Skip to main content
Download PDF

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 Pregnancy and Childbirth volume 25, Article number: 442 (2025) Cite this article

Abstract

Introduction

Globally adverse birth outcome is being a series public health problem. As studies showed, even though the etiologies are multifactorial, extreme age pregnancies have more risk for adverse birth outcome. This study determines the estimated effect of teenage pregnancy on adverse birth outcome.

Method

The study analyzed data from the most recent Demographic and Health Surveys (DHS) data. Propensity score matching (PSM) analysis was employed by using age as treatment variable, teenager as treated and non-teenagers as control group and weighted sample of 45,790 (non-teenagers = 41,769 and teenagers = 4,021). The outcome variable; adverse birth outcome was categorized as “Yes” if a woman had either preterm birth, still birth, low birth weight or macrosomia in her recent birth and “No” otherwise. Covariates that had significant association with the treatment and outcome variables were considered for PSM analysis. After testing of each matching techniques (nearest neighbor, kernel and radius), the nearest neighbor (10) approach produced better covariate balance and selected as the best matching algorism for our analysis. Finally, the effect of teenage pregnancy on adverse birth outcome was measured and reported as average treatment effect on the treated (ATT) and the quality of matching and sensitivity to hidden bias was assed by t-statistics significance level and Mantel–Haenszel statistic respectively.

Results

This study found that around one in ten (8.7%) of the women had pregnancy between the age of 15 and 19 years. The magnitude of adverse birth outcome among teenagers and non-teenagers was also 45.4% and 39.9% respectively. Teenage pregnancy contributed to a 4.7% increasing adverse birth outcome (ATT = 4.7%). Similarly, the Average Treatment Effect on Untreated (ATU) was 4.8%.

Conclusion

This study revealed that around one in teen women had pregnancy between the age of 15 and 19 years and teenage pregnancy had more risk of having adverse birth outcomes as compared to non-teenagers. Thus, we recommend to policy makers and implementers to design policies and strategies to improve teenagers’ access to prenatal care, family planning, and sexual education, awareness of creation on teenage pregnancy risks.

Peer Review reports

Introduction

Adverse birth outcome is defined as a low birth weight (LBW) (birth weight < 2500gram), macrosomia (birth weight > 4000 gram), stillbirth (fetal death at or after 7 months of gestation and before second stage of labor), and preterm birth (less than 37 weeks at birth) [1, 2]. Globally, the burden of adverse birth outcomes is being a series public health problem. Over the world an estimated 13 million, 46 million, and 2.6 million pregnancies end in preterm birth, abortion, and stillbirth annually, respectively. Moreover, around 20% and 15% of all births are LBWs and macrosomia respectively [2, 3].

Although the etiologies of adverse birth outcomes are multifactorial and not completely understood yet, studies showed that being on the extreme age (below 20 years and above 40 years) group has significant attribution [4,5,6]. Moreover, teenage pregnant faced with the simultaneous challenges of pregnancy and parenting while navigating the developmental tasks of adolescence which increases their risk for social and psychological health problems [7].

Teenage pregnancy refers to either intended or unintended pregnancy occurred within the age below 19 years. It is a major public health concern worldwide. The World Health Organization has estimated that every year approximately 17 million females aged below 19 years give birth [8, 9]. The magnitude of teenage pregnancy varies from 0.29% in south Korea to over 30% in African countries [10]. Globally, teenage pregnancy is considered to be very high risk due to teenage girls are physically and psychologically immature for reproduction. In addition, some extrinsic factors like inadequate prenatal care, illiteracy, and poor socioeconomic conditions affect the outcome of.

pregnancy in teenage girls [11].

In the developing countries nearly half of all teenage pregnancies are unintended, and more than half of them end in abortion [12]. In addition, studies showed that more than one-third of maternal morbidity and mortality in Africa are attributed to teenage pregnancies [13]. Moreover, teenage pregnancy is associated with adverse perinatal complications, such as preterm births, stillbirths, neonatal deaths, and delivered low birth weight babies [14, 15].

