A systematic review and meta-analysis was conducted to evaluate the pooled prevalence of congenital anomalies among newborns in Ethiopia. The review was conducted in accordance with the guideline of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) checklist [9]. The review was registered in the International Prospective Register of Systematic Reviews (PROSPERO) 2024 CRD42023426258 Available from https://www.crd.york.ac.uk/PROSPERO/view/CRD42023426258.

The literature search was conducted across multiple databases, including PubMed, Scopus, Google Scholar, Web of Science, and Wiley Online Library. During exploration of the available study the following key search terms and Medical Subject Headings [MeSH] “Prevalence”, “Magnitude”, “Congenital Abnormalities”, “Congenital Malformation” “Congenital Anomalies” “Birth defects” “Associated Factors”, “Predictors”, “Determinant”, “Newborns” “Infants” and “Ethiopia” were used separately or in combination with the Boolean operator’s terms “AND” and “OR” (Supplementary File 1). Additionally, the reference lists of all retrieved articles were examined to find more relevant studies. The search incorporated studies available up to the 28th of February 2025. All identified articles were imported into EndNote version X20, and duplicate records were removed. Subsequently, a comprehensive screening process was conducted, including evaluation of titles, abstracts, and full-texts, as well as an assessment of article quality based on predefined eligibility criteria. Finally, the eligible studies were compiled together for analysis.

Inclusion and exclusion criteria: All primary studies that reported the prevalence of congenital anomalies and/or its predictors among newborns in Ethiopia were illegible for the study. However, the following studies were excluded: those that were not fully accessible despite two attempts to contact the corresponding author via email or if the outcome of interest could not be determined or calculated based on the available data; and studies that received a poor quality score according to the predefined criteria.

The selected studies were carefully examined, and the essential data was independently extracted and summarized by two authors using a standardized data extraction spreadsheet format created in Microsoft Excel. If any disagreements arose between the data extractors, a third author was consulted to resolve them. For each included study, information was extracted with regard to the corresponding author, publication year, region of the study, study design, sample size, the magnitude of congenital anomalies, and identified predictors. Two-by-two data were also extracted for each identified factors.

The methodological quality of each included study was independently assessed by two authors using the Newcastle-Ottawa Scale (NOS), adapted for assessing the quality of observational studies [10]. Any disagreements between reviewers regarding the quality assessment of individual articles were resolved through discussion among all authors, with the final decision reached by consensus. Ultimately, only studies that scored 5 or more out of 10 across the three domains of the modified NOS were included in the analysis.

In addition, three authors independently evaluated the risk of bias in the included studies using the risk of bias tool designed for prevalence studies developed by Hoy et al. [11]. The assessment tool included ten criteria, with each unmet criterion scored as 0 (no). The total score was calculated by summing the responses, yielding an overall quality score between 0 and 10. Any discrepancies at the time of data abstraction were fixed by discussion and consensus. When available data were insufficient to make a judgment on a specific criterion, the corresponding authors were contacted for clarification. If uncertainty persists, the item was scored as 0, indicating a high risk of bias. Eventually, studies receiving 8 or more ‘yes’ responses were classified as having a low risk of bias, those with 5–7 ‘yes’ responses as moderate risk, and those with 4 or fewer ‘yes’ responses as high risk.

Heterogeneity among the included studies was assessed using the I² statistic. The results indicated substantial heterogeneity (I² = 99.8%; P < 0.001). Therefore, a random-effects model was used to estimate the pooled prevalence. The pooled prevalence, together with its corresponding 95% confidence interval (CI), was calculated and displayed using a forest plot. Potential sources of heterogeneity were further explored using subgroup analyses based on study region, year of publication, sample size, and risk of bias. Furthermore, meta-regression analyses were performed using year of publication and sample size as covariates to examine their contribution to heterogeneity. To evaluate the influence of each study on the overall assessment, a sensitivity analysis was performed by systematically excluding each study one at a time. Additionally, publication bias was assessed using funnel plot symmetry, Egger’s regression test, and Begg’s test. Finally, the various predictors were expressed using pooled odds ratios (PORs) with corresponding 95% CI. Heterogeneity tests, publication bias, and sensitivity analysis also carried out for each identified predictors accordingly. All statistical analyses were performed using STATA™ version 14 software (StataCorp LP, College Station, TX, USA).

A total of 692 records regarding the prevalence of congenital anomalies and its predictors among newborns in Ethiopia were retrieved using electronic database searches. Of the identified records, 72 studies were excluded due to duplication. Following title and abstract screening, an additional 567 studies were excluded as they were not relevant for this systematic review and meta-analysis. The remaining 53 full-text articles were assessed for eligibility based on the predefined criteria, resulting in further exclusion of 31 studies. Ultimately, 22 studies that met all eligibility requirements were included in the final systematic review and meta-analysis (Figure 1).

