Prevalence of teenage pregnancy and associated factors in Uganda: a meta-analysis and systematic review Authors: Babatunde Abdulmujeeb Aremu, Ismail Bamidele Afolabi, Awunor Nyemike Simeon, Salaam Mujeeb Published: 2026 ABSTRACT AbstractObjectiveTeenage pregnancy is a major public health problem with huge consequences for maternal health and pregnancy outcomes. This calls for a concerted effort from different stakeholders involved in issues of teenagers’ sexual reproductive health to achieve the SDG target of 2030 on the reduction of maternal mortality. The objective of this systematic review was to estimate the prevalence of teenage pregnancy and its associated factors in Uganda.DesignsWe conducted a systematic review and random-effects meta-analysis according to Preferred Reporting Items for Systematic Review and Meta-analysis standards.Data sourcesAfter prospective registration (PROSPERO CRD42023486460), a literature search was conducted in PubMed, African Journals Online and Google Scholar in December 2023.Eligibility criteria for selecting studiesObservational studies that reported the prevalence of teenage pregnancy among teenagers were included.Data extraction and synthesisArticles were screened for relevance at the title, abstract and full-text levels by multiple reviewers. The quality of studies was independently appraised using the Joanna Briggs Institute Critical Appraisal Checklist and Newcastle-Ottawa Scale. Any discrepancies were resolved by reaching a consensus. The process of extracting data included study information, geographical categorisation, type of studies, population, sample size, results/outcomes and prevalence of teenage pregnancy. The risk of bias was determined by Egger’s test. The results were pooled using random-effects models to calculate the prevalence of teenage pregnancy with a CI of 95% and I² statistic.ResultsOf 918 records screened, 10 studies (4143 participants) were included. The pooled prevalence of teenage pregnancy was 25.5% (95% CI 13.7% to 39.5%; I²=98.9%), with individual study estimates ranging widely from 3.0% to 56.4%, reflecting substantial heterogeneity across study settings, populations and periods of data collection. Higher estimates were observed in the Central–Eastern region (31.7%, 95% CI 19.5% to 45.1%). The overall certainty of evidence was rated as very low using the Grading of Recommendations, Assessment, Development and Evaluation framework, primarily due to very serious inconsistency across included studies (I²=98.9%).ConclusionsPooling studies of mixed quality revealed a slightly higher prevalence of teenage pregnancy in Uganda. This review provides valuable insight for policymakers and healthcare providers in developing interventions and policies to reduce the burden of teenage pregnancy from disadvantaged backgrounds.PROSPERO registration numberCRD42023486460.Keywords: Prevalence, Adolescent, Pregnancy, REPRODUCTIVE MEDICINE Strengths and limitations of this study The restriction to English-language publications may have excluded relevant studies, although a supplementary search in French found no additional eligible studies. The restriction to English-language publications may have excluded relevant studies, although a supplementary search in French found no additional eligible studies. The small number of included studies (n=10) limits the statistical power of subgroup analyses and meta-regression, and reduces the reliability of CIs for regional estimates. The small number of included studies (n=10) limits the statistical power of subgroup analyses and meta-regression, and reduces the reliability of CIs for regional estimates. The MedCalc software used does not implement the Hartung-Knapp-Sidik-Jonkman correction for small meta-analyses; the reported CIs may therefore be narrower than they would be with this adjustment, and should be interpreted with appropriate caution. The MedCalc software used does not implement the Hartung-Knapp-Sidik-Jonkman correction for small meta-analyses; the reported CIs may therefore be narrower than they would be with this adjustment, and should be interpreted with appropriate caution. Three studies collected data during the COVID-19 pandemic (2020–2022), a period marked by school closures, disruption of reproductive health services and increased vulnerability of adolescent girls; these factors may have temporarily elevated pregnancy rates in some settings, contributing to heterogeneity and potentially inflating the pooled estimate. Three studies collected data during the COVID-19 pandemic (2020–2022), a period marked by school closures, disruption of reproductive health services and increased vulnerability of adolescent girls; these factors may have temporarily elevated pregnancy rates in some settings, contributing to heterogeneity and potentially inflating the pooled estimate. Likelihood recall and social desirability bias may affect self-reported pregnancy data in questionnaire-based studies, particularly in settings where teenage pregnancy carries stigma. Likelihood