregnancy can increase the risks of various adverse outcomes, including infections⁷ and complications such as obstetric fistula,⁸ preeclampsia/eclampsia, low birth weight, preterm birth,⁹ and postpartum hemorrhage¹⁰ because the bodies of young adolescents may not be physically or biologically mature enough to handle pregnancy and childbirth.^4,11,12 According to previous studies, mothers aged 10 to 19 have a significantly higher risk (6.7%) of developing preeclampsia/eclampsia globally, and the risk becomes nearly double (11.5%) in LMICs.¹³ Similarly, adolescent pregnancy is associated with postpartum hemorrhage.⁷ Adolescent pregnancy remains a leading cause of maternal mortality.^1,14 According to 2023 data, maternal conditions were among the top five causes of morbidity and mortality globally in adolescent girls aged 15 to 19 years.⁴ Newborns are also at risk; studies show that infants born to adolescent mothers have a higher risk of mortality.¹⁵ Furthermore, premature membrane rupture, cephalo-pelvic disproportion, and low birth weight were among the adverse birth outcomes associated with adolescent pregnancy.^6,16–18 In addition to the direct health risks, pregnancy during adolescence has socioeconomic repercussions, including school dropout, drug abuse, unemployment, poverty, and intimate partner violence.^6,19

A concerted and coordinated effort is required in LMICs to achieve the Sustainable Development Goals (SDGs), particularly Goals 3, 4, and 5, with a focus on the sexual and reproductive health and rights of adolescents. Goal 3 specifically aims to ensure healthy lives and promote well-being for all individuals from birth onward. This broad goal encompasses critical health-related targets, including reproductive, maternal, neonatal, child, and adolescent health.²⁰ Adolescents are disproportionately affected by sexual violence, especially adolescents who live in rural areas and have lower educational levels.^21,22 To effectively prevent the risks that expose adolescents to pregnancy and unsafe abortions, interventions must be grounded in reliable evidence. This is essential to achieving Goal 5, which seeks to empower all women and girls and attain gender equality.²³ Adolescent pregnancy is a key issue that the World Health Organization (WHO) is actively addressing. In collaboration with partners, WHO has been raising awareness of the problem, establishing evidence-based actions, developing program tools and supportive policies, building capacity, and assisting countries in tackling adolescent pregnancy.⁶ WHO also prioritizes adolescent health by preparing guidelines, expanding comprehensive sexuality education, and emphasizing sexual and reproductive health research for the coming years to prevent adolescent pregnancy and its associated consequences.²⁴ Furthermore, providing comprehensive sexuality education to all young people, especially adolescent girls and young women, is crucial for preventing unintended pregnancies.²³

Adolescent girls aged 10 to 19 have a significantly higher proportion of unmet needs related to pregnancy prevention compared to all women of reproductive age.⁵ They also face unmet needs in sexuality education and have limited access to desired healthcare services. These factors increase their vulnerability to child marriage and adolescent pregnancy.^5,25 Reducing adolescent pregnancies and preventing pregnancy-related mortality and morbidity are essential for achieving the SDGs concerning maternal and newborn health, as well as for promoting optimal health outcomes throughout life.⁶

Early marriage, poverty, social norms, cultural beliefs, and lack of education contribute to the high prevalence of adolescent pregnancy.^26,27 It is essential to break the cycle of adolescent pregnancies by empowering girls with access to accurate information, enabling them to make informed decisions about when to have children and their reproductive futures. However, there is limited evidence regarding global and regional coverage, as well as the factors contributing to adolescent pregnancy. Identifying the magnitude, understanding the contributing factors, and developing prevention strategies for adolescent pregnancy are crucial. To support young mothers and prevent future adolescent pregnancies by addressing these contributing factors, this study aimed to assess the proportion and associated factors of adolescent pregnancy in LMICs, as well as to pinpoint target areas for intervention. Our study provides nuanced, policy-relevant evidence by focusing on demographic and health survey (DHS)-participating LMICs.

Methods and materials

Study area and data source

A community-based cross-sectional study design was employed. Data were obtained from six WHO regions, as stated in Table 1, classified according to the DHS, encompassing LMICs from 2015 to 2024. A total of 47 eligible countries were selected for analysis. When more than two standard DHS datasets were available after 2015, the most recent dataset was used. The surveys used a multistage cluster sample and were created to collect and present data on demographic and key health indicators. The DHS program utilizes standardized model questionnaires that are developed through a rigorous protocol. According to the guidelines provided in the DHS record manuals (DHS-7 and DHS-8), the women’s questionnaire is reviewed by a technical advisory group. It is then adapted to fit the local context, translated, and must undergo pre-testing in the field to ensure its validity and reliability before the main survey.^28,29 Each survey employed a cross-sectional study design, collecting data on both the independent and outcome variables simultaneously. LMICs represent diverse nations and are categorized as low-income, lower-middle-income, and upper-middle-income countries, based on the World Bank’s 2025 classification. The analysis included 13 low-income, 26 lower-middle-income, and 8 upper-middle-income countries.³⁰ Among these, 28 countries were from sub-Saharan Africa (SSA), 11 countries from South and Southeast Asia, 4 countries from North Africa/West Asia/Europe, 2 countries from Latin America and the Caribbean, and 1 country each from Central Asia and Oceania. The data were accessed online via the link https://dhsprogram.com/data/dataset_admin/index.cfm after a formal request.

Table 1.

The weighted frequencies of the surveys for the study of adolescent pregnancy in 47 low- and middle-income countries, demographic and health surveys from 2015 to 2024.

S. No.	Countries of LMICs with different regions	Country code and DHS round	Survey years	Frequency	Weighted frequency
01	Sub-Saharan Africa			115,626	113,928
1	Angola	AO7	2015–2016	4,053	4,079
2	Benin	BJ7	2017–2018	4,102	4,118
3	Burkina Faso	BF8	2021	4,572	4,587
4	Burundi	BU7	2016–2017	4,790	4,681
5	Cameroon	CM7	2018	3,968	3,920
6	Cote d’Ivoire	CI8	2021	3,754	3,867
7	Ethiopia	ET7	2016	4,387	4,176
8	Gabon	GA7	2019–2021	2,234	2,074
9	Gambia	GM8	2019–2020	3,234	3,162
10	Ghana	GH8	2022	3,428	3,262
11	Guinea	GN7	2018	3,118	3,176
12	Kenya	KE8	2022	7,608	7,233
13	Lesotho	LS8	2023–2024	1,586	1,488
14	Liberia	LB7	2019–2020	2,065	1,984
15	Madagascar	MD8	2021	5,181	5,198
16	Malawi	MW7	2015–2016	6,399	6,442
17	Mali	ML7	2018	2,799	2,649
18	Mauritania	MR7	2019–2021	4,343	4,399
19	Mozambique	MZ8	2022–2023	3,731	3,698
20	Nigeria	NG7	2018	10,466	10,539
21	Rwanda	RW8	2019–2020	3,862	3,806
22	Senegal	SN8	2023	4,882	4,654
23	Sierra Leone	SL7	2019	4,133	4,081
24	South Africa	ZA7	2016	1,799	1,705
25	Tanzania	TZ8	2022	3,673	3,613
26	Uganda	UG7	2016	5,170	5,147
27	Zambia	ZM7	2018	3,744	3,627
28	Zimbabwe	ZW7	2015	2,545	2,563
02	North Africa/West Asia/Europe			4,421	4,519
29	Albania	AL7	2017–2018	1,969	1,976
30	Armenia	AM7	2015–2016	925	895
31	Jordan	JO8	2023	331	291
32	Turkey	TR7	2018	1,196	1,357
03	Central Asia			2,316	2,350
33	Tajikistan	TJ7	2017	2,316	2,350
04	South and Southeast Asia			186,986	185,828
34	Afghanistan	AF7	2015	3,550	3,453
35	Bangladesh	BD8	2022	3,433	3,591
36	Cambodia	KH8	2021–2022	3,642	3,524
37	India	IA7	2019–2021	148,541	148,484
38	Indonesia	ID7	2017	9,333	8,914
39	Maldives	MV7	2016–2017	1,216	1,371
40	Myanmar	MM7	2015–2016	2,220	2,205
41	Nepal	NP8	2022	3,406	3,280
42	Pakistan	PK7	2017–2018	1,148	933
43	Philippines	PH8	2022	6,897	6,606
44	Timor-Leste	TL7	2016	3,600	3,467
05	Oceania			3,588	3,604
45	Papua New Guinea	PG7	2016–2018	3,588	3,604
06	Latin America and the Caribbean			14,457	13,536
46	Colombia	CO7	2015	10,539	9,787
47	Haiti	HT7	2016–2017	3,918	3,749
Total				327,394	323,767

LMICs: Low- and middle-income countries.