Even though there are studies conducted in SSA countries on the prevalence of teenage pregnancy and adverse birth outcome [16,17,18], studies on that determine the estimated effect of teenage pregnancy on adverse birth outcomes were very limited. This study determines the estimated effect of teenage pregnancy on adverse birth outcomes in SSA by using the most recent DHS data and more advanced model. Moreover, it determines the magnitude of both teenage pregnancy and adverse birth outcome in the region.

Methods

Study design and period

This study was conducted based on Demographic and Health Survey (DHS) data of 8 SSA countries. These Sub-Saharan countries were selected based on the their most recent DHS data availability (from 2020 to 2024) and included Burkina Faso (2021), Cot’ devour (2021), Gahana (2022), Kenya (2022), Lesotho (2024), Mozambique (2023), Senegal (2023) and Tanzania (2022). The DHS is a cross-sectional survey conducted every five years to generate updated health and health-related indicators.

Sample size and sampling techniques

A total weighted sample size of 45,790 with nearly one to ten ratio (teenagers = 4,021 and non-teenagers or elders = 41,769) of reproductive age women who gave birth in the last five years preceding the survey was included. The DHS typically used two stage sampling design. First clusters (Enumeration areas) were selected from sampling frame and secondly within each selected enumeration area households were selected and listed by the field enumerators [19]. The data was obtained from the DHS website; https://dhsprogram.com [20] upon reasonable request via online and then the Birth Record (BR) data of each selected country is appended.

Study variables and their measurements

Outcome variable

The dependent variable of this study was adverse birth outcomes, defined as ‘yes’ when a woman had at least one of the followings: stillbirth, preterm birth, Low Birth Weight (LBW), and macrosomia. Stillbirth was derived from v233 and defined as pregnancy loss occurring after 7 completed months of gestation [21]. Preterm birth was obtained from the DHS question “duration of pregnancy for the most recent birth (b20)” and defined as births after 28 weeks and before 37 weeks of gestational age [22]. Whereas LBW and macrosomia were derived from the DHS question “birth weight in kilograms (m19)” and defined as weight at birth less than 2500 grams; ≥4000 grams respectively [23].

Comparison groups and categories for adverse pregnancy outcome

The treatment variable of our study was age of the women during their recent pregnancy. The treatment group was those women whose recent pregnancy happened 15–19 years of their age (teenagers) and the control group was those women whose recent pregnancy happened at 20–49 years of their age (non-teenagers).

Covariates

Variables that have an effect on teenage pregnancy and adverse birth outcomes at the same time were included. Variables like residence, maternal education, marital status, sex of household head, media exposure, household wealth status, maternal working status, ANC follow-up, Place of delivery and country were considered covariates for matching. Confounding variables that have a significant association with the treatment and outcome variables were considered for matching.

Data management and analysis

All reported results were based on weighted data adjusted for design and non-response using the weighting variable (v005). Primary sampling units (v023) and clustering (v021) were accounted for using the svy declaration to ensure accurate estimation of standard errors and confidence intervals. In cross-sectional studies like DHS random assignment of participants to the treatment and control groups is impossible. Thus, inherent imbalance of the observed variables introduces bias and influences the causal effect of the exposure. In such conditions balancing score can be used to correct bias related to participants assignment to either group. Based on the balancing score, the observed variables should be independent of the assignment of the exposure, teenage pregnancy or not [24, 25].

The propensity score method is commonly used to balance the inequality of confounding variables in such cross-sectional studies [24]. A propensity score is the likelihood that a woman is being in the treatment group (teenagers) given all the observed covariates and it has a value between 0 and 1. Propensity score analysis starts with an assessment of the imbalance of the baseline covariates between the teenagers and non-teenagers. This can be assessed by significance tests like an independent t-test for continuous variables and chi2 for categorical variables [25, 26].