In this study total of 22 original studies published between 2013 and 2025 years that showed the magnitude of congenital anomalies and/or its predictors among newborns in Ethiopia were included. The sample size of the included studies ranged from 295 in North Wollo Zone [12] to 45951 in Southwest Ethiopia [13]. The highest prevalence of congenital anomalies was stated in the study conducted in Butajira Hospital 57.5 per 1,000 birth [14] while the lowest was stated in the study conducted in Kersa district Eastern Ethiopia 3.8 per 1,000 [15] (Table 1).

The pooled prevalence of congenital anomalies among newborns was estimated to be 21.25 per 1,000 (95% CI: 17.93–24.58) with a significant level of heterogeneity as evidenced by I² statistic (I² = 99.8%; p < 0.001). This finding indicates a great variability in the prevalence of congenital anomalies among newborns across the included studies. Hence, a random-effects model was employed to estimate the pooled prevalence (Figure 2).

To identify the possible source of heterogeneity across the included studies, subgroup analysis based on the region where the study was conducted, years of publication, and sample size were conducted. Although only one study done in Harari and Somali region with prevalence of 32.40% and 14.30%, respectively, sub- group analysis shows the highest prevalence was observed in Southern Nation Nationality and People Region (SNNPR) with prevalence of 28.6% followed by Tigray region 26.25% while the lowest prevalence was observed in Addis Ababa 16.33%. Subgroup analysis also revealed that the pooled prevalence of congenital anomalies was higher in studies with a sample size smaller than the median of the total sample size (<1,081) compared to those with a sample size equal to or greater than the median (27.61% vs. 15.58%) (Table 2). Additionally, a meta-regression analysis was performed using year of publication and sample size as covariate. However, none of these variables were statistically significant for the presence of heterogeneity.

Of the total neonatal anomalies identified, central nervous system abnormality was the commonest type occurring in 48.7%, followed by anomalies with more than one system involvement (14.9%) (Table 3).

Sensitivity analysis was performed to evaluate the effect of each study on the pooled prevalence of congenital anomalies among newborns by using the leave-one-out approach, excluding each study step-by-step from the analysis. The results showed that there was no a single study that significantly affects the overall prevalence of congenital anomalies among newborns (Supplementary File 2).

The graphical inspection of the funnel plot was symmetrical in distribution indicating the absence of publication bias among the included studies. Egger’s (p = 0.769) and Begg’s test (p = 0.386) computed to prove the existence of publication bias, also revealed no evidence of publication bias among the included studies.

Data regarding 25 identified predictors of congenital anomalies among newborns were extracted from the available studies and the results pooled odds ratio for each identified variable were presented by forest plot (Supplementary File 3). For each identified predictors, sensitivity analysis was also carried out by excluding each study one by one, but the result showed that there was no strong evidence for the effect of a single study on the overall results. Moreover, for each identified variable publication bias was assessed and no publication bias was detected. Pooled odds ratio (POR), level of significance, heterogeneity, and publication bias for each identified variable were summarized below (Table 4).

More than two out of every one hundred fetuses were born with congenital anomalies. Behavioral characteristics of mother such as cigarette smoking, alcohol consumption, pesticide exposure, use of medications during pregnancy, and lack of folic acid supplementation during pregnancy were significantly associated with congenital abnormalities among newborns in Ethiopia. Additionally, a history of maternal chronic illness, twin pregnancy, and low birth weight (<2.5 kg) were also significantly linked to the occurrence of congenital anomalies.

Therefore, in order to combat congenital anomalies, healthcare providers in collaboration with relevant stakeholders should focus on strengthening pre-conceptional folate programs, propose strict agricultural protection policies for pregnant women, reducing substance use and inappropriate medication during pregnancy.

This systematic review and meta-analysis has some limitations that need to be considered. Most of the studies included the reviews were cross-sectional. As a result, a cause-effect relationship cannot be ascertained. Moreover, there was substantial heterogeneity among the included studies and the cause/s for the heterogeneity thus remains undetected. Some of Ethiopian regions were underrepresented, with certain areas having only one study or none at all. This may reduce generalizability of the study to the entire country. Exclusion of studies with poor quality score may also under/or overestimate the prevalence of congenital anomalies. Variations in reporting practices, diagnostic skills and equipment in different facilities may have contributed to misclassification bias or underreporting of certain types of congenital anomalies.