To ensure consistency across countries and survey rounds, adolescent pregnancy and its associated reproductive, demographic, and socioeconomic factors were assessed and classified using the standard DHS definitions and coding from the DHS7 and DHS8 Recode Manuals.^28,29

Population

The target population comprised all adolescent girls aged 15 to 19 in LMICs, while the study population included all adolescent girls aged 15 to 19 in 47 selected LMICs.

Inclusion criteria and exclusion criteria

This study included all recorded adolescent girls aged 15 to 19 in the selected LMICs. Participants with missing or unknown values for adolescent birth, history of termination, or pregnancy within 5 years of data collection in the given age range were excluded.

Sample size determination and sampling procedure

Since this study was based on secondary data, we did not calculate the sample size. The sample includes all recorded adolescent girls who gave birth, had a history of terminated pregnancy, or were pregnant before age 20 within 5 years of data collection. Therefore, according to Table 1, the current study involved 327,394 participants, which, after weighting, was adjusted to 323,767. Of these, 45.86% were from India (weighted). The countries’ sample sizes, which constitute the overall sample size, are presented in Table 1.

Study variables

Dependent variable: Adolescent pregnancy, also known as teenage pregnancy, is an individual giving birth before the age of 20 years.¹¹ It is based on self-reported birth history and is categorized as yes (coded 1) if the aggregate of v212 (age at first birth), v213 (currently pregnant), an v228 (history of terminated pregnancy) had codes of an adolescent who gave birth, had a history of terminated pregnancy, or was pregnant during and/or within 5 years of data collection, and no (coded 0) otherwise. Adolescent pregnancy in DHS is assessed through self-reported reproductive histories, including age at first birth (v212), current pregnancy status (v213), and terminated pregnancies (v228). While DHS applies standard and widely accepted data collection methods, measures could be susceptible to reporting bias. Where early marriage is legally prohibited, reporters might over-report their age at marriage or first birth. Also, within settings where pregnancy during adolescence is being stigmatized, there may be unrecorded or underreported pregnancies that end in unsafe abortion.

Independent variables

Individual-level factors: Assessed for their association with the outcome variable included the respondent’s educational attainment (no education, primary, secondary, and above), wealth index (poor, middle, rich), respondent’s age (15–17, 18–19), sex of household head (male, female), condom use (yes, no), contraceptive use (yes, no), mobile phone ownership (has, does not have), bank account ownership (has, does not have), and knowledge of contraception methods (knows, no knowledge).

Community-level variables included WHO region (North Africa/West Asia/Europe, SSA, Central Asia, South and Southeast Asia, Oceania, Latin America, and the Caribbean), residence (urban or rural), and media exposure (none or media exposure).

Data collection procedure and quality

Data collection procedure

The research utilized data from the DHS, collected using a standardized women’s questionnaire that included multiple sections, some of which were specifically tailored for women and households. Trained data collectors conducted face-to-face interviews at participants’ homes with eligible women aged 15 to 19 who had given birth within the past 5 years. The questionnaire included women’s health characteristics within sociodemographic, socioeconomic, and pregnancy-related contexts.³¹

Data quality control

The goal of the DHS is to generate high-quality, easily accessible data consistently and systematically. However, national surveys in developing countries often encounter issues such as omitted or partially reported responses. Additionally, complex questions frequently yield inconsistent answers, resulting in incomplete or unreliable datasets that complicate analysis.³² To address these challenges, the DHS employs editing and imputation techniques to ensure the dataset is suitable for research and accurately represents the population.^32,33

Data management, processing, and analysis

The data extracted from DHS datasets, available online, were collected from 47 LMICs between 2015 and 2024. The data were cleaned, recoded, and analyzed using STATA version 17. Following the survey report’s recommendations, the data were weighted for sampling probabilities and non-response using the weighting factor to restore representativeness and produce reliable statistical estimates before analysis. All analyses applied DHS sample weights (v005) and cluster numbers (v001) to account for the complex survey design. Consequently, the proportions reported represent nationally weighted estimates for female adolescents. Multicollinearity was validated for independent variables and assessed using variance inflation factors (VIFs), revealing no significant issues. Bivariable multilevel logistic regression analyses with a p-value < 0.20 were included in the final model. Multivariable multilevel logistic regression was employed for the outcome variable. The p-values less than 0.05 were considered statistically significant, and adjusted odds ratios (AORs) with 95% CIs were calculated from the final model. These AORs and their 95% CIs served as measures of association in the full model, which examined the relationship between adolescent pregnancy and independent factors at individual and community levels. Assumptions of chi-square tests were validated and fulfilled.

We obtained statistically accurate estimates of the regression coefficients by incorporating clustering information. Before appending the datasets, cluster numbers were recoded sequentially, starting from one, to prevent overlapping of cluster identifiers across multiple countries. Therefore, a multilevel analysis was performed to estimate mixed effects, including fixed effects for predictors at both the individual and community levels, and a random effect to capture variation between clusters. The likelihood ratio test was employed to assess the significance of the random intercept variation. Additionally, the intra-class correlation coefficient (ICC), proportional change in variance (PCV), and median odds ratio (MOR) were used to quantify the variation attributable to the random effects.

To determine the overall variation at each model attributable to individual and/or community-level characteristics, we employed PCV. PCV was calculated using the formula PCV = ((VA−VB)/VA) × 100, where VA represents the variance of the empty model and VB represents the variance of the model with additional components. The candidate model was evaluated using the Deviance Information Criterion (DIC), calculated as deviance = −2 × log-likelihood.

Model building and comparison

Four models incorporating variables of interest were fitted to identify the factors contributing to adolescent pregnancy. Without explanatory variables, the null model was fitted to test for random variation in the intercept and to estimate ICC. The ICC is a measure of variation within a cluster. Model I examined how individual-level characteristics affect adolescent pregnancy. Model II analyzed the factors at the community level that influence the outcome. Model III (the full model) analyzed how individual and community-level factors jointly impact the outcome of interest. The PCV, which represents the change in community-level variance between the empty model and subsequent models, and ICC, which indicates the percentage of variance explained by the higher level, were reported as measures of variation for community-level variables.

Model equation

Let P(Y_ij = 1) be the probability of adolescent pregnancy for individual I in cluster j. Then, the multilevel logistic regression model can be expressed as:

Logit (P (Y_{i j} = 1)) = β_{0} + \sum_{p = 1}^{P} β_{p} Χ_{P i j} + \sum_{q = 1}^{Q} β γ_{q} Z_{q j} + u_{0 j}

where Y_ij is the binary outcome variable (1 if adolescent $i$ in cluster j is pregnant, 0 otherwise); Logit(P(Y_ij = 1)) is the log-odds of adolescent pregnancy, and this is the link function for logistic regression, transforming the probability into a linear scale; β₀ is the overall intercept; Χ_pij is the individual-level variables for the individual $i$ in cluster j; β_p is the fixed effect coefficient for the individual-level predictors; Z_qj is the cluster-level predictor variables for cluster j; γ_q is the fixed effect coefficients for the cluster-level predictors; and u_0j is the random intercept for cluster j.

Results

Sociodemographic characteristics

Table 2 reveals that a total of 327,394 (weighted 323,767) adolescents were included in the current study. Most participants, 240,321 (74.23%), had secondary education and higher. Nearly two-thirds of respondents, 202,236 (62.46%), had media exposure; they were exposed to media at least once a week (radio, newspapers, or television). Out of all participants, two-fifths, 131,309 (40.56%), of women had poor wealth status. Among all study participants, nearly two-thirds, 205,823 (63.57%), were rural dwellers. Almost three-fifths, 185,829 (57.40%), of women were from the South and Southeast Asian region (see Table 2)

Table 2.

Sociodemographic characteristics of adolescent age (15–19 years) women who experienced adolescent pregnancy: evidence from recent surveys in low- and middle-income countries demographic health surveys from 2015 to 2024.