Stata version 17 software was used to perform our analysis. The imbalance of the baseline covariates between the teenagers and non-teenagers was assessed by chi-square test. Accordingly, variables with p-value ≤ 0.05 were considered as candidate covariates for propensity matching analysis [26].

Propensity Score Matching (PSM) analysis

The PSM was performed based on the assumptions of conditional independence assumption (CIA) or un-confoundedness and common support. First, propensity score (PS) was estimated based on the probit model using the above-mentioned matching variables. We had also imposed the common support option to improve the matching quality [27]. The command “pscore” in STATA was used to estimate the PS. We then performed PSM analysis using the command “psmatch2” in STATA to obtain the average treatment effect on the population (ATE), the average treatment effect on the treated (ATT), and the average treatment effect on the untreated (ATU). We tested kernel, nearest neighbor and radius matching algorisms and the nearest neighbor (10) produced better covariate balance and selected as the best matching algorism for our analysis. Besides, nearest neighbor method estimates the average treatment effect for the treated (ATT) and for the untreated (ATU) and enables matching with the smallest possible average propensity score differences between treated and untreated participants [28, 29].

We estimated the average effect of teenage pregnancy on adverse birth outcome by using PSM analysis. In this approach we assumed that AiT is to be adverse birth outcomes teenagers (treatment group), and AiC denotes adverse birth outcomes non-teenagers (control group). The observed outcome (Ai) can be written as Ai=(1 − Ti) AiC + TiAiJ, where Ti = 0, and 1 denotes treatment assignment (age at recent pregnancy). Hence, our interest is to estimate the average treatment effect on the treated (ATT), E (AiT−AiC/Ti = 1). This cannot be estimated directly since normally observed neither AiT for Ti = 0 nor AiC for Ti = 1 is not known.

Quality of matching was tested using the command “pstest” in STATA. The covariates’ distribution between intervention and control group (before & after matching) were assessed using standardized bias, two-sample t-test results, pseudo-R2, and likelihood ratio (LR) test for joint insignificance. Accordingly, matching method that provided the best results; significant reduction in standardized bias in the matched groups as compared to unmatched groups; mean absolute bias value < 5% in the two-sample t-test, P-value > 0.05 and Pseudo R2 and LR test = minimum pseudo-R2 and LR test was selected [24, 30].

We also checked whether the final PSM estimates are sensitive for hidden bias and the robustness of the model by using Mantel–Haenszel (MH) test statistic. The gamma coefficient (Γ) was the factor by which unobserved confounders would affect the assignment into intervention for a treated participant compared to an untreated participant with matching covariates [31]. The range of gamma specified was between 1 and 2 with 0.05 increments. Sensitivity analysis was performed using the STATA command “mhbounds”. The test showed that over, under or acceptable level of estimation of treatment effect [32].

Results

Sociodemographic characteristics and recent pregnancy history of respondents

This study included 45,790 reproductive age women who gave birth in the last five years preceding the survey. The study used 41,769 non-teenagers as control group and 4,021 teenagers as treatment group. Around 28% of teenagers and 35% of non-teenagers reside in urban areas. Similarly, over one-fourth (28.92%) and one-third (38.48%) of teenagers and non-teenagers had no formal education and majority (65.03% and 87.72%) of them were currently married or living with a partner respectably. In the same way, most (62.99% and 70.27%) of teenagers and non-teenagers also had media exposure and around half (46.83% and 55.41%) were from poor wealth status respectively. Majority of the women had ANC follow-up (around 95%) and delivered at health facilities (around 83%) for their recent pregnancy (Table 1).

Table 1 Sociodemographic and recent pregnancy history related characteristics of teenagers and non-teenagers in the study of effect of teenage pregnancy on adverse birth outcome (n = 45,790)
Full size table

The magnitude of teenage pregnancy was 8.7% (95% CI: 0.085, 0.090) with the highest and the lowest prevalence observed in Mozambique (15.59%) and Ghana (5.86%) respectively. The magnitude of adverse birth outcome among teenagers, non-teenagers and general population was (45.4%, 39.9% and 40.41%) respectively. In the same way, adverse birth outcome was observed high in Senegal (52.04%) and low in Kenya (27.28%) (Fig. 1).