recall and social desirability bias may affect self-reported pregnancy data in questionnaire-based studies, particularly in settings where teenage pregnancy carries stigma. Introduction As per a report by the United Nations Population Fund (UNFPA), every year, 14 million girls between the ages of 15 and 19 give birth.1 More than 90% of these births occur in developing countries.2 The pregnancy rate per 1000 women in 2021 was 42.5. The regions with the highest rates of pregnancy were Latin America at 53.2 and Sub-Saharan Africa at 101 per 1000 women.3 The reason for concern and the gap between the potential impact of initiatives for adolescents is highlighted by these statistics. In East Africa, close to 10% of young females give birth by age 16.4 Specifically, Uganda has the largest percentage of women who give birth prior to age 20 (63%) and also has the largest total fertility rate (TFR) in East Africa at 6.2.5 6 The country also has a young population, with 48% of its population below age 15.7 The population aged 10–19 years constitutes 25.7%, while that aged 10–24 years is at 35.2%.8 Despite the Uganda Demographic and Health Survey findings that TFR declined from 7.4 children per woman in 1988–1989 to 5.4 in 2016, it is seen that fertility rates have increased for teenagers.9 These high rates of teen pregnancies are linked to negative health consequences; for women in Uganda aged 15–19 years, the primary causes of mortality and morbidity are related to pregnancies, unsafe abortions and deliveries.10 Adolescent pregnancies may have negative socioeconomic and psychological influences on the young mother and her siblings due to links with health complications, Sexually Transmitted Infections and family planning.11 Variables associated with teenage pregnancies include maternal age during pregnancies, maternal education during pregnancies, residence, employment status, contraceptive or non-contraceptive use, poverty, parental home stability, gender inequality and decision-making, house chores, monthly income, lack of communication on reproductive health between parents and their children, maternal history of teenage pregnancies, religion, ethnic group membership, premarital sexual relations before 15 years for females and marriage before 18 years.12,14 Addressing teenage pregnancy in Uganda is key to meeting the United Nations’ Sustainable Development Agenda 2030 targets on ensuring healthy lives and well-being at all ages and National Development Plan III, Health Sector Strategic Plans I and II and Vision 2040 for Sustainable Development. Moreover, a report on teenage pregnancies by the UNFPA showed that the percentage of teenage pregnancy will increase globally by 2030, particularly in Sub-Saharan African countries.1 This calls for a concerted effort from different stakeholders involved in issues of teenagers’ sexual reproductive health in order to achieve the SDG target 2030 on the reduction of maternal mortality. Despite the concern about teenage pregnancy in different regions in Uganda, these data have not been summarised using meta-analysis among teenagers in Uganda. A previous review published on teenage pregnancy among teenagers focused on Sub-Saharan African countries, as only a few studies from Uganda were included.15 The result of this systematic review and meta-analysis will benefit policymakers and stakeholders to improve the prevention as well as management strategies for teenage pregnancy. Given the above background, this study aimed to pool the prevalence data and factors associated with teenage pregnancy in Uganda. Specific objectives of this review are: To provide an estimate of the prevalence of teenage pregnancy among teenagers in Uganda according to existing published studies. To provide an estimate of the prevalence of teenage pregnancy among teenagers in Uganda according to existing published studies. To establish the risk factors associated with the prevalence of teenage pregnancy among teenagers in Uganda, according to existing published studies. To establish the risk factors associated with the prevalence of teenage pregnancy among teenagers in Uganda, according to existing published studies. To critically appraise the methodological quality of the prevalence studies to highlight areas for improvement in future research. To critically appraise the methodological quality of the prevalence studies to highlight areas for improvement in future research. Methods Systematic review registration and reporting The protocol for this review was registered within the PROSPERO database (registration number CRD42023486460).16 This study was reported according to Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) standards.17 Data sources and search strategies All the primary studies that investigated the prevalence and factors associated with teenage pregnancy in Uganda were identified through the following databases: PubMed, Google Scholar and African Journal Online conducted by BAA, IBA, SM and ANS in December 2023, using specific terms. The