Variables	Category	Adolescent pregnancy		Subtotal (%)
Variables	Category	No (%)	Yes (%)	Subtotal (%)
Women’s education	No formal education	17,273 (52.77)	12,69 (47.23)	29,342 (9.06)
	Primary Education	37,571 (69.44)	16,534 (30.56)	54,105 (16.71)
	Secondary and higher education	210,282 (87.50)	30,038 (13.04)	240,320 (74.23)
Wealth index	Poor	100,883 (76.83)	30,426 (23.17)	131,309 (40.56)
	Middle	53,785 (81.03)	12,595 (18.97)	66,380 (20.50)
	Rich	110,457 (87.61)	15,620 (12.39)	126,078 (38.94)
Respondent’s current age	<18 years	151,015 (94.10)	9,462 (5.90)	160,477 (49.57)
Respondent’s current age	18–19 years	114,111 (69.88)	49,180 (30.12)	163,290 (50.43)
Contraceptive use	No	251,128 (85.58)	42,313 (14.42)	293,441 (90.63)
Contraceptive use	Yes	13,997 (46.15)	16,329 (53.85)	30,326 (9.37)
Sex of household head	Male	205,350 (81.70)	46,004 (18.30)	251,354 (77.63)
Sex of household head	Female	59,776 (82.55)	12,637 (17.45)	72,413 (22.37)
Condom use	No	257,143 (82.08)	56,129 (17.92)	313,272 (96.76)
Condom use	Yes	7,982 (76.05)	2,513 (23.95)	10,495 (3.24)
Knowledge of contraception	Has no knowledge	23,496 (87.08)	3,487 (12.92)	26,983 (8.33)
Knowledge of contraception	Has knowledge	241,629 (81.42)	55,155 (18.58)	296,784 (91.67)
Marital status	Unmarried/separated	243,413 (95.40)	11,728 (4.60)	255,141 (78.80)
Marital status	Married	21,713 (31.64)	50,564 (68.36)	68,626 (21.20)
Having a bank account	Has no	243,841 (81.44)	55,559 (18.56)	299,400 (92.47)
Having a bank account	Has	21,284 (87.35)	3,083 (12.65)	24,367 (7.53)
WHO regions	SSA	81,695 (71.71)	32,234 (28.29)	113,929 (35.19)
	North Africa/West Asia/Europe	4,185 (92.61)	334 (7.39)	4519 (1.40)
	Central Asia	2,031 (86.40)	320 (13.60)	2,350 (0.73)
	South and Southeast Asia	163,023 (87.73)	22,806 (12.27)	185,829 (57.40)
	Oceania	2,932 (81.35)	672 (18.65)	3,604 (1.11)
	Latin America/Caribbean	11,260 (83.19)	2,276 (16.81)	13,536 (4.18)
Media exposure	Has no media exposure	92,185 (75.85)	29,345 (24.15)	121,530 (37.54)
Media exposure	Has media exposure	172,940 (85.51)	29,297 (14.49)	202,237 (62.46)
Place of residence	Urban	101,873 (86.37)	16,071 (13.63)	117,944 (36.43)
Place of residence	Rural	163,252 (79.32)	42,571 (20.68)	205,823 (63.57)

The magnitude of adolescent pregnancy

The pooled proportion of adolescent pregnancy is 23% (20–26%). Across regions, it ranges from 10% (6–14%) in North Africa/West Asia/Europe to 28% (25–32%) in SSA. The proportion across countries also ranges from 4% (3–5%) in the Maldives to 45% (43–47%) in Mali (see Figure 1).

Figure 1.

The pooled proportion of adolescent pregnancy in 47 low- and middle-income countries, as evidenced from DHS 2015–2024.

The prevalence of adolescent pregnancy in several countries which have higher than 30% was as follows, as shown in Figure 1, Mali (45%), Mozambique (43%), Angola (42%), Madagascar (41%), Liberia (40%), Malawi (39%), Zambia (36%), Bangladesh (34%), Guinea (34%), Uganda (33%), Pakistan (32%), Cameroon (31%), Benin (30%), Cote d’Ivoire (29%), Sierra Leone (29%), Nigeria (28%), Tanzania (28%), Zimbabwe (28%), and Jordan 27%. There are also countries with a proportion of 10% or less: the Maldives (4%), Albania (5%), Turkey (6%), Armenia (7%), the Philippines (7%), and Rwanda (9%), Indonesia (10%), Myanmar (10%), and Timor-Leste (10%).

Factors associated with adolescent pregnancy

With a p-value of less than 0.2 in the bivariable analysis, all variables for both (individual and community level) factors were included in the multivariable multilevel analysis. In the final model, both the individual-and community-level factors were included simultaneously; women’s education, respondent’s age, household wealth index, sex of household head, contraceptive use, condom use, marital status, having a bank account, and knowledge of contraception were individual-level factors significantly associated with adolescent pregnancy.

Table 3 shows that women who had no formal education increased the likelihood of adolescent pregnancy 1.10 times (AOR = 1.10, 95% CI: [1.04, 1.16]), while women who had primary education increased the likelihood of adolescent pregnancy by 58% (AOR = 1.58, 95% CI: [1.51, 1.66]) compared with women who had secondary and higher education. The odds of adolescent pregnancy decreased by 17% (AOR = 0.83, 95% CI: [0.80, 0.87]) among women from middle-income families and by 42% (AOR = 0.58, 95% CI: [0.55, 0.61]) among women from wealthy households, compared to those from low-income households. The odds of adolescent pregnancy were increased more than four times (AOR = 4.33, 95% CI: [4.17, 4.49]) among women aged between 18 and 20 compared with those less than 18 years of age. Similarly, the likelihood of adolescent pregnancy was increased by 40% (AOR = 1.40, 95% CI: [1.35, 1.46]) among female household heads. The odds of adolescent pregnancy were reduced by 56% (AOR = 0.44, 95% CI: [0.42, 0.46]) among participants who used contraception, and the likelihood of adolescent pregnancy decreased to 0.38 (AOR = 0.38, 95% CI: [0.35, 0.41]) among individuals who knew contraceptive methods. Among condom-using individuals, the odds of adolescent pregnancy decreased by 21% (AOR = 0.79, 95% CI: [0.74, 0.85]). The odds of adolescent pregnancy among married women increased 9 times (AOR = 9.02, 95% CI: [8.91, 9.73]).

Table 3.

Multilevel analysis of factors associated with adolescent pregnancy in adolescent women 15 to 19 years in 47 low- and middle-income countries: based on demographic health surveys from 2015 to 2024.

Variables	Category	Model I AOR (95% CI)	Model II AOR (95% CI)	Model III AOR (95% CI)
Women’s education	No education	1.92 [1.82, 2.03]	—	1.10 [1.04, 1.16]**
	Primary education	2.70 [2.59, 2.81]	—	1.58 [1.51, 1.66]***
	Secondary and higher education	1	—	1
Wealth index	Poor	1	—	1
	Middle	0.92 [0.88, 0.96]	—	0.83 [0.80, 0.87]***
	Rich	0.71 [0.68, 0.74]	—	0.58 [0.55, 0.61]***
Respondent’s age	15–17 years	1	—	1
Respondent’s age	18–19 years	3.75 [3.62, 3.90]	—	4.33 [4.17, 4.49]***
Sex of household head	Male	1	—	1
Sex of household head	Female	1.66 [1.60, 1.72]	—	1.40 [1.35, 1.46]***
Contraception use	No	1	—	1
Contraception use	Yes	0.40 [0.38, 0.42]	—	0.44 [0.42, 0.46]***
Condom use	Not used a condom	1	—	1
Condom use	Condom used	1.06 [0.98, 1.15]	—	0.79 [0.74, 0.85]***
Marital status	Not married	1	—	1
Marital status	Married	12.85 [9.24, 14.41]	—	9.02 [8.91, 9.73]***
Having a bank account	No	1	—	1
Having a bank account	Yes	0.56 [0.52, 0.60]	—	0.89 [0.83, 0.96]**
The respondent owns a mobile phone	No	1	—	1
The respondent owns a mobile phone	Yes	1.20 [1.16, 1.26]	—	1.01 [0.98, 1.07]
Knowledge of contraception	Does not know	1	—	1
Knowledge of contraception	Know	0.56 [0.52, 0.60]	—	0.38 [0.35, 0.41]***
WHO region	North Africa/West Asia/Europe	—	1	1
	SSA	—	5.28 [4.59, 6.08]	8.79 [6.96, 10.24]***
	Central Asia	—	1.71 [1.36, 2.15]	2.96 [2.16, 3.67]***
	South and Southeast Asia	—	1.36 [1.18, 1.56]	1.65 [1.35, 2.03]***
	Oceania	—	2.15 [1.78, 2.61]	4.11 [2.66, 5.01]***
	Latin America/Caribbean	—	3.01 [2.58, 3.51]	5.83 [4.93, 6.49]***
Media exposure	Has no exposure	—	1	1
Media exposure	Has exposure	—	0.78 [0.75, 0.81]	0.86 [0.83, 0.89]**
Place of residence	Urban	—	1	1
Place of residence	Rural	—	2.07 [1.99, 2.13]	1.01 [0.97, 1.07]
Statistic	Null model	—	—	—
AIC	284,541.7	148,939.1	276,096.5	141,043.7
BIC	284,563.1	149,110.2	27,6182.1	141,279.1
Cluster-level variance	1.423794	1.215778	1.011827	1.142183
ICC	0.3020569	0.2731721	0.2360279	0.2533126
MOR	3.12	3.01	2.62	2.97
PCV	Reference	14.61%	28.61%	19.78%
Deviance	2,476.08	1,231.824	1,549.37	932.436
Mean VIF	—	1.13	1.03	1.18

CI: confidence interval; ICC: intra-class correlation coefficient; MOR: median odds ratio; PCV: proportional change in variance; SSA: sub-Saharan Africa; VIF: variance inflation factors

P-value < 0.05. **P-value < 0.01. ***P-value < 0.001.