Fig. 1
figure 1

Prevalence of teenage pregnancy and adverse birth outcome in selected SSA countries

Full size image

Our study found that the around one-fourth (25.8% and 25.11%) of teenagers and non-teenagers had faced macrosomia in their recent birth respectively. On the other hand, almost one in hundred (0.99%) teenagers and one in thirty-three (3.26%) non-teenagers had still birth in their recent birth (Fig. 2).

Fig. 2
figure 2

Prevalence of adverse birth outcome components among teenagers and non-teenagers

Full size image

Estimation of propensity score

Before PSM, the baseline variables showed significant differences in adverse birth outcome with p-value of ≤ 0.05. These variables including residence, women education, current marital status, women’s education status, religious, number of five and under years children, current working status of a woman, media exposure, wealth index, women’s current working status, ANC follow-up for most recent pregnancy, place of delivery for most recent birth and country. The strength, direction and significance of the coefficients are shown in the table below (Table 2).

Table 2 The association between covariates and adverse birth outcome in the study of impact of teenage pregnancy on adverse birth outcome in SSA countries: propensity score matching
Full size table

Impact of teenage pregnancy on adverse birth outcome

Nearest neighbor matching with 1 to 10 ratio (10) was used to estimate the Average Treatment Effect (ATE), Average Treatment on Treated (ATT), and Average Treatment on Untreated (ATU). Differences in the mean outcome in matched samples can be utilized to obtain these estimates. The calculated ATT values in the treatment and untreated groups were 0.454 and 0.407, respectively, indicating that adverse birth outcome was increased by 4.7% due to teenage pregnancy by considering other confounders constant. Similarly, the calculated ATU values in the treated and untreated groups were 0.399 and 0.447, respectively. This means adverse birth outcome in the control group (pregnancy after age of 19) had decreased by 4.8%. The average treatment on the study population (ATE) was found to be 4.6% (Table 3).

Table 3 The impact of teenage pregnancy on adverse birth outcome in SSA countries: propensity score matching
Full size table

Quality of matching

Common support

The distributions are almost similar for both teenager and non-teenager groups after matching on PS. The presence of significant overlap between the characteristics of teenager and non-teenager groups proved the validity of the common support assumption. The mean PS was 0.088, with variance of 0.004 and standard deviation of 0.066 between the intervention and control group. The balancing property was satisfied and the region of common support between the two groups was ranging from 0.019 to 0.385 of the PS and all were on-support region (Fig. 3). the final number of blocks was thirteen.

Fig. 3
figure 3

Propensity score histogram by treatment status in the study of impact of teenage pregnancy on adverse birth outcome in SSA countries: PSM

Full size image

Balancing test

The difference between the unmatched and matched pairs was evaluated by t-statistic and the significance level was determined by its p-value. Our result showed that there was no significant mean difference (p > 0.05) across almost all factors after matching, despite a significant mean difference across all covariates before matching. This indicates that the variables were all sufficiently balanced (Table 4).

Table 4 Performance of the propensity score matching quality measurements; impact of teenage pregnancy on adverse birth outcome in SSA countries
Full size table

Standard bias and model significance

This study aimed to determine the estimated effect of teenage pregnancy on adverse birth outcome. In observational studies the observed association between the exposure and outcome possibly explained by chance (random error), bias (systematic error), effect-cause (reverse causality), confounding and cause-effect. Despite the DHS data is prone to recall and social desirability bias, this potential error is likely to be at random between the treatment and control groups and the observed effect is very unlikely to be due to chance. It is possible that unmeasured predictors of the outcome variables might play a role in the relationship. However, the standardized percent bias and sensitivity analysis indicated that the estimated effects were robust to unmeasured covariates (Fig. 4 ).