original search was conducted in December 2023 and was updated in April 2025 to include literature published through March 2025. The following concepts were captured by keywords and specific key terms: “Prevalence” AND Teenage pregnancy, Adolescent pregnancy OR Young maternal pregnancy AND Determinant factors OR Associated factors AND Uganda were used in combination with the search terms. The electronic search was limited to two decades, starting in December 2003. Eligibility criteria and study selection Specifically, studies were included if they: reported the prevalence of teenage pregnancy among females aged 10–19 years (consistent with the WHO definition of adolescence); were conducted in Uganda; used observational study designs (cross-sectional or case-control) and reported sufficient quantitative data to extract prevalence estimates. Reviews, systematic reviews, commentaries, case studies and case series were excluded. Studies were further excluded if they: were not published in English, were not from Uganda, provided insufficient details on work-related teenage pregnancy, did not provide full tests, were not accessible or provided insufficient data about the sampling technique. Title and abstract screening were performed independently by two reviewers (BAA and IBA). Any disagreements were resolved by a third reviewer (ANS). Full-text screening of potentially eligible articles was conducted independently by two reviewers (SM and ANS), with disagreements resolved by consensus among all four reviewers (BAA, IBA, SM and ANS) during weekly evaluation meetings. The inter-rater agreement at the abstract screening stage was assessed using Cohen’s kappa, which indicated substantial agreement (κ=0.76). Quality assessment and data extraction Articles were screened using their titles, abstracts and full paper review before being included in the meta-analysis. According to the Uganda Bureau of Statistics, Uganda was classified into four different regions: Central, Western, Eastern and Northern regions.18 The Joanna Briggs Institute (JBI) critical appraisal checklist was used to assess the quality of the selected studies.19 A modified JBI critical appraisal checklist for prevalence studies was used to assess the quality of the selected cross-sectional studies.19 The checklist was adapted from the full 9-item JBI tool and condensed to five key domains relevant to prevalence estimation: (1) Sample representativeness, (2) Participant recruitment method, (3) Adequacy of sample size, (4) Sufficiency of data analysis coverage and (5) Exposure assessment. Each item was scored using a binary system (1=Yes; 0=No/Unclear/Not applicable), yielding a total score range of 0–5, with scores classified as high quality (≥4), medium quality (3) and low quality (≤2) (online supplemental table 1). In the case-control study, the Newcastle-Ottawa Scale was used. Studies rated 7–9 stars were rated as high quality, 5–6 stars as moderate quality and ≤4 stars as low quality, indicating progressively higher risk of bias as one moves through the categories of selection, comparability and outcome or exposure (online supplemental table 2). The data extraction tool captured necessary details from every included article, such as: first author, year of publication, region of publication, study design and type, study population, sample size, the outcome measured and prevalence of teenage pregnancy. Data extraction was performed independently by two reviewers (BAA and IBA) using a standardised pre-designed extraction form. A third reviewer (ANS) independently verified all extracted data against the original publications. Discrepancies in numerical values, study design classification or prevalence reporting were resolved by consensus, with all four authors consulted where agreement could not be reached between the two primary extractors. Statistical analysis The rates representing the prevalence of teenage pregnancy in individual studies were pooled using random-effects models, and a 95% CI was estimated. The heterogeneity was assessed graphically using forest plots and statistically through I2 statistics, which give percentages of variation across the studies.20 The Egger’s test (p<0.05) was explored for publication and related biases in funnel plot analysis.21 The subgroup analysis was performed to examine the source of heterogeneity attributed to geographical regions. Analyses were performed using the statistical software MedCalc V.19.13.22 Although MedCalc does not include the adjustment for the Hartung-Knapp-Sidik-Jonkman adjustment in meta-analysis, it was recognised as relevant in meta-analysis studies where the number of studies was small. Therefore, the findings were interpreted cautiously. The direction and consistency, rather than the accuracy, of the CI were highlighted. Sensitivity analyses were pre-planned as part of this review. A leave-one-out analysis was performed to assess the robustness of the pooled estimate by sequentially removing each study and re-pooling the remaining studies. An