Additionally, the likelihood of teenage pregnancy decreased by 11% (AOR = 0.89, 95% CI: [0.83, 0.96]) among women who had bank accounts.

Regarding community-level factors in the final model analysis, the WHO region and media exposure were significant predictors of adolescent pregnancy. In the SSA and Central Asia regions, the odds of adolescent pregnancy were 8.79 times (AOR = 8.79, 95% CI: [6.96, 10.24]) and 2.96 times (AOR = 2.96, 95% CI: [2.16, 3.67]) higher, respectively, compared to the North Africa/West Asia/Europe region. Likewise, the likelihood of adolescent pregnancy rose by 1.65 times (AOR = 1.65, 95% CI: [1.35, 2.03]) in South and Southeast Asia, and by 4.11 times (AOR = 4.11, 95% CI: [2.66, 5.01]) in Oceania, relative to the North Africa/West Asia/Europe region.

The likelihood of adolescent pregnancy was 5.83 times (AOR = 5.83, 95% CI: [4.93, 6.49]) in Latin America and the Caribbean region. Among individuals who had media exposure, the likelihood of adolescent pregnancy was decreased to 14% (AOR = 0.86, 95% CI: [0.83, 0.89]) compared with those who had not.

Random effects and model comparison

According to Table 3, in the empty model, the intra-enumeration area correlation coefficient (ICC) revealed that variations between cluster areas account for 30.21% (ICC = 0.3020569) of the overall variation in adolescent pregnancy, with individual differences accounting for the remaining 69.79% of the variability. In the final model (Model III), the total variation in adolescent pregnancy at the cluster level was reduced to 25.33% (ICC = 0.2533126) and may be attributable to other unobserved community-level factors. Model fitness was compared using deviance, and the model with the lowest deviance (Model III) was considered the best-fitting model. The final model (Model III), which has the lowest deviance value of 932.436 compared to the other models, is utilized to determine significant factors associated with adolescent pregnancy among mothers aged 15 to 19 in the DHS data from 47 LMICs. The effects of community heterogeneity were also minimal, as evidenced by the lowest MOR value of Model III (2.97), which was lower than the MOR of the null model (3.12). According to the PCV, the complete model, which incorporated factors at both the individual and community levels, accounted for 25.34% of the differences in adolescent pregnancy across communities. Additionally, the average VIF across all models was very low (<10), indicating that there was no multicollinearity among the predictor variables (see Table 3).

Discussion

Adolescent pregnancy remains a significant global public health issue. Approximately 21 million girls aged 15 to 19 become pregnant every year, with nearly half of these pregnancies being unintended. More than half (55%) of unintended pregnancies end in abortion.⁵

The pooled proportion of adolescent pregnancy among all adolescent girls in 47 LMICs was 23% (95% CI: [20, 26]) with $I^{2}$ > 99%. By region, the percentage was greater in SSA at 28% (95% CI: [25, 32]), whereas North Africa/West Asia/Europe had a lower percentage of 10% (95% CI: [6, 14]), which restricts the generalizability of the findings. There was notable variation among countries, with proportions ranging from 4% (95% CI: [3, 5]) in the Maldives and 5% (95% CI: [5, 7]) in Albania to as high as 45% (95% CI: [43, 47]) in Mali. This heterogeneity demonstrates that adolescent pregnancy is influenced by multiple social, cultural, structural, knowledge of reproductive and sexual health, peer influence, and economic factors.³⁴ In several countries, the percentage was notably above the pooled average, including Mali, Mozambique, Angola, Madagascar, Liberia, Malawi, Zambia, Bangladesh, Guinea, Uganda, Pakistan, Cameroon, Benin, Côte d’Ivoire, Sierra Leone, Nigeria, Tanzania, Zimbabwe, and Jordan. The pooled proportion is extremely higher than the estimated global average in 2023 by the United Nations, which was 4.13% (41.3 births per 1,000 girls).³⁵ There was a greater discrepancy in the proportion of adolescent pregnancy between low-income and high-income countries. According to some recent studies, the proportion in high-income countries was below 5%.³⁶

Previous evidence indicates that child marriage rates in most SSA countries were higher,^37,38 which might contribute to the increased prevalence of adolescent pregnancy in the region. Child marriage in countries such as Cote d’Ivoire, Guinea, Liberia, Mali, Mauritania, Nigeria, Sierra Leone, Uganda, and most SSA countries continues to be high.^39,40 According to the findings of this study, which was supported by other studies,⁴¹ married adolescents were nine times more likely to be pregnant compared to unmarried ones. Modern contraceptive coverage in SSA countries is still extremely low,⁴² and some studies revealed that in most countries in the region, adolescent women have unmet needs for contraception.^43,44 Additionally, this study included all adolescent girls who had a birth, a history of terminated pregnancy, or who were pregnant during data collection, which could contribute to the increased reported rates of adolescent pregnancy. However, underreporting of terminated pregnancy and over-reporting of age at first marriage for various reasons are still possible in the majority of restrictive settings.⁴⁵ In Bangladesh, Pakistan, and Jordan, the individual record datasets included only ever-married women, unlike in other countries. This could contribute to the increased adolescent pregnancy rate. In South Asia, perspectives on gender have mostly stayed unchanged. In most developing countries, a lack of awareness among adolescents about sexual and reproductive health, social and cultural influences, family beliefs, and low educational and financial status remain significant drivers of adolescent pregnancy.³⁴ Despite improvements in social indicators, gender inequality continues to be an important concern in South Asia,⁴⁶ which may contribute to the substantially higher prevalence of adolescent pregnancy in these countries. Studies in Bangladesh and South Asia revealed a significantly high rate of adolescent marriage,^47,48 which could be linked to the increased prevalence of adolescent pregnancy. In Pakistan, earlier studies showed that adolescent births accounted for 42.5%,⁴⁹ and family planning coverage was below 50%.⁵⁰ This low contraceptive utilization indicates an unmet need for such services among women, particularly adolescents. In Jordan, the higher prevalence of adolescent pregnancy may be attributed to the inclusion of refugees in the data, whether formally residing in refugee camps or informally living in communities outside camps.⁵¹

It was 10% or less in some countries, such as the Maldives, Albania, Armenia, Turkey, the Philippines, Rwanda, Indonesia, Myanmar, and Timor-Leste. An explanation for the lower proportion in these countries might be the inclusion of all unmarried and ever-married adolescents. This proportion is also likely linked to stronger investments in education, reproductive health services, and different gender norms.⁵² Child marriage in some countries, such as Albania, Rwanda, Turkey, Indonesia, Myanmar, the Philippines, and Timor-Leste, was relatively lower.^39,41,48 The increase in women’s ability to choose marriage and make decisions within the family in Albania⁵² might contribute to lowering the rate of adolescent births.