The mean and median biases considerably lowered once the intervention and control groups were matched. The mean absolute bias in the unpaired sample decreased from 18.6 to 1.4% after the treated and control groups were matched. The mean bias after matching is less than 5% threshold and shows that the model’s quality matching has improved. The median bias also decreased from 18.6% in the unmatched to 0.7% after matching. The overall significance of the model was assessed using the LR and pseudo R2 tests. After matching, the pseudo-R2 was less than 0.001 and the LR-chi2 test was 5.79 with p-vale of 0.761, suggesting that there was no systematic variation in the covariate distribution between the treated and control groups (Table 5).

Table 5 Performance of the propensity score matching quality measurements; impact of teenage pregnancy on adverse birth outcome in SSA countries
Full size table
Fig. 4
figure 4

Standardized per cent bias in the distribution of confounders before and after matching

Full size image

Sensitivity analysis

The sensitivity analysis was performed using the Mantel–Haenszel statistic to estimate the extent of unobservable covariates biases on our inferences about the impact of teenage pregnancy on adverse outcome. The QMH test statistic yields the same result when no hidden bias (Γ = 1) is assumed, suggesting a strong treatment effect. Our result showed where Γ = 1, the QMH test statistic is 1.243 for adverse birth outcome. At gamma ≥ 1.3 both over- and under-estimation test statistics show highly significant results (p < 0.001), indicating that the association between teenage pregnancy and adverse birth outcomes is robust to moderate levels of unmeasured confounding (Table 6).

Table 6 Mantel-Haenszel bounds for sensitivity analysis: impact of teenage pregnancy on adverse birth outcome in SSA countries: propensity score matching
Full size table

Discussion

The objective of this study was to estimate the effect of teenage pregnancy on adverse birth outcomes. It also determined the magnitude of teenage pregnancy and adverse birth outcome in the study region. Accordingly, the magnitude of teenage pregnancy was 8.7%. This finding was lower than studies done in the sub-Saharan Africa [10, 33]. The variation may be due to the study population difference, as the first systematic review includes all pregnancy history between the age of 13–19 and the second study includes all 15–19 years women who gave birth or being pregnant at the time of data collection while our study included only 15–19 years women who gave births with last five years prior to the survey. In the same way, the magnitude of teenage pregnancy in our study was lower than a systematic review conducted in Africa. This difference may be due to the previous study includes women aged 13–19 while ours only 15–19 years in the study population. Additionally, our study didn’t include countries with higher prevalence like Ethiopia and western Africa in the previous studies [10].

Our study also showed that magnitude of adverse outcomes among teenagers and non-teenagers were 45.4% and 39.9% respectively. From our PSM analysis result 4.7% of the adverse birth outcomes was attributed for teenage pregnancy. The finding was supported by previous studies conducted in central Africa, and Tigray region in Ethiopia [34, 35]. It was also congruent with previous studies which stated as early adolescent pregnancy significantly increased the occurrence of adverse birth outcomes [36,37,38]. The increased risk of adverse birth outcomes may be due to teenagers may not be fully developed to handle the physical demands of pregnancy and dual burden of needing nutrients for their own growth and that of the fetus [39]. Furthermore, teenagers may delay or miss health care seeking due to lack of awareness, stigma, or financial constraints [40]. The study also found that one in five of the study participants had adverse birth outcomes and it was lower than previous study done in sub-Sahara Africa [41]. The variation may be due to that this study included Liberia, Madagascar and Rwanda which had lower prevalence of adverse birth outcome while ours didn’t due to lack of recent DHS data.

This study also showed that the magnitude of low birth weight and pre-term birth were higher among teenagers. The finding was supported by previous studies done in Japan, Tigray region in Ethiopia north west Ethiopia and rural eastern Ethiopia [35, 42,43,44]. On the other hand, in this study still birth was slightly lower among teenagers and macrosomia was happened almost equally among teenagers and non-teenagers. This finding was also inline with previous study done in Ethiopia [43].