additional sensitivity analysis was performed, restricted to studies rated as high quality. Both analyses were conducted using random-effects models. Meta-regression was also performed to examine whether the year of publication was a statistically significant predictor of the pooled prevalence. Assessment of certainty of evidence The overall certainty of the body of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluation approach across five domains: risk of bias, inconsistency, indirectness, imprecision and publication bias. The assessment was applied to the primary outcome (pooled prevalence of teenage pregnancy) and reported in a summary of findings format. Meta-regression We performed a meta-regression analysis to show that there was no significant statistical relationship between the year of publication and the pooled prevalence of teenage pregnancy. Although a downward trend in prevalence over time was observed, the association was not statistically significant (p>0.05). This suggests that the robustness of the studies could not directly be influenced by the year of publication, as shown in the list of figures, showing meta-regression figure 1. Showing meta-regression. Results The exploration of electronic databases identified 918 citations, as shown in the list of figures, showing PRISMA flow chart of reviewed studies figure 2. Of these, 18 were selected based on the article/abstract for full-text evaluation. After the eligibility assessment, 8 studies were excluded. Finally, 10 studies were included in this meta-analysis. Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow chart of reviewed studies. AJOL, African Journal Online. Study characteristics and quality appraisal The detailed study characteristics are provided as shown in the online supplemental table 3. Among the 10 studies included in the systematic review, 8 cross-sectional studies were assessed using JBI; Letaru,23 Okot et al,24 Manzi et al,25 Chemutai et al26 Akanbi et al,27 Musinguzi et al,28 Agaba29 and Tuyiragize et al.30 The other two studies were case-control studies, which were assessed using NOS; Ochen et al,31 and Kasozi Kirungi et al.32 The total population of teenagers who participated in these studies was 4143. The quality appraisal of 8 cross-sectional studies revealed that 5 (over 80%) were of high methodological quality and 3 (79%–50%) were of medium methodological quality; the 2 case studies were of high quality (7–9). Pooled prevalence of teenage pregnancy in Uganda The prevalence estimates were log-transformed before being pooled because of a variance-stabilising transformation that is recommended when conducting meta-analyses of prevalence rates, as guided by CSA.33 The prevalence rates were then back-transformed for interpretation. The rationale for this is that it aims to improve normality assumptions and also reduce the impact of outliers in prevalence rates. The pooled estimated prevalence of teenage pregnancy in Uganda was 25.5% (95% CI 13.7% to 39.45%), and the observed heterogeneity was significant (I2=98.94%) as shown in the list of figures, showing the pooled prevalence of teenage pregnancy in Uganda from pooled studies figure 3. The regional prevalence of teenage pregnancy across the Central and Eastern regions has the highest prevalence of teenage pregnancy at 31.67% (CI 19.47% to 45.11%, I2=94.75%), the Northern region has a prevalence of 23.5% (CI 1.10% to 62.03%, I2=93.01%) and the Western region has a prevalence of 22.64% (CI 6.17% to 45.55%, I2=93.01%), as shown in the list of figures, showing the regional prevalence of teenage pregnancy in Uganda from the pooled studies figure 4. Showing the pooled prevalence of teenage pregnancy in Uganda from the pooled studies. Showing the regional prevalence of teenage pregnancy in Uganda from the pooled studies. The pooled estimate of 25.5% encompasses a wide range of study-level prevalence estimates, from as low as 3.0%, a school-based study in Hoima district, Kasozi Kirungi et al,32 to as high as 56.4% a study in Arua district among community respondents by Letaru.23 This heterogeneity is substantial (I²=98.9%) and is discussed further in the sensitivity and subgroup analyses below. Factors associated with teenage pregnancy As summarised in online supplemental table 4, rural residential area OR=0.2 (95% CI 0.04 to 0.61), no formal education OR=0.4 (95% CI 0.27 to 0.48), no history of contraceptive use OR=0.2 (95% CI 0.10 to 0.79) and lack of knowledge of contraceptives OR=0.2 (95% CI 0.05 to 0.70) were statistically significant factors associated with teenage pregnancy as shown in online supplemental table 2. Sensitivity analysis A leave-one-out sensitivity analysis was performed to assess the robustness of the pooled prevalence estimate. Excluding any single study did not substantially alter the pooled estimate, which remained between 21.8% and 28.2% across all iterations, consistent with the overall pooled estimate of 25.5% (online supplemental file). The two studies that had the greatest individual influence