The pooled proportion of adolescent pregnancy in the current study was higher than the findings from Ethiopia, Ghana, SSA, and a systematic review of global studies.^9,53–55 One possible explanation for this discrepancy is demographic differences. For instance, whereas a previous study was conducted in a single town using a cross-sectional method, our research analyzed combined data from several countries, including some of the most marginalized populations globally. This broader scope may have led to a higher overall rate of adolescent pregnancy in our analysis. Adolescent pregnancy rates are strongly influenced by socioeconomic status and tend to be relatively low in most developed countries.^36,56 Conversely, our findings are lower than those reported in studies from East Africa, Ethiopia, Tanzania, and Zambia.^56–60 However, some previous studies were based on single cross-sectional designs,^58,59 making it difficult to directly compare a single snapshot study with the pooled proportion derived from global data. For example, a study in Ethiopia was conducted in a refugee camp,⁶⁰ which may have elevated the prevalence due to higher risks of sexual violence and unmet needs of contraception in such settings.^61,62 Furthermore, the East African study used only women who had ever had sex as the denominator, unlike our study, which could have inflated the reported prevalence.⁵⁶

Adolescent pregnancy is consistently associated with substantial adverse maternal and neonatal outcomes, and these risks are amplified in LMICs where maternal healthcare services are poor and with considerable disparities across countries.^15,63,64 Evidence shows that adolescent mothers face a higher risk of obstetric complications, including hypertensive disorders, anemia,^9,65 obstructed or prolonged labor due to cephalo-pelvic disproportion, postpartum hemorrhage,¹⁰ and a greater likelihood of requiring cesarean section (CS) delivery.^65–67 CS is a common incidence in adolescents, due to some indications, such as non-reassuring fetal status and labor dystocia.⁶⁸ Prolonged or obstructed labor can lead to CS delivery in adolescents.⁶⁹ CS delivery is currently on the rise in all parts of the world.⁷⁰ Neonatal risks are also elevated, with increased risks of stillbirth, preterm birth,^9,65 low birth weight,^7,65 birth trauma,⁷¹ and neonatal mortality.¹⁵ Systemic challenges, such as limited access to quality antenatal and intrapartum care and delays in obtaining emergency obstetric services, combined with broader structural and socioeconomic constraints in LMICs, significantly worsen maternal and neonatal outcomes among adolescents.^72,73 Recent studies in LMICs reported that adolescents experienced nearly a fourfold higher risk of stillbirth and a twofold increase in neonatal mortality compared to adult mothers.¹⁵ Adolescent pregnancy also leads to higher risks of maternal and neonatal mortality,^4,15 a greater likelihood of mental health problems, and constraints for educational and economic prospects, which contribute to cycles of poverty and inequality that can span across generations.^6,19 Its impact extends beyond the mother and her child to affect the entire society.^11,49

Several socioeconomic and sociodemographic factors were found to be significantly associated with adolescent pregnancy, such as lower educational attainment, being married, the respondent’s age, living in a female-headed household, and WHO regions, which were positively associated with adolescent pregnancy. Conversely, higher wealth index, media exposure, contraceptive use, knowledge of contraception, condom use, and owning a bank account were negatively associated with the outcome. As the education level increased, the likelihood of adolescent pregnancy decreased, and vice versa. This finding aligns with studies conducted in Tanzania, Zambia, and South Asia.^57,58,74 Education level is a critical factor influencing adolescent fertility.⁷⁵ Findings from a systematic review in LMICs are consistent with our results, showing that higher educational attainment protects adolescents from early pregnancy.⁷⁶ Adolescent girls who remain in education longer have a greater chance of delaying pregnancy.^49,77 Conversely, lack of formal education, inadequate sexuality education, and limited knowledge significantly increase the risk of unplanned early pregnancy.^78,79 This finding aligns with studies conducted in Tanzania, Pakistan, and South Asia, which show that wealth status is significantly associated with adolescent pregnancy.^49,57,58,74 LMICs are often characterized by low socio-economic communities.⁸⁰ Poverty is a critical factor contributing to adolescent pregnancy, particularly in these countries.^77,81 Low educational attainment, unemployment, and poverty often come together.³⁴ Addressing the impact of poverty and empowering adolescent women are essential strategies to reduce adolescent pregnancy.⁸² Similarly, in the current study, having a bank account among adolescents was associated with a decreased likelihood of adolescent pregnancy. Apart from DHS data, no other studies have examined this variable. Beyond reducing the risk of early motherhood, economic empowerment opportunities can positively influence their lives.¹¹

Consistent with our findings, studies conducted in Zambia have shown that women with regular access to media have lower odds of adolescent pregnancy.⁵⁷ Media access is an effective tool for reducing adolescent pregnancy,⁸³ as it facilitates sexuality and reproductive health education, enabling young women to protect themselves from risky behaviors.⁵⁷ Adolescents exposed to important information about adolescent pregnancy are better equipped to prevent unplanned pregnancies at an early age.^84,85 The media serves as a powerful tool for promoting sexual and reproductive health, significantly influencing adolescents’ health behaviors.^86–88 However, excessive exposure to social media can increase risky sexual behavior, leading to a higher incidence of unwanted pregnancies among adolescents.⁸⁹ In today’s digital world, adolescents must be selective about the information they consume and actively engage in sexual and reproductive education. Media can provide adolescents with valuable information about contraceptive use, thereby enhancing their utilization of effective contraceptive methods.⁸⁴

Place of residence was not a significant factor in the outcome of interest in this study. However, findings from previous studies conducted in Pakistan, Ghana, LMICs, and other regions, rural residency increased the likelihood of adolescent pregnancy.^2,49,81,90 Similarly, individuals residing in rural areas are more likely to face poverty and lack of formal education among adolescents and their parents, which may increase the likelihood of pregnancy.⁹¹ However, other studies conducted in Uganda and Tanzania have reported findings that urban residency increased the risk of adolescent births.^58,92

Contraception offers significant benefits by preventing unintended and unwanted pregnancies. This, in turn, reduces adverse perinatal outcomes, including maternal deaths and neonatal complications.^93,94 Contraceptive use can also decrease the likelihood of adolescent fertility.⁹⁵ These findings are consistent with a systematic review of studies conducted in LMICs and a cross-sectional study done in Cameroon.^2,79,82 This may be because adolescents often begin using contraception after becoming pregnant.

Additionally, poverty and limited resources can lead to the unavailability of contraceptives.³⁴ The availability of contraceptives plays a crucial role in reducing adolescent pregnancy rates.⁶⁶ However, availability alone is insufficient to address the issue. Therefore, knowledge of contraceptive use is an additional key predictor. Awareness of modern contraceptive methods increases their utilization, thereby reducing unplanned pregnancies at an early age.⁹⁶ Limited knowledge of contraception leads to lower service utilization among adolescents, resulting in unintended pregnancies. Effective knowledge of contraception can mitigate this risk by encouraging the use of available services.^97,98 Adolescent women who engage in sexual activity without proper information or access to contraception face a higher likelihood of pregnancy.¹¹ Physical, psychological, and social unmet needs, including unmet contraceptive needs, are particularly high in the developing world. Unmet contraceptive needs are a significant factor in LMICs.^78,99 Reports indicate a substantial gap between the demand for contraception and actual access in these regions. Addressing the unmet contraceptive needs of all women in LMICs could reduce unintended pregnancies by more than two-thirds and decrease unsafe abortions by 72%.⁵ Condom use is associated with a lower likelihood of adolescent fertility. Promoting correct condom use and dual protection, preventing both sexually transmitted infections and pregnancy, is an important strategy.^100,101 Condoms are an effective method to avoid adolescent conception, and ensuring private access to condoms may improve their utilization and consequently reduce adolescent fertility.^102,103

In our study, marital status emerged as a significant predictor of early childbearing, consistent with research conducted in Uganda, Ethiopia, and Nigeria.^59,92,104 The findings of a systematic review and meta-analysis in Africa support our results.⁵³ Additionally, increasing age was a strong predictor of adolescent pregnancy in this study. Adolescents aged 18 or older were more likely to experience an adolescent pregnancy compared to those younger than 18 years. Studies from Ethiopia and Uganda corroborate these findings.^59,92 This association may be explained by the fact that, in our study, 84% of married adolescents were 18 or older. Married adolescents have a higher likelihood of pregnancy due to increased exposure to sexual intercourse.

The sex of the household head is another significant factor in the current study. Living in a female-headed household increased the likelihood of adolescent pregnancy. This finding aligns with a study done in Malawi,¹⁰⁵ which may be attributed to the fact that female-headed households often face poverty. Additionally, women in these households typically spend most of their time caring for their children.^106,107 They may be less educated, and less educated women residing in rural areas may have restricted decision-making autonomy,¹⁰⁸ which could limit their life choices and could increase their vulnerability to child marriage and early pregnancy. However, this finding contradicts the findings of a study conducted in SSA.⁹⁰ A possible explanation is that adolescents living in male-headed households are more likely to be married.