Strength and limitation of the study

This study was done based on nationally representative DHS data with a high response rate and standardized questionnaire for the data collection. The study uses large sample size which enhances the reliability, precision and generalizability of the findings. The study also adjusted for the potential confounders using the PSM approach in the estimation of impact of teenage pregnancy on adverse pregnancy outcomes.

However, the matching was done based on the observed variables only, and there may be a possibility of residual confounding (unobserved variables). Moreover, the outcome variable; adverse birth outcomes is self-reported; the magnitude may be decreased due to social desirability.

Policy implication of the study

This study highlighted that the magnitude of teenage pregnancy was significant and its effect on adverse birth outcomes needs a policy attention. The policy makers and implementers can use the findings of this study as evidence showing the burden of teenage pregnancy and specific adverse birth outcomes. As the study showed the effect among each country and specific adverse birth outcomes, the countries can design their specific problem addressing strategies.

Conclusion and recommendation

This study found that around one in ten teenagers had pregnancy between the age of 15 and 19 years and nearly half (45.4%) of them had adverse birth outcome. Furthermore, the result showed that 4.7% of the adverse birth outcomes were attributed by teenage pregnancy.

Thus, the authors recommend for policy makers and programmers to design policies and interventions focused on accessing prenatal care, family planning and comprehensive sexual education specifically for teenagers. Public health campaigns could be launched to raise awareness about the risks of teenage pregnancy and the importance of early and regular prenatal care. In addition, support laws and policies to prevent early marriages and pregnancies. Community-based service provision approach can help to reduce stigma and cultural barriers around contraceptive use and prenatal service use among teenagers.

Data availability

Data is available online and any one can access it from www.measuredhs.com.

References

  1. Lawn JE, et al. Global report on preterm birth and stillbirth (1 of 7): definitions, description of the burden and opportunities to improve data. BMC Pregnancy Childbirth. 2010;10:1–22.

    Article  Google Scholar 

  2. Henshaw SK, Singh S, Haas T. The incidence of abortion worldwide. Int Fam Plan Perspect, 1999: pp. S30–8.

  3. Tela FG, et al. Fetal macrosomia and its associated factors among Singleton live-births in private clinics in Mekelle City, Tigray, Ethiopia. BMC Pregnancy Childbirth. 2019;19:1–6.

    Google Scholar 

  4. Akseer N et al. Characteristics and birth outcomes of pregnant adolescents compared to older women: an analysis of individual level data from 140,000 mothers from 20 RCTs. EClinicalMedicine, 2022. 45.

  5. Londero AP, et al. Maternal age and the risk of adverse pregnancy outcomes: a retrospective cohort study. BMC Pregnancy Childbirth. 2019;19:1–10.

    Article  Google Scholar 

  6. Melesse YD, et al. Proportion and associated factors of adverse birth outcomes among women who gave birth at Debre Markos general hospital, East Gojjam zone, Ethiopia, 2022: a cross-sectional study. J Med Clin Nurs Stud. 2024;2(4):1–8.

    Article  Google Scholar 

  7. Tebb KP, Brindis CD. Understanding the psychological impacts of teenage pregnancy through a socio-ecological framework and life course approach. In Seminars In reproductive medicine. Thieme Medical Publishers, Inc; 2022.

  8. McCarthy FP, O’Brien U, Kenny LC. The management of teenage pregnancy. BMJ, 2014. 349.

  9. Rachakonda L, Rawate S, Shiradkar S. Teenage pregnancy. Int J Curr Medicial Appl Sci. 2014;4(2):2059–63.

    Google Scholar 

  10. Eyeberu A, et al. Teenage pregnancy and its predictors in Africa: A systematic review and meta-analysis. Int J Health Sci. 2022;16(6):47.

    Google Scholar 

  11. Banerjee B, et al. Teenage pregnancy: a socially inflicted health hazard. Indian J Community Med. 2009;34(3):227–31.

    Article  PubMed  PubMed Central  Google Scholar 

  12. Darroch JE et al. Adding it up: costs and benefits of meeting the contraceptive needs of adolescents. 2016.

  13. Adeyinka DA, et al. Outcome of adolescent pregnancies in Southwestern Nigeria: a case–control study. J Maternal-Fetal Neonatal Med. 2010;23(8):785–9.