on the pooled estimate were Letaru,23 which contributed the highest individual prevalence of 56.4%, and Kasozi Kirungi et al,32 which contributed the lowest estimate of 3.0%. An additional sensitivity analysis restricted to the five studies of high methodological quality (JBI score ≥4 or NOS score ≥7) yielded a pooled prevalence of 24.1% (95% CI 12.8% to 37.9%; I²=97.3%), which was not meaningfully different from the overall pooled estimate (25.5%), indicating that the inclusion of medium-quality studies did not introduce material bias into the pooled estimate. Assessment of publication bias Objective assessment of publication bias was performed using Egger’s linear regression test, with p<0.05 indicating statistically significant publication bias.20 Visual inspection of the funnel plot showed approximate symmetry, with no strong evidence of publication bias. Egger’s linear regression test confirmed this finding (Kendall’s τ=−0.045; p=0.86; 95% CI −12.21 to 51.97), indicating no statistically significant publication bias (p>0.05). These results should be interpreted with caution given the limited number of included studies (n=10), which reduces the statistical power of publication bias tests, as shown in the list of figures, showing Forest publication bias figure 5. Forest plot showing publication bias. Discussion Teenage pregnancies continue to be an important public health issue, in addition to being one of the top contributors to morbidity and mortality among adolescent girls in Uganda. This systematic review and meta-analysis bring together existing evidence on the prevalence of teenage pregnancies, as well as their determinants in Uganda, using ten studies that have fulfilled the criteria for inclusion. As far as this review has established, this study forms the first systematic review and meta-analysis to have estimated, using primary literature, the combined prevalence of teenage pregnancies in Uganda. The undertaking of this study has followed PRISMA guidelines. The general prevalence of teen pregnancies stood at 25.5%, which is a high prevalence, with variations in different regions in Uganda. The eastern region recorded a higher prevalence, which is in line with the Uganda Demographic Health Survey estimates that stood at 26% in 2006, 25% in 2011 and 26% in 2016.34 The high levels of teenage pregnancy in the country have persisted over the years despite various efforts that have been made. However, the pooled value is comparable to the country prevalence of about 25%, but it is higher than the values reported in some African countries; these include 13.3% in Cameroon,35 1.5% in Nigeria,36 23.59% in Ethiopia37 and 15.4% in Ghana.38 On the other hand, the value was lower than the one reported in some countries; these include 60.5% in one study39 and 54.6% in the pooled value of East Africa.40 On the Ugandan side, the heterogeneity in the country in terms of socioeconomic conditions, poverty levels and the availability of sexual and reproductive health services may account for the large variations in the value observed in the different regions. The findings by Wasswa et al34 are consistent with the above assertion in the sense that the study showed the large variations in the prevalence of teenage pregnancies in Uganda attributable to the heterogeneity in the population. The wide range of prevalence estimates (3.0% to 56.4%) observed across included studies warrants specific discussion. The lowest estimate (3.0%) was reported by Kasozi Kirungi et al32 in a school-based study in Hoima district. School-based sampling is well-known to underestimate teenage pregnancy prevalence because girls who become pregnant are more likely to have already dropped out of school at the time of data collection, a form of selection bias that is difficult to adjust for. In contrast, the highest estimate (56.4%) was from Letaru,23 a community-based study conducted in Arua district in north-western Uganda, a region historically affected by armed conflict, internal displacement and limited access to reproductive health services. Community-based designs, while more representative, may capture a more socioeconomically vulnerable population with structurally higher pregnancy risk. This conflicting evidence is consistent with several findings from comparable Sub-Saharan African settings studies, where teenage pregnancy prevalence has been reported to be relatively low in some urban and better-served populations, including about 3% in Addis Ababa,33 41 Ethiopia42 and about 5% in Rwanda,43 while remaining substantially higher in rural and socioeconomically disadvantaged settings. This variation highlights the importance of context-specific interpretation. The research shows that various factors are significantly associated with teen pregnancies, such as living in a rural area, not attaining formal education, not practising contraceptive use, lack of knowledge about contraceptives and not being formally employed. These variables, when considered together, show that structural factors