Regarding WHO regions, all regions were found to have a higher likelihood of adolescent pregnancy compared to North Africa/West Asia/Europe. Most of the countries in those regions are low- and lower-middle-income countries, which were included in the current study, unlike the reference region, from which upper-middle-income countries were included.³⁰

Limitations of the study

Even though DHS data from 47 middle- and low-income countries provide a large, nationally representative, and standardized dataset that enables strong cross-country comparisons, the data used to determine adolescent pregnancy are based on self-reported reproductive histories, which are susceptible to recall bias, over-reporting of age at first birth in contexts where underreporting in restricted settings where adolescent pregnancy is stigmatized or leads to unsafe or illegal abortion. These factors likely decrease the report of adolescent pregnancy. The cross-sectional nature of the surveys limits causal inference, and these factors may result in underestimation or misclassification. Since we used secondary data, some important variables could not be included due to their absence or missing values, such as unmet needs for contraception, partners’ education, age at marriage, and other behavioral factors. Another limitation involves using combined data from 47 LMICs, which poses challenges due to significant regional and sociocultural variations, as reflected by the high heterogeneity ( $I^{2}$ >99%). Nevertheless, this issue was somewhat mitigated by adjusting the multilevel analysis to account for regional effects. Similarly, surveys cover nearly 10 years, which may have seen changes in settings and experiences. Furthermore, since secondary data were used in this study, the sample size was preset, and no power calculation was done beforehand. Our ability to detect associations for less frequent subgroups might have been limited. Finally, the interaction term between associated predictors was not tested. All predictors were categorical with multiple subcategories, which would have significantly increased model complexity and decreased interpretability when evaluating interaction terms.

Conclusion and recommendations

The pooled proportion of adolescent pregnancy is high in the study area with significant variability. The findings indicate that LMICs require a strong commitment to reduce this proportion to an acceptable level. Given the well-documented significant impact of adolescent pregnancy on maternal and child health, addressing this issue in developing countries demands careful attention. Factors positively associated with adolescent pregnancy include lower educational attainment, being married, and age under 18 years, living in female-headed households, and specific WHO regions. Conversely, a higher wealth index, media exposure, contraceptive use, condom use, knowledge of contraception, and owning a bank account are negatively associated with adolescent pregnancy.

Enhancing women’s education, increasing media exposure, improving knowledge and access to contraception, and promoting financial support for adolescent girls are identified as key modifiable factors to reduce adolescent pregnancy. The WHO regions, such as SSA, South and Southeast Asia, require urgent action to lower adolescent pregnancy rates. Significant efforts are essential to achieve the health-related SDGs by 2030. Countries including Mali, Mozambique, Angola, Madagascar, Liberia, Malawi, Zambia, Guinea, Uganda, Cameroon, Benin, Côte d’Ivoire, Sierra Leone, Nigeria, Tanzania, and Zimbabwe require particular attention due to their notably high rates of adolescent pregnancy.

To improve these modifiable factors, the involvement of multiple stakeholders is essential. To reduce adolescent pregnancy, the WHO and the health authorities of these LMICs must take immediate action, particularly in countries where rates exceed the pooled average. Nations in SSA and South and Southeast Asia are strongly encouraged to prioritize efforts to combat adolescent pregnancy. Health authorities, healthcare professionals, and educators should emphasize providing comprehensive sexuality and reproductive health education, especially through schools, mainstream media, and various social media platforms. Expanding community access to media and delivering health education to mothers and influential community leaders via targeted campaigns are effective strategies to enhance knowledge and awareness of the risks and consequences of adolescent pregnancy. For researchers, this study offers a valuable resource for conducting implementation research in diverse contexts within this field.

Footnotes

Acknowledgements

The authors thank the DHS program for allowing us to use online data.

ORCID iDs

Mequanent Dessie Bitewa

Thomas Kidanemariam Yewodiaw

Hiwot Tezera Endale

Mihret Getnet

Author contributions

MDB led in conceptualization, data cleaning, formal analysis, methodology development, software use, validation, writing, review, and visualization in drafting and revising the original manuscript. TKY, THE, and MG contributed to data cleaning, formal analysis, methodology, software use, manuscript drafting, writing, review, and editing. All authors reviewed and approved the final version of the manuscript.

Ethical considerations

The ICF Institutional Review Board and the health service ethical review committees of the included countries approved data collection for all DHS surveys. The benefits and risks of the surveys were explained to the participants. Before distributing the Women’s Questionnaire, eligible respondents provided written informed consent. For respondents with no formal education or limited literacy, the consent statement was read aloud in the local language by the interviewers, and verbal informed consent was documented by them. For adolescents under 18, DHS obtained written, and verbal informed consent from their legally authorized representatives (parents or guardians). The survey was entirely voluntary. The names and identification numbers of the respondents were not included in the final datasets. Additional ethical approval for this study was not required and was therefore waived/exempted. For additional information about data and ethical standards, visit .

Consent for publication

Not applicable.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

References

World Health Organization. Adolescent health, https://www.who.int/health-topics/adolescent-health/#tab=tab_1 (2024, accessed 23 June 2025).

Chung

Kim

Lee

J-E.

Comprehensive understanding of risk and protective factors related to adolescent pregnancy in low- and middle-income countries: a systematic review. J Adolesc 2018; 69: 180–188.

Innocent

Vasavada

Eneh

, et al. Systematic review evidence on the factors influencing teenage mothers utilization of maternal health services in Sub Saharan Africa. Acad J 2023; 38(2): 9–23.

UNICEF. Early childbearing. https://data.unicef.org/topic/child-health/early-childbearing/ (2024, accessed 23 June 2025).

Sully

Biddlecom

Darroch

, et al. Adding it up: investing in sexual and reproductive health 2019. https://www.guttmacher.org/report/adding-it-up-investing-in-sexual-reproductive-health-2019 (2020, accessed 26 June 2025).

World Health Organization. Adolescent pregnancy. https://www.who.int/news-room/fact-sheets/detail/adolescent-pregnancy (2024, accessed 26 June 2025).

Staniczek

Manasar-Dyrbuś

Stojko

, et al. Adolescent pregnancy: a comparative insight into the prevalence and risks of obstetric complications in a polish cohort. J Clin Med 2024; 13(19): 5785.

Whitcomb

Obstetric fistulas in Sub-Saharan Africa. Ballard Brief 2024; 2024: 1.

Abate

Sendekie

Alamaw

, et al. Prevalence, determinants, and complications of adolescent pregnancy: an umbrella review of systematic reviews and meta-analyses. AJOG Glob Rep 2025; 5(1): 100441.

10.

Todhunter

Hogan-Roy

Pressman

(eds.) Complications of pregnancy in adolescents. Semin Reprod Med 2022; 40(1-02): 98–106.

11.

Pregnancy Archive. Exploring the reasons behind the rising rates of teenage pregnancy – unveiling the complex factors contributing to this global issue, (2024, accessed 3 September 2025).

12.

Kawakita

Wilson

Grantz

, et al. Adverse maternal and neonatal outcomes in adolescent pregnancy. J Pediatr Adolesc Gynecol 2016; 29(2): 130–136.

13.

Macedo

TCC

Montagna

Trevisan

, et al. Prevalence of preeclampsia and eclampsia in adolescent pregnancy: a systematic review and meta-analysis of 291,247 adolescents worldwide since 1969. Eur J Obstetr Gynecol Reprod Biol 2020; 248: 177–186.

14.

Plan International. Teenage pregnancy. https://plan-international.org/srhr/teenage-pregnancy/ (accessed 5 July 2025).

15.

Noori

Proctor

Efevbera

, et al. The effect of adolescent pregnancy on child mortality in 46 low- and middle-income countries. BMJ Glob Health 2022; 7(5): e007681.

16.

Ogawa

Matsushima

Urayama

, et al. Association between adolescent pregnancy and adverse birth outcomes, a multicenter cross sectional Japanese study. Sci Rep 2019; 9: 2365.

17.

Wahyudi

SPE

Amalia

Frety

EE.

Correlation between adolescent pregnancy with labor complication in Puskesmas Galis Bangkalan Regency. J Midwifery 2023; 8(1): 131.

18.

Kassa

Arowojolu

, et al. Adverse neonatal outcomes of adolescent pregnancy in Northwest Ethiopia. PLOS ONE 2019; 14: e218259.

19.

Visacri

Komoda

Meza

, et al. Pregnancy in adolescence: a public health issue. CIPF, Redondela, Spain, 2023, pp. 7–30.

20.

UNICEF. Goal 3: Good health and well-being: Ensure healthy lives and promote well-being for all at all ages. https://data.unicef.org/sdgs/goal-3-good-health-wellbeing/ (2025, accessed 23 June 2025).

21.

Wado

Mutua

Mohiddin

, et al. Intimate partner violence against adolescents and young women in sub-Saharan Africa: who is most vulnerable? Reprod Health 2021; 18(1): 119.

22.

Azizi

Abbasi

Nazem

, et al. The most prevalent intimate partner violence; physical, sexual, verbal, or emotional in early marriage: a narrative review. J Pediatr Rev 2021; 9(1): 27–36.

23.