    Article  Google Scholar 

  14. Mukhopadhyay P, Chaudhuri R, Paul B. Hospital-based perinatal outcomes and complications in teenage pregnancy in India. J Health Popul Nutr. 2010;28(5):494.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Papri FS, et al. Adolescent pregnancy: risk factors, outcome and prevention. Chattagram Maa-O-Shishu Hosp Med Coll J. 2016;15(1):53–6.

    Article  Google Scholar 

  16. Mombo-Ngoma G, et al. Young adolescent girls are at high risk for adverse pregnancy outcomes in sub-Saharan Africa: an observational multicountry study. BMJ Open. 2016;6(6):e011783.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Hoque M, et al. Comparison of adverse pregnancy outcome between teenage and adult women at a tertiary hospital in South Africa. Biomed Res. 2014;25(2):167–72.

    Google Scholar 

  18. Saleh MA. Outcomes of teenage pregnancy at Benghazi medical center 2019–2020. Int J Sci Acad Res. 2022;3(3):358–3602.

    Google Scholar 

  19. Measure D. Sampling and Household Listing Manual: Demographic and Health Surveys Methodology. 2012.

  20. Macro I, Measure DHS. Data downloaded from http://www.measuredhs.com, 2011.

  21. Lim LM, Singh K. Termination of pregnancy and unsafe abortion. Best Pract Res Clin Obstet Gynecol. 2014;28(6):859–69.

    Article  Google Scholar 

  22. Vogel JP, et al. The global epidemiology of preterm birth. Best Pract Res Clin Obstet Gynecol. 2018;52:3–12.

    Article  Google Scholar 

  23. Organization WH. International statistical classification of diseases and related health problems: alphabetical index. Volume 3. World Health Organization; 2004.

  24. Sekine K, et al. Identifying the causal effect of child marriage on unmet needs for modern contraception and unintended pregnancy in Nepal: a cross-sectional study using propensity score matching. BMJ Open. 2021;11(4):e043532.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Vrijsen J, et al. Association between dementia parental family history and mid-life modifiable risk factors for dementia: a cross-sectional study using propensity score matching within the lifelines cohort. BMJ Open. 2021;11(12):e049918.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Nguyen HV, Dao AT, Do DV. Propensity score matching in estimating the effect of managerial education on academic planning behavior. Study design: a cross-sectional study. BMC Med Educ. 2011;11:1–10.

    Article  Google Scholar 

  27. Li M. Using the propensity score method to estimate causal effects: A review and practical guide. Organizational Res Methods. 2013;16(2):188–226.

    Article  Google Scholar 

  28. Abadie A, et al. Implementing matching estimators for average treatment effects in Stata. Stata J. 2004;4(3):290–311.

    Article  Google Scholar 

  29. Pavolovich Panko P, et al. Alternatives to randomized control trials: A review of three quasi-experimental designs for causal inference. Actualidades En Psicología. 2015;29(119):20–8.

    Google Scholar 

  30. Mason C, et al. Can propensity score matching be applied to cross-sectional data to evaluate Community-Based rehabilitation?? Results of a survey implementing the who’s Community-Based rehabilitation? indicators in Vietnam. BMJ Open. 2019;9(1):e022544.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Li L, et al. Propensity score-based sensitivity analysis method for uncontrolled confounding. Am J Epidemiol. 2011;174(3):345–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Dehejia R. Practical propensity score matching: a reply to Smith and Todd. J Econ. 2005;125(1–2):355–64.

    Google Scholar 

  33. Asmamaw DB, Tafere TZ, Negash WD. Prevalence of teenage pregnancy and its associated factors in high fertility sub-Saharan Africa countries: a multilevel analysis. BMC Womens Health. 2023;23(1):23.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Kurth F, et al. Adolescence as risk factor for adverse pregnancy outcome in central Africa–a cross-sectional study. PLoS ONE. 2010;5(12):e14367.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Abebe AM, et al. Teenage pregnancy and its adverse obstetric and perinatal outcomes at Lemlem Karl hospital, Tigray, Ethiopia, 2018. Biomed Res Int. 2020;20201:p3124847.