have a great influence on teen reproductive outcomes. It was also discovered that adolescents living in rural areas had an increased risk of teenage pregnancy, which might be attributed to a lack of information and reproductive health services compared with adolescents living in urban areas. This aligns with another systematic review of adolescent pregnancy in Africa.34 Moreover, adolescents’ lack of knowledge on contraception and those not practising contraception are also at a higher risk of early pregnancies. Inadequate access to contraceptive services and SRH education in developing countries has been known to contribute to high teenage pregnancy rates in Sub-Saharan Africa for a long time.44,46 The unmet need for contraception by adolescents is a concern and is also a cause for increased unplanned pregnancies.47 Lack of formal education came up as an important factor that contributed to an increased risk of teenage pregnancies. Education can improve the ability of adolescents to make well-informed decisions about their sexual conduct and the prevention of pregnancy. This study agrees with systematic reviews carried out in Ghana and other low and lower-middle-income nations, where lack of formal education emerged as an important risk factor for teenage pregnancies.38 48 Lastly, the high degree of heterogeneity among the included studies is also a reflection of the varying nature of their settings and time of data collection, which includes periods of the COVID-19 pandemic. Outliers among the included studies also added to heterogeneity. Conclusion This review provides the first national pooled estimate of teenage pregnancy in Uganda, which shows a high prevalence with a great variation by region. The observed heterogeneity is most likely due to differences in study designs, whether community- or facility-based, regional differences in socioeconomic status and education levels, as well as the time of data collection, which in some regions could have been during the COVID-19 pandemic. The review highlighted various factors associated with teenage pregnancy in Uganda, such as lack of education, lack of contraceptive use, lack of knowledge of contraceptives and rural residence. Prevention of early marriage, promotion of girl child education, creating awareness on the use of contraceptives and ensuring their accessibility, especially in rural areas within Uganda, is recommended. It is further recommended that policymakers, the education ministry and healthcare providers in Uganda jointly develop interventions and policies to identify the structural barriers, including poor rural access, stockouts, stigma and the lack of services that are friendly to adolescents, and to overcome these, including the use of mobile outreach clinics, school-based education and community-based delivery of contraception. This will reduce the burden of teenage pregnancy to achieve the SDG target 2030 on the reduction of maternal mortality. Dissemination of findings The findings from the data synthesis will be published in relevant peer-reviewed journals and conferences targeting adolescent and reproductive health. Additionally, the findings will be presented during the monthly academic seminars at the Faculty of Health Sciences, Habib Medical School, Islamic University in Uganda. Supplementary material Acknowledgements We thank the management of the Islamic University in Uganda, Habib Medical School, Faculty of Health Science, for allowing us to access the university-subscribed databases. Footnotes Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors. Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-101827). Provenance and peer review: Not commissioned; externally peer reviewed. Patient consent for publication: Not applicable. Ethics approval: Not applicable. Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research. Author note: BAA is working as a senior lecturer of public health and global health researcher, in the faculty of health science, at Habib Medical School, Islamic University in Uganda. He has been working on musculoskeletal disorders in African settings. IBA is a public health specialist and global health researcher with numerous years of experience in infectious and non-infectious diseases in sub-Saharan Africa. He is currently a doctoral student in public health at the University of Ottawa, Ottawa, Ontario, Canada. ANS is an Associate Professor of public health and community medicine at Delta State University, Abraka, Nigeria, and was a past senior lecturer in Public Health at Habib Medical School, Islamic University in Uganda. His research interests include reproductive and family health. SM is a Medical Laboratory Scientist and Public Health Specialist. He is a lecturer in the Department of Pathology at the Islamic University in Uganda, where he has numerous years of experience in laboratory practice and public health research. Data availability statement No data are available. References Supplementary Materials Data Availability Statement No data are available.