Advocates for Youth. Advancing the sexual and reproductive health and rights of adolescent girls and young women, https://www.advocatesforyouth.org/resources/fact-sheets/advancing-the-sexual-and-reproductive-health-and-rights-of-adolescent-girls-and-young-women/ (2025, accessed 8 July 2025).

24.

World Health Organization. WHO guideline on preventing early pregnancy and poor reproductive outcomes among adolescents in low- and middle-income countries. Geneva: World Health Organization, 2025.

25.

World Health Organization. Policy and education: ways to end child marriage and prevent adolescent pregnancy, https://www.who.int/news-room/feature-stories/detail/policy-and-education—ways-to-end-child-marriage-and-prevent-adolescent-pregnancy (2025, accessed 5 July 2025).

26.

Gerbaka

Richa

Tomb

Early and forced marriage. In: Gerbaka

Richa

Tomb

(eds.) Child sexual abuse, exploitation and trafficking in the Arab region. Cham: Springer International Publishing; 2021. pp.229–271.

27.

Dubik

Aniteye

Richter

Socio-cultural factors influencing teenage pregnancy in the East Mamprusi Municipality, Ghana. Afr J Reprod Health 2022; 26(5): 120–30.

28.

The DHS Program. DHS7 recode manual: household, woman’s, and man’s questionnaires. Rockville, MD: ICF, https://dhsprogram.com/pubs/pdf/DHSG1/Guide_to_DHS_Statistics_DHS-7.pdf (2018, accessed 9 July 2025).

29.

The DHS Program. DHS8 recode manual: household, woman’s, and man’s questionnaires. Rockville, MD: ICF, https://www.dhsprogram.com/pubs/pdf/DHSG1/Guide_to_DHS_Statistics_DHS-8.pdf (2021, accessed 9 July 2025).

30.

World score card. Income Levels by Country. https://worldscorecard.com/world-facts-and-figures/income-levels-by-country-2025/ (2025, accessed 9 July 2025).

31.

DHS Program. Available Dataset accessed online. http://dhsprogram.com/data/available-datasets.cfm (accessed 23 June 2025).

32.

DHS program. Data quality and use https://dhsprogram.com/data/Data-Quality-and-Use.cfm (accessed 23 June 2025).

33.

Croft

DHS data editing and imputation,https://dhsprogram.com/pubs/pdf/DHSG3/DHS_Data_Editing.pdf (1991, accessed 24 June 2025).

34.

Sulaiman

SKB

Idris

Hod

, et al. A mini-review on the determinants and risk factors of adolescent pregnancy in developing countries. Malays J Med Health Sci 2023; 19(3): 341–348.

35.

United Nations Department of Economic and Social Affairs. Good health and well-being. https://unstats.un.org/sdgs/report/2024/Goal-03/ (2024, accessed 5 September 2025).

36.

Sedgh

Finer

Bankole

, et al. Adolescent pregnancy, birth, and abortion rates across countries: levels and recent trends. J Adolesc Health 2015; 56: 223–230.

37.

Ahinkorah

Budu

Seidu

A-A

, et al. Girl child marriage and its association with maternal healthcare services utilization in sub-Saharan Africa. BMC Health Serv Res 2022; 22(1): 777.

38.

Ahinkorah

Hagan

Seidu

A-A

, et al. Association between female genital mutilation and girl-child marriage in sub-Saharan Africa. JBiosoc Sci 2023; 55(1): 87–98.

39.

Shasha

Phiri

Namayawa

, et al. Prevalence and factors associated with early childbearing in sub-saharan Africa: evidence from demographic and health surveys of 31 countries. BMC Womens Health 2023; 23(1): 430.

40.

Fatusi

Adedini

Mobolaji

JW.

Trends and correlates of girl-child marriage in 11 West African countries: evidence from recent Demographic and Health Surveys. AAS Open Res 2021; 4: 35.

41.

Hailu

Beyene

Adolescent marriage, maternity, and limited access to education in 106 countries: Bayesian analysis of prevalence, trend, and prediction. Sci Rep 2025; 15(1): 9584.

42.

Boadu

Coverage and determinants of modern contraceptive use in sub-Saharan Africa: further analysis of demographic and health surveys. Reprod Health 2022; 19(1): 18.

43.

Osborne

James

Bangura

, et al. Exploring the drivers of unmet need for contraception among adolescents and young women in Sierra Leone: a cross-sectional study. Contracept Reprod Med 2024; 9(1): 25.

44.

Sidibé

Grovogui

Kourouma

, et al. Unmet need for contraception and its associated factors among adolescent and young women in Guinea: a multilevel analysis of the 2018 Demographic and Health Surveys. Front Glob Womens Health 2022; 3: 932997.

45.

Ushie

Juma

Akuku

, et al. Assessing induced abortion underreporting in restrictive settings using prospective morbidity surveys from Kenya, Liberia, and Sierra Leone. Sci Rep 2025; 15(1): 28756.

46.

Bussolo

Ezebuihe

Munoz Boudet

, et al. Social Norms and gender equality: A descriptive analysis for South Asia. The World Bank Group, 2022.

47.

Bhowmik

Biswas

Hossain

Child marriage and adolescent motherhood: a nationwide vulnerability for women in Bangladesh. Int J Environ Res Public Health 2021; 18(8): 4030.

48.

Kamal

Ulas

Child marriage and its impact on fertility and fertility-related outcomes in South Asian countries. Int Sociol 2021; 36(3): 362–377.

49.

Ali

Khaliq

Lokeesan

, et al. Prevalence and predictors of teenage pregnancy in Pakistan: a trend analysis from Pakistan Demographic and Health Survey datasets from 1990 to 2018. Int Health 2022; 14(2): 176–182.

50.

Sreeramareddy

Acharya

Tiwari

Inequalities in demand satisfied with modern methods of family planning among women aged 15–49 years: a secondary data analysis of Demographic and Health Surveys of six South Asian countries. BMJ Open. 2022; 12(6): e049630.

51.

Department of Statistics (DoS) [Jordan] and ICF. Jordan population and family and health survey 2023. Amman, Jordan, and Rockville, Maryland, USA: DoS and ICF, 2024.

52.

Beqiraj

Family law in Albania: regulatory responses to the jurisprudence of the ECTHR. 2022.

53.

Kassa

Arowojolu

Odukogbe

, et al. Prevalence and determinants of adolescent pregnancy in Africa: a systematic review and Meta-analysis. Reprod Health 2018; 15(1): 195.

54.

Mathewos

Mekuria

Teenage pregnancy and its associated factors among school adolescents of Arba Minch Town, Southern Ethiopia. Ethiop J Health Sci 2018; 28(3): 287–298.

55.

Mohammed

Analysis of national and subnational prevalence of adolescent pregnancy and changes in the associated sexual behaviours and sociodemographic determinants across three decades in Ghana, 1988–2019. BMJ Open 2023; 13(3): e068117.

56.

Worku

Tessema

Teshale

, et al. Prevalence and associated factors of adolescent pregnancy (15–19 years) in East Africa: a multilevel analysis. BMC Pregnancy Childbirth 2021; 21: 1–8.

57.

Malunga

Sangong

Saah

, et al. Prevalence and factors associated with adolescent pregnancies in Zambia: a systematic review from 2000–2022. Arch Public Health 2023; 81: 27.

58.

Moshi

Tilisho

The magnitude of teenage pregnancy and its associated factors among teenagers in Dodoma Tanzania: a community-based analytical cross-sectional study. Reprod Health 2023; 20(1): 28.

59.

Mezmur

Nega

Alemayehu

Teenage pregnancy and its associated factors in Eastern Ethiopia: a community-based study. Int J Womens Health 2021; 13(null): 267–278.

60.

Adhena

Fikre

Teenage pregnancy matters in refugee setup: early pregnancy among adolescent girls in Kule refugee camp, Gambella, Ethiopia. BMC Pregnancy Childbirth 2023; 23(1): 861.

61.

Urindwanayo

Richter

Teenage pregnancy in refugee camps: a narrative synthesis. J Int Womens Stud 2020; 21(1): 255–270.

62.

Moodley

Hidden realities: lived experiences with teenage pregnancy, motherhood, and HIV in Botshabelo, South Africa. PhD Thesis, University of Oxford, 2021.

63.

Amoadu

Hagan

Ansah

EW.

Adverse obstetric and neonatal outcomes of adolescent pregnancies in Africa: a scoping review. BMC Pregnancy Childbirth 2022; 22(1): 598.

64.

Baten

Biswas

Kendal

, et al. Utilization of maternal healthcare services in low- and middle-income countries: a systematic review and meta-analysis. Syst Rev 2025; 14(1): 88.

65.