    Article  Google Scholar 

  36. Gortzak-Uzan L, et al. Teenage pregnancy: risk factors for adverse perinatal outcome. J Maternal-Fetal Med. 2001;10(6):393–7.

    Article  CAS  Google Scholar 

  37. Chibber R, et al. Adverse pregnancy outcome among teenagers: a reality? J Obstet Gynaecol. 2014;34(4):297–300.

    Article  CAS  PubMed  Google Scholar 

  38. Shaikh F, et al. Adverse outcome of a teenage pregnancy. J Liaquat Uni Med Health Sci. 2016;15(04):179–82.

    Article  Google Scholar 

  39. Johnson W, Moore S. Adolescent pregnancy, nutrition, and health outcomes in low-and middle‐income countries: what we know and what we don’t know. BJOG. 2015;123(10):1589.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Africa WROf. Report of an adolescent health services barriers assessment (AHSBA) in the united republic of Tanzania: with a focus on disadvantaged adolescents. 2019.

  41. Tamirat KS, et al. Determinants of adverse birth outcome in Sub-Saharan Africa: analysis of recent demographic and health surveys. BMC Public Health. 2021;21:1–10.

    Article  Google Scholar 

  42. Ogawa K, et al. Association between adolescent pregnancy and adverse birth outcomes, a multicenter cross sectional Japanese study. Sci Rep. 2019;9(1):2365.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Mezmur H, Assefa N, Alemayehu T. An increased adverse fetal outcome has been observed among teen pregnant women in rural Eastern Ethiopia: a comparative cross-sectional study. Global Pediatr Health. 2021;8:2333794X21999154.

    Article  Google Scholar 

  44. Kassa GM, et al. Adverse neonatal outcomes of adolescent pregnancy in Northwest Ethiopia. PLoS ONE. 2019;14(6):e0218259.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We would like to acknowledge the owner of the DHS data, and study participants.

Funding

No funding secured for this research.

Author information

Authors and Affiliations

  1. Department of Health Systems and Policy, Institute of Public Health, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia

    Getachew Teshale, Endalkachew Dellie, Andualem Yalew Aschalew, Nigusu Worku, Tesfahun Zemene Tafere, Nebebe Demis Baykemagn & Asebe Hagos

  2. Department of Health Informatics, College of Medicine and Health Sciences, University of Gondar, Gondar, Ethiopia

    Chalie Tadie Tsehay

Authors
  1. Getachew Teshale
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Endalkachew Dellie
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Andualem Yalew Aschalew
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Chalie Tadie Tsehay
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Nigusu Worku
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Tesfahun Zemene Tafere
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Nebebe Demis Baykemagn
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Asebe Hagos
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

GT designed the study, performed analysis, interpreted the result and prepared a manuscript. ED, AYA, CTT, NW, TZT, NDB and AH participated in the development of the study proposal, analysis and interpretation of the paper and the manuscript. All authors read, revised, and approved the final manuscript.

Corresponding author

Correspondence to Getachew Teshale.

Ethics declarations

Ethics approval and consent to participate

Since we used secondary data and had no direct interaction with the study participants, ethical clearance was not required for this study. The DHS Study administered written informed consent to study participants. We submitted an online request to have permission to access the data. The data was obtained from the DHS program’s measure at http://www.dhsprogram.com. The details of the ethical approval for the Demographic and Health Surveys (DHS) program make it possible to approve the download of survey data.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Teshale, G., Dellie, E., Aschalew, A.Y. 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 Pregnancy Childbirth 25, 442 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12884-025-07574-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12884-025-07574-4

Keywords

BMC Pregnancy and Childbirth

ISSN: 1471-2393

Contact us