Lawot

Bajracharya

Acharya

Consequences of adolescent pregnancy on maternal and neonatal health: a narrative review. J Nurs Educ Nepal 2023; 14(1): 80–86.

66.

Maheshwari

Khalid

Patel

, et al. Maternal and neonatal outcomes of adolescent pregnancy: a narrative review. Cureus 2022; 14(6): e25921.

67.

Teshale

Dellie

Aschalew

, et al. Estimated effect of teenage pregnancy on adverse birth outcome in Sub-Saharan African countries: propensity score matching analysis of recent demographic and health survey data. BMC Preg Childbirth 2025; 25(1): 442.

68.

Yu Wu

Berwick

Incidence, indications, and complications of cesarean delivery among adolescent childbirths: national inpatient sample 2019–2021. J Pediatr Adolesc Gynecol 2025; 38(2): 255.

69.

Idris

Menghisteab

Cesarean section delivery rates, determinants, and indications: a retrospective study in Dekemhare Hospital. Global Reprod Health 2022; 7(1): e56.

70.

Betran

Moller

A-B

, et al. Trends and projections of caesarean section rates: global and regional estimates. BMJ Global Health 2021; 6(6): e005671.

71.

Matei

Poenaru

Dimitriu

MCT

, et al. Obstetrical soft tissue trauma during spontaneous vaginal birth in the Romanian adolescent population—multicentric comparative study with adult population. Int J Environ Res Public Health 2021; 18(21): 11491.

72.

Amoadu

Hagan

Ansah

EW.

Adverse obstetric and neonatal outcomes of adolescent pregnancies in Africa: a scoping review. BMC Preg Childbirth 2022; 22(1): 598.

73.

Chauke

Improving access to emergency obstetric care in low- and middle-income countries. Best Pract Res Clin Obstet Gynaecol 2025; 98: 102572.

74.

Poudel

Razee

Dobbins

, et al. Adolescent pregnancy in South Asia: a systematic review of observational studies. Int J Environ Res Public Health 2022; 19(22): 15004.

75.

da Conceição Chagas de Almeida

Aquino

. The role of education level in the intergenerational pattern of adolescent pregnancy in Brazil. Int Pers Sex Reprod Health 2009; 35(3): 139–146.

76.

Pradhan

Wynter

Fisher

Factors associated with pregnancy among adolescents in low-income and lower middle-income countries: a systematic review. J Epidemiol Commun Health 2015; 69(9): 918–924.

77.

Mohr

Carbajal

Sharma

BB.

The influence of educational attainment on teenage pregnancy in low-income countries: a systematic literature review. J Social Work Global Commun 2019; 4(1): 2.

78.

Salami

Ayegboyin

Adedeji

IA.

Unmet social needs and teenage pregnancy in Ogbomosho, South-western Nigeria. Afr Health Sci 2014; 14(4): 959–966.

79.

Yakubu

Salisu

WJ.

Determinants of adolescent pregnancy in sub-Saharan Africa: a systematic review. Reprod Health 2018; 15: 1–11.

80.

World Bank. Low and italics lower middle income countries list 2024–2025, https://ipwgnet.org/doc/IPWG2025/List_of_LIC&LMIC2024.pdf (2024, accessed 6 July 2025).

81.

Asare

BY-A

Baafi

Dwumfour-Asare

, et al. Factors associated with adolescent pregnancy in the Sunyani Municipality of Ghana. Int J Afr Nurs Sci 2019; 10: 87–91.

82.

Donatus

Sama

Tsoka-Gwegweni

, et al. Factors associated with adolescent school girl’s pregnancy in Kumbo East Health District North West region Cameroon. Pan Afr Med J. 2018; 31: 138.

83.

Bhatta

Bhandary

Factors associated with adolescent pregnancy in selected rural municipalities of Lalitpur District: a mixed method study. J Gen Pract Emergency Med Nepal. 2022; 9(13): 2363.

84.

Kearney

Levine

PB.

Media influences on social outcomes: the impact of MTV’s “16 and Pregnant” on teen childbearing. Am Econ Rev 2015; 105(12): 3597–3632.

85.

Brindis

Decker

Gutmann-Gonzalez

, et al. Perspectives on adolescent pregnancy prevention strategies in the United States: looking back, looking forward. Adolesc Health Med Ther 2020; 11(null): 135–145.

86.

Bailey

Mann

Wayal

, et al. Sexual health promotion for young people delivered via digital media: a scoping review. Public Health Res 2015; 3(13): 1–119.

87.

Levin-Zamir

Lemish

Gofin

Media Health Literacy (MHL): development and measurement of the concept among adolescents. Health Educ Res 2011; 26(2): 323–335.

88.

Scull

Kupersmidt

Malik

, et al. Examining the efficacy of an mHealth media literacy education program for sexual health promotion in older adolescents attending community college. J Am Coll Health 2018; 66(3): 165–177.

89.

Aminatussyadiah

Wardani

SFP

. Overview of premarital sexual behavior at risk of pregnancy termination in adolescent girls in Indonesia. Jurnal Kebidanan 2025; 14(1): 8–15.

90.

Odimegwu

Mkwananzi

Factors associated with teen pregnancy in sub-Saharan Africa: a multi-country cross-sectional study. Afr J Reprod Health 2016; 20(3): 94–107.

91.

Miriti

Mutua

MN.

Teenage pregnancies and girls education capabilities in Nzambani Ward Kitui County in Kenya: school and community perspective. Int J Gend Studies 2019; 4(1): 59–84.

92.

Gideon

Factors associated with adolescent pregnancy and fertility in Uganda: analysis of the 2011 demographic and health survey data. Am J Sociol Res 2013; 3(2): 30–35.

93.

Cleland

Conde-Agudelo

Peterson

, et al. Contraception and health. Lancet 2012; 380(9837): 149–156.

94.

Ganatra

Faundes

Role of birth spacing, family planning services, safe abortion services and post-abortion care in reducing maternal mortality. Best Pract Res Clin Obstet Gynaecol 2016; 36: 145–155.

95.

Sánchez-Páez

Ortega

JA.

Adolescent contraceptive use and its effects on fertility. Demographic Research. 2018;38:1359-88.

96.

Anna Karolina Lages de

Nery

. Knowledge about contraception and factors associated with pregnancy planning in adolescence. Cogitare enfermagem 2018; 23(2): e55841.

97.

Costa

GPO

Costa

GPO

de Paula Farias

, et al. Contraception knowledge, attitudes and practices of adolescents at risk of pregnancy in northeastern Brazil. Open J Obstet Gynecol 2014; 4(6): 300–308.

98.

Serván-Mori

Cerecero-García

Morales-Vazquez

, et al. The role of effective knowledge on contraceptive methods use in the replication of mother-daughter adolescent pregnancy in Mexico. Arch Sex Behav 2022; 51(8): 4035–4046.

99.

Reda Mohamed Ismail

Ragab

Mohamed

, et al. Rural teenage pregnant female unmet needs. Egyptian J Health Care 2022; 13(3): 203–216.

100.

Adolescence

O’Brien

Warner

, et al. Condom use by adolescents. Pediatrics 2013; 132(5): 973–981.

101.

Wiyeh

Mome

RKB

Mahasha

, et al. Effectiveness of the female condom in preventing HIV and sexually transmitted infections: a systematic review and meta-analysis. BMC Public Health 2020; 20(1): 319.

102.

Osaikhuwuomwan

Osemwenkha

AP.

Adolescents’ perspective regarding adolescent pregnancy, sexuality and contraception. Asian Pacific J Reprod 2013; 2(1): 58–62.

103.

Todd

Black

Contraception for adolescents. J Clin ResPediatr Endocrinol 2020; 12(Suppl 1): 28.

104.

Akanbi

Ope

Adeloye

, et al. Influence of socio-economic factors on prevalence of teenage pregnancy in Nigeria. Afr J Reprod Health 2021; 25(5s): 138–146.

105.

Baruwa

Mkwananzi

Amoateng

, et al. Teenage pregnancy among unmarried teenagers in Malawi: does sex of the household head matter? Afr J Reprod Health 2020; 24(4): 51–57.

106.

Buvinić

Gupta

GR.

Female-headed households and female-maintained families: are they worth targeting to reduce poverty in developing countries?

Econ Dev Cult Change 1997; 45(2): 259–280.

107.

Ciciolla

Luthar

SS.

Invisible household labor and ramifications for adjustment: mothers as captains of households. Sex Roles 2019; 81(7): 467–486.

108.

Idris

Hamis

Bukhori

ABM

, et al. Women’s autonomy in healthcare decision making: a systematic review. BMC Womens Health 2023; 23(1): 643.